The Unz Review • An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersiSteve Blog
Benford's Law
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
AgreeDisagreeThanksLOLTroll
These buttons register your public Agreement, Disagreement, Thanks, LOL, or Troll with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used three times during any eight hour period.
Ignore Commenter Follow Commenter
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks

From Wikipedia:

Benford’s law, also called the Newcomb–Benford law, the law of anomalous numbers, or the first-digit law, is an observation about the frequency distribution of leading digits in many real-life sets of numerical data. The law states that in many naturally occurring collections of numbers, the leading digit is likely to be small. In sets that obey the law, the number 1 appears as the leading significant digit about 30% of the time, while 9 appears as the leading significant digit less than 5% of the time. If the digits were distributed uniformly, they would each occur about 11.1% of the time.

For example, consider all baseball pitchers who have won at least 100 games in their major league careers:

There is one whose number of career wins begins with a 5: 1: Cy Young (511)

With a 4: 1: Walter Johnson (417)

With a 3: 22

With a 2: 94

With a 1: 503

What’s going on is that winning a major league baseball game is difficult, so the number of men drops off the higher you count. Many social statistics look like this, following Benford’s Law.

On the other hand, if you look at career winning percentages among major league pitchers with at least 100 career decisions, only two begin with a 7: Albert Spalding, the 1870s pitcher who retired at 27 to go into the sporting goods business (a rather successful man) and Spud Chandler, the Joe DiMaggio Era Yankees pitcher. But then are 125 pitchers in the .600s, with Clayton Kershaw of the current Dodgers the highest at just below the .700.

I suspect there are more pitchers in the .300s than in the .200s and more in the .200 than in the .100s. That’s because they don’t let you stick around in the majors if you are really bad. Among pitchers with 100 career decisions, only two are in the .200s, while many more are in the .300s.

Many other social statistics looks like this, not following Benford’s Law.

On the third hand, if you want to know the maximum number of players on the roster of a major league team through August, it’s almost always going to be a number beginning with the digit 2. That’s because the rules allowed 25 players on the roster until 2020, when it was increased to 26.

And many social statistics look like this.

You see a lot of splitting or lumping due to administrative decisions.

For example, if a voting precinct grows so large that the physical facility gets overcrowded with voters and voting slows down, the precinct is likely to eventually be split into two. On the other hand, if the number of voters in a precinct turns out to be too small to justify the fixed costs of voting facility, the precinct is likely to be lumped in with another precinct.

Hence, there’s little reason to expect the number of votes recorded in precincts to follow Benford’s Law.

 
Hide 190 CommentsLeave a Comment
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
    []
  1. anonymous[124] • Disclaimer says:

    What about the millions of deleted/switched Trump votes?

  2. Thank you, Steve Sailer, for applying your considerable talent for statistics and getting around to this thing that some of us have mentioned here.

  3. Hockamaw says:

    Seems fairly obvious to me what happened in the election. Dem nonprofit groups took advantage of COVID vote by mail rules to do massive ballot harvesting efforts (i.e. going door to door in apartment buildings, housing projects, group homes, nursing homes, etc. and collecting ballots) in densely populated urban areas in key swing states. But doing this costs money so you can’t do it in every state, which explains why you see the massive, frankly unbelievable, turnout numbers in Detroit and Milwaukee, while you don’t see the same thing in Cleveland. While perhaps untoward, this is not illegal. The Democrats won the election. They may have done it using underhanded tactics, but they’re Democrats. What did anyone expect? TRUMP failed to stop them, so he lost. I think it’s frankly a sad state of affairs that so many on the right seem to prefer living in fantasy world than coming to terms and dealing with the reality of the situation.

    • Agree: Jack D
  4. Interesting, but how about looking beneath the hood at saves versus complete games. I’d posit the latter is the benchmark for determining the better of today’s pitchers.

    BTW, I thought the original application of Benford’s law to the count of votes went to the number of precincts that reported values ending in zero.

  5. Because if there’s one thing that will really convince your average American voter, it’s math.

    • LOL: Hangnail Hans
    • Replies: @Carol
  6. So why did the normal looking cities follow Benford’s Law and the fraudulent looking ones did not?

    • Agree: Daniel H
    • Replies: @Deadite
  7. Walsh2 says:

    If there’s power at stake then there’s fraud involved. This said, proving fraud is a tough get at any level. I suspect moreso when the perpetrators are practiced and adept.

    • Agree: bomag
  8. @anonymous

    Well maybe 2020 isn’t done with fun surprises. DJT had better hope so, because otherwise he’s a little bit deeper in that hole with each passing day. Maybe the Electoral College vote on Dec 14 will be a watershed for him.

  9. As I understand it, Benford’s Law applies primarily to counting stats, not ratios (such as pitchers’ winning percentages), and only those counting stats which are not artificially constrained. I haven’t looked it up, but I’ll bet if you looked at the career win totals (or loss totals) of all pitchers in major league history, they would follow Benford’s Law. (You don’t have to restrict it to those with 3-digit totals like Steve did in his first example; the Law applies to the first digit, regardless of how many follow it.)

    The number of votes in each precinct should be expected to follow the Law. Precinct sizes do get adjusted as Steve explained, but not in such a way as to keep them within a narrow range like baseball roster sizes.

    According to what I’ve read, Trump’s reported precinct-by-precinct vote totals did indeed follow Benford’s Law. Biden’s did not, at least in the swing states.

    • Agree: ScarletNumber
    • Replies: @res
    , @arfoo
  10. Then the question is, why is it that Trump-leaning precincts do follow Benford’s law, but Biden-leaning precincts don’t? Maybe the reason is that conservative districts are more rural, so consolidation/splitting is less convenient because of distances and administrative boundaries.

    Does the down-ballot vote in Biden-leaning precincts follow Benford’s law? Did past presidential votes in Biden-leaning districts follow Benford’s law?

    As of now, it’s not looking good for the Trump/Giuliani election fraud lawsuits. What I’ve learned from this is that the evidentiary standards required to accomplish what needs to be accomplished are very strict. You’ve got a crime, and the perps are in charge of granting access to the evidence. It would be a slam dunk, if the standards were like those needed to prove racial discrimination.

  11. The most suspicious statistic I have seen is that in a lot of PA districts, the on-election-day tally gave Trump leads varying from a few % to 10s of %. Then the postal ballots were counted and they each flipped to Biden by very small margins.

    If you wait to see how many votes you need to tip the result and then add just enough, this is the pattern you would expect.

    This looks like Illinois in 1960, when Mayor Daley stopped the Chicago count until the downstate results came in, and then stuffed just enough ballots to give the state to JFK.

    All the votes ought to counted at the same time.

  12. Coemgen says:

    How ’bout probabilities instead of stats?

    What is the probability that a “hundred year virus” would show up in any given year?

    One in a hundred maybe?

    How ’bout that virus shows up in a particular month in a given year?

    One in a thousand perhaps?

    How ’bout the virus shows up just about the time that the impeachment turns into an exposé of the Biden’s questionable behavior, the existence of the “interagency,” and that Sauron is above the law _AND_ the virus is used, in multiple ways, against Trump in the 2020 election?

    One in a googol perchance?

    • Agree: BenKenobi
    • Replies: @Anonymous
  13. Anonymous[179] • Disclaimer says:
    @International Jew

    It would be a slam dunk, if the standards were like those needed to prove racial discrimination.

    Could you make the case, to illustrate for us how it might look?

    • Replies: @jon
  14. @International Jew

    Agree. My understanding is that Benford’s Law as applied to election precincts is breached only by some Biden-voting areas and additionally these precincts are in the swing states for the most part.

    If precinct vote counts did not conform to Benford’s Law, then this could be demonstrated widely across the whole country and regardless of political affiliation. It cannot.

  15. I simply cannot follow Steve’s Baseball Political Theory.

  16. Ano says:

    Hi Steve,

    Benford’s Law isn’t one of the legal firms representing Trump is it?

    If it is, can you please cut out from old newspapers individual letters which by gluing on to a sheet of paper form the words of the below sentence. Then mail it to him.

    Mr Ben FoRd. wE knOW nameS of your cHildRen- and We kNow wheRe theY go to scHool….

    https://www.zerohedge.com/political/shocking-reversal-wayne-county-election-board-republicans-rescind-certifications-claim

    https://www.steynonline.com/10762/an-ambulance-with-no-chasers

    Thanks Steve for playing your part in the restoration of democrat, er, oops, I mean democratic norms.

    Yours,

    Mr Wayne County (D)

  17. OT- great film review….

    https://www.timesofisrael.com/soros-squanders-an-opportunity-with-poor-picture-of-controversial-billionaire/

    ‘Soros’ squanders an opportunity with poor picture of controversial billionaire

    The documentary, out in select cinemas in the US and virtual theaters on Friday, is a one-dimensional advertisement for George Soros, and less informative than his Wikipedia page
    …………………

    Directed by Jesse Dylan (Bob’s son), this portrait of the 90-year-old Hungarian-born billionaire is not, in any way, a “real” film. It feels like one of those videos they make you watch before you look at a time share. It is an 86-minute advertisement for just how great George Soros is, and an intensely boring 86 minutes at that. While some of the Great Man’s offspring do show up to heap praise on Pop, you will not glean a single thing about Soros’s inner life, what makes him tick, or how he actually got all his money. But you will come away, so Dylan hopes, convinced that George Soros is the greatest human being who ever lived, and anyone who says different is a simpleton.
    ……

  18. Thoughts says:

    I had never heard of Benford’s Law before this election…

    When I first heard people talk about it, I rolled my eyes and went ‘Big word Name Thing finished by the word ‘law”

    And then I thought Democrat disinformation to confuse white male libertarian sillies who think they are good at Math

    The only thing I’d like more explanation on is the Dominion Biden to Trump vote Ratio that Sidney Powell mentioned….was it a weighted average…1.25 to .75 as in her example…or did the software add X votes per Y vote…That’s interesting

    I’d like more info

    At some point, I need to access the data for one county showing blatant fraud and do all the counts myself and have my own graphs and spreadsheet generator as a party trick for Family and Friends for the Holidays.

  19. @International Jew

    As others have pointed out, the Dems have no problem with this math proving discrimination, just election fraud.

  20. Thoughts says:

    My other thought of the day is…

    This is so a script guys. This is all scripted. The election against Hillary, with Trump winning. Now this election…

    What would be the BEST Script? Trump winning of course, and the U.S. burning. That would be epic movie-making. Everyone would hate each other and wear masks and fun would be had by all!

    Trump voters would have purpose, David and Goliath, and Liberals would have purpose hating Trump voters. Who needs Football or Religion when you have A Political Scripted Soap Opera…that you’re a key part of!

    Trump hasn’t done enough because he is controlled. He is a part of the Matrix.

    The real question is…Is there any chance of Escaping the Hollywood Script? Is there any chance of getting Real Leaders? Or are all the gates closed by the gate keepers in this country?

    We need to anti-Matrix the U.S.

  21. @International Jew

    why is it that Trump-leaning precincts do follow Benford’s law

    I wish I had the link, but others have found that voting precincts in other elections tend to follow Beneford’s Law.

  22. Polynikes says:
    @International Jew

    That’s the question. Why do other tallies follow the law but not Biden’s in the counties at question? Also, why has Benfords law been used in other election cases but not applicable here? Maybe there are good reasons for that. I don’t know enough about it.

    Also, you’re starting to see Trump’s legal strategy. It isn’t too switch enough votes to him. It’s to drag Biden to 269 or below, meaning the election will go to a vote of the state’s.

  23. The intriguing thing I’ve heard about Benford’s Law and the election is that the social media giants are muffling anyone who dares mention the concept and the relevant anomalies — as if ol’ Benford might be getting too close to the truth. I don’t use social media so can anyone confirm or refute that?

  24. You’re kidding right? Benford’s Law???

    is a manifestation of the human mind’s tendency to single out things for comparison. We pick the smaller quantity because we want to isolate it and hold it up before us. So, we say, “one out of ten sheep is black”, not “nine out of ten sheep are white”. We focus on the exception and try to diagnose why they don’t follow the norm. Why did one man, Ted Williams, bat 400 and not why do the majority of hitters bat 250? (or whatever). We don’t care about the 67%. We care about the 1%. So we talk about it more.

  25. pseudo999 says:

    What about 2nd digit Benford’s Law analysis?

  26. Mr. Anon says:

    OT – The WHO concludes Remdesivir doesn’t work:

    https://www.nytimes.com/2020/11/19/health/remdesivir-covid-19.html

    This is a conundrum for Facebook and Twitter. Do they ban the WHO because it disagrees with Anthony Fauci? Or do they ban Anthony Fauci because he disagrees with the WHO?

    Anyway, Gilead made a bundle. And that’s what is really important.

  27. @International Jew

    Then the question is, why is it that Trump-leaning precincts do follow Benford’s law, but Biden-leaning precincts don’t?

    Exactly. In sussing out fraud, the point is to compare the pro-Biden and pro-Trump votes.

    For example, the crazy post-election vote dumps for Biden like the reported 150K mail votes, 96% for Biden, at 6:31 a.m. on Nov.4 in Michigan, are suspicious as hell.

    Unfortunately, even prima facia evidence of vote fraud is not going to get the initial election results changed. Trump is going to lose because the standards of proof are too high, time is too short. And, most importantly, there is no realistic remedy once fraudulent votes are mixed in with the real ones.

    So the issue becomes: by pointing out possible fraud are you “undermining confidence in our democracy” as the media claim?

    In 1960 during the Cold War that might have been a reasonable view for Nixon to take. In 2020, for Trump, not so much.

    • Thanks: TWS
    • Replies: @Achmed E. Newman
  28. Michael S says:

    Benford’s Law isn’t supposed to prove conclusively whether or not numbers are real, it only tells you if a distribution appears to be naturally-occurring. And it’s an effective fraud-detection tool, not because it “must” be obeyed because of some iron universal law, but because it is obeyed, almost all the time.

    You’ve given examples of distributions that are not naturally-occurring. That’s exactly what the analysis detects – red flags. A red flag might prove to be benign; just as you can give reasons in your folksy baseball analogies why not every conceivable distribution will follow Benford’s Law, I can do the same for accounting, e.g. if invoice numbers are generated randomly instead of sequentially. But an investigator is likely to look at that and say “hey, this is not normal, nobody does this, why did you do it?” And you then should have a convincing explanation, otherwise it is likely to trigger a detailed audit, because it’s possible you did it specifically to cover up fraud.

    Baseball is not a good domain for this because most of the numbers are so small. But if you insist on straining the analogy then it would go something like this: if you sequentially number every game that’s ever been played, you find that every single team has a win distribution following Benford’s Law, except one, and furthermore that all of the skew comes from winning 98% of their home games. Perhaps there is a good reason why they play so much better at home. But, as the commissioner of Major League Baseball, you are going to want to do a thorough investigation and have several people watching several of their home games very very carefully.

    That’s what happened here. Every county in every state does follow Benford’s Law on the balloting, including absentee ballots in other counties and even including the normal ballots in the counties with late-night vote dumps! Only the dumps themselves were off. That kind of outlier should automatically trigger a detailed audit, even if we ignore all the other outliers, which I’m not going to rehash here.

    In any case, it’s two weeks late to be having this discussion because we’ve already moved into the “how”, not “if”. We’ve got a pretty clear picture of what happened in Detroit, even if the lamestream won’t report it. And I’d say it’s a pretty good bet that something very similar happened in the other disputed counties.

    • Thanks: Almost Missouri
    • Replies: @Dieter Kief
  29. ziel says:

    I learned a number of years ago about Benford’s Law – much to my horror – as I used to make up expense numbers for my wife’s small business, as gathering the data from receipts (when she kept them) was an insane amount of work. The IRS uses Benford’s law to find possible malfeasance. Seeing the error of my ways, I was actually able to make up much more plausible numbers.

    But nowadays with online banking and credit cards and Amazon purchasing I can download everything and quickly categorize it all using pivot tables in Excel, greatly relieving my stress level.

    As an aside, the great challenge for non-retail small businesses with taxes is that almost all the revenue is easily discernible since few pay cash – while with expenses, aside from payroll, it’s entirely on you to actually have it all accounted for properly, and proper accounting is pretty expensive.

  30. anon[245] • Disclaimer says:

    Off Topic

    Here is an article that outlines the sort of “abundance of caution” that was part of the elaborate non political slow walking. https://www.statnews.com/2020/10/20/dry-technical-but-important-why-an-fda-advisory-panels-meeting-on-covid-19-vaccines-matters/

    Striking that the non political Ezekiel Emanuel manages to insert himself.

    Articles like this may disappear.

  31. @International Jew

    It would be a slam dunk, if the standards were like those needed to prove racial discrimination.

    You betcha. This is a case of “disparate impact,” wherein mail-it-in voting has a negative, disparate impact on Trump voters.

    This is the logic upheld and applied by our esteemed judges.

    • Agree: Mike Tre
  32. Tiny Duck says:

    Trump actually improved on his 2016 popular vote count by — at this writing — roughly 7 million.This was after he failed to respond to COVID, botched the economy, alienated our allies and emboldened our enemies, undermined every government institution, extorted Ukraine, occupied Portland and declared war on Lafayette Square, embraced white supremac. After that , 7 million more people cast their ballots for Trump. Faced with a clear choice between good and evil, America did the right thing, barely. That is sobering and profoundly disappointing.

    And it strips bare all the glossy claims about who we are as a country, underscoring the fact that in a meaningful sense, we are not one country at all anymore, but two sharing the same borders. The last time that happened, it took four years and 750,000 lives to force us back into some semblance of oneness. Even then, the seams of the fracture were always visible.

    Unlike that break, this one is not starkly geographic: South versus North. No, this one is future versus past. Meaning that yesterday, this was a nation where white people were the majority, and tomorrow it will be one where they are not.

    The fear and resentment that inspires in many white people cannot be overstated. It has warped our politics for years, culminating in the disaster of Trump. Now Biden is elected on a promise to heal those breaks, but that will require more than a good man’s good intentions. It will require white Americans to divest a system of white supremacy that, let’s face it, has been very, very good to them.

    Unfortunately, it has been less good for the country. So a moral reckoning is required here. It is time more white Americans finally recognize that white supremacy is not something you compromise with or rationalize. It must be a deal breaker, always. And it isn’t, as evidenced by the fact that the man who called Mexicans rapists and Haiti, El Salvador and the nations of Africa “shithole countries,” who described neo-Nazis as “very fine people” and told four Congresswomen of color to “go back” where they came from just won 7 million more votes than he did in 2016.

    That’s “who we are as a people.” Let us stop kidding ourselves about that.

    And start figuring out how to become something worthy

    • Replies: @anon
    , @Bill
  33. Since 1980, out of the thousands of counties in the US, 19 counties have gotten the Electoral College right.

    This year, we’re expected to believe only 1 was right.

    We’re also expected to believe that the average accuracy of the most 174 accurate bellwether counties (since 1976) could also plummet from 85% accurate to 16% accurate.

    The main stream (Wall Street Journal) explanation for this is that the country has become too polarized for these bellwether counties to matter much. That, of course, is a total crock – why they were all right in 2016, when the country was hardly less polarized?

    That, plus the fact that the last time an election featured one candidate losing despite winning both Ohio and Florida was 1960, which was our most recent aggressively stolen election.

    But Sam Giancana had a funny way of helping out in karma’s reward for JFK.

  34. Ano says:

    Hi Stevie!

    Benford’s Law reminds me of Bendover’s Law: the firm where, as a practising barrister, I worked under all sorts of powerful influential men, giving a lot of satisfaction with my performance (and for every blow I received in my climb up the ladder, I gave far more to advance my career!), while all the while spreading myself wide in gaining experience in the intersexting worlds of law and politics.

    XOXO

    Your all-American Kamala

    • LOL: TomSchmidt
  35. @International Jew

    The other real question is, why do Biden’s results only or primarily violate it in the controversial swing states, plus Chicago (which has a certain tradition of voter fraud)?

    As for PA, you can’t just look at Philly, by the way. This was a statewide scam here.

  36. @John Burns, Gettysburg Partisan

    Since 1980, out of the thousands of counties in the US, 19 counties have gotten the Electoral College right in every election.

  37. J.Ross says:

    If you can physically expel court mandated observers and still crow about “no evidence” then the system doesn’t work and the election is still a failed election.

  38. George says:

    If the CIA FBI MI5 Mossad rigged the election, they would have hired people who know about Benford’s law, and how to spit out suitably random numbers.

    On the other hand if morons self organized a vote rigging…

    Judge of elections charged in darkening of ballot choices, Lehigh County DA says
    https://www.msn.com/en-us/news/politics/judge-of-elections-charged-in-darkening-of-ballot-choices-lehigh-county-da-says/ar-BB19IDYl

    Judge Erika Bickford, also known as Everett Edward Bickford and Reverend Everett E. Bickford, was possibly sifting through ballots with no oval filled for certain offices, probably judgeships and other stuff people don’t vote on at all because they don’t care and don’t know any of the candidates and refuse to vote party line all the way down the column. Mx Brickford then possibly filled in the ovals for that office. But what if possibly there were lots of ballots with no vote for president? Perhaps from Rs that refused to vote Trump, D or waste their vote by darkening the oval belonging to Jo Jorgenson.

    Just imagine an army of Judge Brickfords.

  39. Hans says:

    “And the fact of the matter is we can’t see any physical way possible for some of those votes to have been in those kind of numbers because they just don’t have the equipment that can produce it in that timing.” – Russ Ramsland, of the Allied Security Operations Group, on Lou Dobbs – https://www.youtube.com/watch?v=LPdD8Cd5PGI

  40. @Hypnotoad666

    And, most importantly, there is no realistic remedy once fraudulent votes are mixed in with the real ones.

    Maybe there is no realistic remedy to unsort the votes, but the States and their electors have plenty of remedies available to them to get it right. I guess you could say those remedies are unrealistic too, nowadays, because the States have ceded all important power to the Feds over the last 50 years, arguably a century and a half. I don’t mean they’ve literally ceded this electoral power, but they not used to standing up to the Feds. The Ferel Gov’t says “jump” (or lower your speed limits, or send us this data…), and the sackhanging States say “high high, Sir?!” Follow the flow of the money. Amendment XVI was an evil one. (See Peak Constitutional Amendment – XVI, Part 3.)

    BTW, Mr. Derbyshire has discussed Benford’s law in his columns more than once. I wonder if this part-time mathematician could chime in here? Hello??

    • Agree: JimB
  41. @ziel

    The IRS uses Benford’s law to find possible malfeasance. Seeing the error of my ways, I was actually able to make up much more plausible numbers.

    Now, there’s some “news you can use”, as they say! Thank you, Ziel.

  42. ic1000 says:
    @Buzz Mohawk

    Steve, these baseball examples are good illustrations of where Benford’s Law shouldn’t apply.

    (I’m no mathematician, but) the idea is that when collections of things span large numbers of orders of magnitudes, the first digit will — non-intuitively — be a “1” much more than one-ninth of the time. (Consider that the first digit will never be a “0”, as then you could argue that it’s always a “0”.)

    The set of the list price of the items by SKU at the Dollar Store won’t work.

    The set of the list price of items at Amazon could follow this (quasi) Law. The price range probably runs, fairly evenly, from 10^0 dollars to 10^2 dollars, so that’s three orders of magnitude.

    (But there are probably deviations, as stores don’t set prices randomly, e.g. $9.99, $19.99, $49.99, and $99.99 might be favored points. If this was correct, we’d see positive deviations from expected values for 1, 4, and 9, with negative deviations for the other six digits.)

    Going back to baseball, the number of games won by pitchers runs from 0 to 511:
    1-99 — who knows
    100-199 — 503
    200-299 — 94
    300-399 — 22
    400-499 — 1
    500-599 — 1
    600-699 — 0
    700-799 — 0
    800-899 — 0
    900-999 — 0

    This is about as non-random a collection of numbers as you can find. And it fails to cover a sufficient range of orders of magnitude.

    The other baseball examples have the same problem, e.g. team rosters cluster in a very narrow range.

    If practical limits impact the size of precincts so they tend to be split in two once reaching a certain size, that will invalidate Benford’s Law as well.

    On the other hand, some collections of vote numbers don’t seem to disqualify for this reason. Pennsylvania seems to have tallies (townships/cities? counties? precincts?) that run from a couple dozen voter to many hundreds of thousands. So that data set might be expected to follow Benford’s Law. As would the subordinate sets of Biden vote numbers and Trump vote numbers.

    • Replies: @Anon
    , @prosa123
    , @Jack D
    , @Roger
  43. Trinity says:

    Here is a baseball statistic on pitchers. The last two pitchers to win 20 games and lose 20 games in the same season were Phil Niekro 21-20 in 1979 and Wilbur Wood who went 24-20 in 1973. Before Wood, the last person to win 20 and lose 20 in the same season was Walter Johnson in 1916. No one since Niekro has accomplished this feat. In fact no one since 1979 has even had 34 decisions. Poor Phil, imagine the number of wins he would have had if he had played with a contender instead of the hapless Braves for most of his career. Even playing for the pathetic Braves for most of his career, Phil managed to pull out 318 wins which is an accomplishment no matter who you played for, but unbelievable when you play on some of the teams Niekro pitched for back then. I think it is safe to say that you will never see a 20-20 pitcher in baseball again.

  44. Don’t confuse an exponential or log-normal distribution for Benford’s law. The baseball examples are not applicable without considering the leading digits of EVERYONE; in other words, you need to include the guys who had 10-19 wins and 20-39 wins, etc.

    • Agree: ic1000
  45. @ziel

    Talking about IRS reporting, part of how the Affordable Care Act (Obama Care) attempted to gather more tax revenue to pay for itself was to increase the reporting requirements for “cost basis” of investments.

    What this means is that if you have a mutual fund through T Rowe Price, the burden is on the fine people at T Rowe Price to calculate your “basis”, that is your initial investment which is deducted from what you get from selling your shares in calculating how much capital gains tax you owe.

    The idea was to “harvest” more tax revenue, but I am perfectly fine if T Rowe Price does the complicated bookkeeping, especially with a mutual fund on which you are growing the investment by automatically reinvesting dividends. I am happier being in compliance with new system.

    The problem is that many of us “little guys” were encouraged to invest in mutual funds and other financial instruments other than bank deposits, certainly since the 1980s. You buy the mutual fund, let it reinvest dividends (and the fund reinvests capital gains from trading on its portfolio) and then you watch it grow. Good luck trying to figure out on your own what your cost basis is when it is time to sell.

    One solution is to die, and then your heirs (at least until the Biden/Harris Administration changes this) gets a “reset of basis.” Easy peasy. But if you want to sell shares you have held before the ACA went into effect, what are people doing? I see these accountants on the Web with their vanity Web pages giving out investment advice essentially scolding you, “You should have kept records.”

    I guess one can ask T Rowe Price for help, and maybe they will supply this information. Another mutual fund company said “do it yourself”, the promised a document dump of statements back to the 1980s to help with this, but they never mailed it.

    With the IRS doing Benford’s law audits, what do people do?

    • Replies: @scrivener3
  46. The expected frequency of a leading digit under Benford’s Law is log(d+1) – log(d), so it’s about .3 for 1, for example.

  47. ic1000 says:

    Here’s an explanation at brilliant.com.

    Benford’s law is an observation about the leading digits of the numbers found in real-world data sets. Intuitively, one might expect that the leading digits of these numbers would be uniformly distributed so that each of the digits from 1 to 9 is equally likely to appear. In fact, it is often the case that 1 occurs more frequently than 2, 2 more frequently than 3, and so on. This observation is a simplified version of Benford’s law. More precisely, the law gives a prediction of the frequency of leading digits using base-10 logarithms that predicts specific frequencies which decrease as the digits increase from 1 to 9.

    Their image that shows why a collection of random numbers that span a wide range of orders of magnitude should have more “1”s than “2”s as first digits (and so on).

  48. Anon[801] • Disclaimer says:
    @ic1000

    Even if you violate the assumptions of a statistical test it does not invalidate the test. It only means you have to be more careful about bias in your conclusions. Look at Democrat counties of the same size and party affiliation where no cheating is alleged and you will find they follow Benford’s Law. Only in the disputed counties and only for Biden do they violate Benford’s Law. Not Trump, not the independents. Only Biden.

    • Agree: Desiderius
  49. @ziel

    But nowadays with online banking and credit cards and Amazon purchasing I can download everything and quickly categorize it all using pivot tables in Excel, greatly relieving my stress level.

    In our little business in the olden time, to borrow the Churchllian locution, all paper receipts would be scotched taped to an 8×11 piece of paper and allocated to a specific “job” (even if they weren’t) and filed under that job.

    Revenue from each job for a given year would get its own “every other” column on a Lotus 1-2-3 spreadsheet. Then, at tax time, we’d pullout each file, enter in the receipt numbers in cells below, by job, put in IRS expense categories in adjacent cells, and do a LookUp table to sum everything up by category.

    It took basically two days and was a fucking nightmare, but it worked.

    Today, we charge all of our expenses to an American Express card, and nothing BUT business expenses, and getting everything organized is a snap. Still use my lookup table format, but it’s all copy-paste-totaled, done.

    • Replies: @ziel
  50. So per your baseball examples, and if I’m understanding this at all, Mario Mendoza is the most famously hapless Benford distribution — hitting in the .100s five seasons and in the .200s — barely — four times as he walked a tightrope along what we now fondly know as The Mendoza Line.

  51. @Achmed E. Newman

    Maybe there is no realistic remedy to unsort the votes, but the States and their electors have plenty of remedies available to them to get it right. I guess you could say those remedies are unrealistic too, nowadays, because the States have ceded all important power to the Feds over the last 50 years, arguably a century and a half. I don’t mean they’ve literally ceded this electoral power, but they not used to standing up to the Feds.

    The “remedy” unfortunately would require balls of steel.

    Republican legislatures in Michigan, Pennsylvania, and Georgia would have to declare the elections in their states invalid, and either refuse to send electors or send Trump electors.

    That’s been the Giuliani-Powell strategy all along. There is simply no way to get true accurate recounts or audits — too many pieces of paper, too little time, too many barriers to actually investigate vote by vote.

    Should the legislators act with the ONLY remedy available, on would come the death threats, the antifa demonstrations, the doxxing, the cancelling, as sure as night follows day.

    Intimidation is VERY effective. Everybody KNOWS the entire enterprise was fraudulent, but they’ll say to themselves, “not worth it now. We’ll come back in 2 years, 4 years.”

    • Agree: Almost Missouri
    • Replies: @J.Ross
  52. Steve, Really sorry to be the bearer of bad news, but as far as this election cycle goes, Benford’s Law jumped the shark about a week ago.

  53. Steve,
    Really looking forward to your review of this. I’m sorry I wrote a negative review of the Half-Blood Prince. At the time, I had not read much of your blogging and don’t think I got you at all. I want to re-read it now, but I don’t want to read any of Obama’s actual claptrap.

    The obvious way to end any commentary on the Obama years is with remarks about the Trump triumph.

    https://mobile.twitter.com/NYDailyNews/status/1329520332751966209

  54. Anon[240] • Disclaimer says:

    I add this election to my list entitled “The official version of events is highly implausible, but I will never know the actual facts.” The Kennedy assassination(s); Flight 800; 911; Epstein’s death; and I could go on.

    • Agree: Bill
  55. Doesn’t limiting the analysis to pitchers with 100 career decisions render it essentially meaningless?

    If you opened it up to all pitchers to ever pitch a game in the majors I expect that the winning percentages would probably follow Benford’s Law pretty closely. I would think that the vast majority of everyone to take the mound in the majors doesn’t get to 100 starts.

  56. Anon7 says:

    OT: In a new Twitter ad, google makes it official:

    “More than 55% of Brazilians are black.
    Yet discrimination and lack of inclusion still persist.”

    Clicking on the ad takes you to a page where google is trying to get black people to become programmers and developers.

    Will it be any different here in the USSA, when the white population dips below half?

  57. ic1000 says:

    Last night, Lester Holt and his squad (my stand-in for the consensus American media) were even more stridently declarative on the subject of 2020 election fraud. Trump and his people are falsely-claiming, there is no-evidence-of, there was no-fraud, Best-Election-Ever, etc.

    Notwithstanding journalists’ abandonment of journalism’s former supposed aspirations: it’s really hard to find solid discussions of credible, fairly-considered evidence of fraud. By that, I mean examinations that are transparent, and link to open-source data, and take the idea of “controls” seriously.

    [MORE]

    For example, if a deviation from Benford’s Law in Biden vote counts in Pennsylvania in 2020 is being evaluated as evidence of fraud, parallel analyses should be performed on multiple equivalent data sets where there is reason to believe that fraud didn’t occur. Which of these follow Benford’s Law: Trump/Clinton in PA in 2016, Romney/Obama in 2012. Ohio in 2020, 2016, 2012. If deviations crop up all over the place, then either (1) this isn’t a good test for fraud, or (2) major fraud of a Benford-violating type is rampant. You tell me which you think it is.

    Some recent resources I have found useful in framing the issue.

    This 4-minute Tucker Carlson excerpt is worth a listen — he rips into Trump’s legal team for unsubstantiated claims of millions of votes switched by voting-machine software.

    Did the Voting Machines Lie?, by Bob Zeiden, The American Spectator (Nov. 18, 2020).

    Twitterer nsram discusses an affidavit filed by Russell James Ramsland, Jr. on 11/17/20, describing a set of anomalies in Michigan’s statewide voting-machine tabulations. 9-page PDF on github here.

    Steven McIntyre’s twitter feed. A thread where McIntyre contests the claim by Shiva Ayyaduri that he’s uncovered clear-cut evidence of vote fraud in Michigan.

  58. OT:

    On graph porn:

    I am seeing people discussing (on Twitter, etc) infection or hospitalization graphs of COVID now without seeing discussion of the demographics of the afflicted or any new science on the course of the disease.

    Without seeing information that suggests the disease is more than a nuisance or low-rate bad luck for the unfit and the non-elderly, I will remain unworried…

  59. I clicked this blog by mistake. I thought it was another one. I was surprised. The author of that other blog never writes stuff like this. You know – nerdy, intellectual stuff – the kind of stuff I too like to post. Then I saw my mistake and all was clear. Sailer of course.

    I should drop in more often. I used to check in on your blog first.

  60. @International Jew

    Then the question is, why is it that Trump-leaning precincts do follow Benford’s law, but Biden-leaning precincts don’t? Maybe the reason is that conservative districts are more rural, so consolidation/splitting is less convenient because of distances and administrative boundaries.

    I thought of this immediately when this data was trotted out. “Biden precincts” simply do not look like “Trump precincts”. If it’s reasonable to have say up to 800-1000 voters in a precinct, then both would have a bunch of suburban precincts that they win with varying hundreds of voters. But then Biden would have a bunch of relatively similar sized urban precincts he blows out 7 or 8 or 9 to 1. And Trump would have a huge collection of rural precincts that he wins handily, but running from full sized down to some tiny burg that collects the 100 farm families within 20 square miles.

    You really have need a comparison with 2016 and 2012, 2008, to see if there’s something uniquely suspicious this time out.

    The real tragedy was that this was close. The Democrats unleashed a wave of riots and street thuggery upon America over an open fraud–both generally and the specific (Saint George) case–an attack upon rule-of-law. Biden–obviously in serious mental decline–chose a running mate a woman whose signature issue (anti-lynching) is an openly fraudulent blood libel upon white people. And Trump ranted and raved but was unable to make this clear to white people.

    • Replies: @Michael S
  61. anon[893] • Disclaimer says:
    @ic1000

    Last night, Lester Holt and his squad (my stand-in for the consensus American media) were even more stridently declarative on the subject of 2020 election fraud. Trump and his people are falsely-claiming, there is no-evidence-of, there was no-fraud, Best-Election-Ever, etc.

    I wonder what those same journo-lists said about the theory that Russia! Russia! Russia! hacked the 2016 election for Trump? That would be a useful 1st order test of their veracity, wouldn’t it?

    Yes, these are rhetorical questions.

  62. JimB says:
    @Achmed E. Newman

    Maybe there is no realistic remedy to unsort the votes, but the States and their electors have plenty of remedies available to them to get it right.

    Think of fake votes as trash and Democrats as the jerks who throw their garbage into the recycling bin. Once collected, there’s no way to separate recyclables from garbage at the recycling plant. That’s why recounts are a waste of time. But the rigged Dominion vote machines are another matter altogether. I really think Trump should be commended for challenging this election result all the way to the electoral college. After four years of establishment gaslighting, Americans need proof their elections aren’t just a fraud.

    I’ve suspected fraud in my local elections for years. The early morning vote dump is a pretty common thing in my district for getting tax and bond measures passed. Don’t think for one moment that the screaming idiots on CNN and MSNBC wouldn’t condone wholesale theft of an election for the benefit their globalist paymasters and most favored politicians. Hunter Biden is essentially the face of Democrat America’s political opportunists, except maybe with a darker suntan.

    • Agree: Desiderius
  63. martin_2 says:

    Surely the baseball pitchers example has to do with the fact that most data sets follow a positively skew distribution, that is to say, most values at the low end and a few values at the extreme high end. Is that the same as Benford’s Law? I didn’t think it was. It “follows” Benford’s Law but I thought that his law models only cases where several orders of magnitude are involved, and was something different to the “20-80 principle”.

  64. prosa123 says:
    @ic1000

    (But there are probably deviations, as stores don’t set prices randomly, e.g. $9.99, $19.99, $49.99, and $99.99 might be favored points. If this was correct, we’d see positive deviations from expected values for 1, 4, and 9, with negative deviations for the other six digits.)

    Similar to the way that dating sites have a peculiarly large number of 29-year-old women and 5’11” men.

    • Replies: @JimB
  65. Michael S says:
    @AnotherDad

    the real tragedy was that this was close.

    It became close only after the fake vote dumps. Before that, Trump was leading by a 5-10 point spread in those states.

    That’s how the scheme works. It’s how it’s always worked. Manufacture just enough votes to win and be outside the automatic recount threshold, but not enough to be totally unbelievable.

    Trouble is, Trump had so much of a lead that coming up with “close” state vote counts still required totally unbelievable results at the county/precinct level. The result is only believable if you force yourself not to notice any details.

    • Replies: @Desiderius
    , @Dieter Kief
  66. You can use second digit analysis, too, but then you need a larger data set because the differences in frequency of the 10 digits is smaller- as you go out to 3rd, 4th, nth digit, the frequency of each approaches 10%.

    A better analysis would be last digit frequency analysis. If you start seeing 30% zeroes or 5s, or 7s, you gotta a fraud.

  67. Jack D says:
    @ic1000

    As stores don’t set prices randomly, e.g. $9.99, $19.99, $49.99, and $99.99 might be favored points.

    Stores don’t but Amazon seems to set prices algorithmically, which appears as semi-random to us mortals. I just searched for “generators” and the first 4 prices that came up on Amazon were $619.87, $919.08 (twice for two different generators with very different specs) and $468.30. And then it threw in a couple of ozone generators at $57.22 and $95.60. And if you were to try this again tomorrow, the numbers would be different. I rarely see x9.99 pricing on Amazon.

  68. johnmark7 says:
    @Buzz Mohawk

    I read some of Walter Mebane’s papers on election fraud (he’s kind of the expert for many), and the one on past Presidential elections. He concludes that Benford’s Law is highly useful at determining fraud, but then, not always due to possible anomalies in some districts.

    So he uses the Law except when he doesn’t want to (like this year).

    The B Law won’t apply in every single district, but it will in almost all of them.

    • Replies: @Prof. Woland
  69. Anonymous[274] • Disclaimer says:

    One has to wonder why these theories aren’t being brought before the courts by Trump’s lawyers….time is running out, and Trump is now trying to influence state legislatures to allocate electors to him.

    There are three possibilities I can think of: 1) his lawyers are incompetent and not using statistical analysis of this sort 2) There are no anomalies and the lawyers aren’t dumb enough to perjure themselves in court by claiming things that they know aren’t true 3) Trump lost

    My guess is that 3) is the most likely reason, but I’m perfectly happy to watch Trump burn the whole electoral system t0 the ground on his way out if he loses. If he wins by flipping some state legislatures, it’ll be popcorn time.

  70. Forbes says:
    @Buzz Mohawk

    A collection of sports trivia data surrounding the game of baseball does not strike me as a real-life set of naturally occurring numbers. By their own definition, e.g. number of wins or winning percentage, by pitchers with over 100 wins, are arbitrary limitations violating the basic premise of Benford’s Law.

  71. @ic1000

    Did the Voting Machines Lie?, by Bob Zeiden, The American Spectator (Nov. 18, 2020).

    This article is worth reading, SOLELY for the comments, where knowledgeable commenters rip Zeiden a couple or three of new ones.

    • Replies: @J.Ross
  72. anon[208] • Disclaimer says:

    What’s going on is that winning a major league baseball game is difficult, so the number of men drops off the higher you count. Many social statistics look like this, following Benford’s Law.

    I have to protest over this example. These statistics have fully 81% of the data beginning with a 1. They definitely do not follow Benford’s law, in which about 30% of the data would begin with a 1.

    Benford’s law doesn’t just say “leading 1 is more common than leading 2”; it’s much more specific than that.

  73. Jack D says:

    What’s going on is that winning a major league baseball game is difficult, so the number of men drops off the higher you count. Many social statistics look like this, following Benford’s Law.

    That’s not what is going on at all. Because Benford’s looks at the FIRST digit, you could have first digits that span many orders of magnitude. Take # of lifetime hits – there are some hitters with over 3,000 hits, a bunch in the 300s, others in the 30s, a few with only 3. As far as Benford is concerned, these are all the same. As a rule of thumb, the more orders of magnitude that the data covers, the more accurately Benford’s law applies (so # of games won is a poor example because it only covers a couple of orders of magnitude). IQ does not follow Benford’s law because it is narrowly constrained.

    If you look at the population of each city, town and village in the US, Benford’s law applies – 30% start with 1 and 5% start with a nine. It could be 1, 10, 100, 1,000, 10,000, etc. as long as it starts with 1, so it has nothing to do with the difficulty of achieve a population that starts with a high first digit. It’s just as easy to have 90 people as to have 110 people. It’s really only when you get to the millions that it becomes hard to hit the 9s but there are only a handful of cities with populations over 1 million so that has little influence on the overall average.

    Benford’s law is hard to understand if you think of integers as integers. But if you think of the distribution of the logarithms of those numbers, it makes sense. For those who are old enough, remember your old slide rule?

    Look at scales C and D, which are logarithmic . Look at the distance between 1 and 2 and then look at the distance between 9 and 10. Those proportions exactly follow Benford’s law. Benford’s law was first discovered when a guy noticed that the 1’s section of a book of log tables was a lot more worn that the other sections. It is purely a mathematical phenomenon and has nothing to do with difficulty.

    • Agree: Desiderius, AnotherDad
    • Replies: @DPG
  74. I’m hanging with Kant on this one. If the numbers don’t manifest randomly, then their apparent rank order is an artifact of the way the mind goes about its business of organizing the world.

  75. @Achmed E. Newman

    Yes, sadly, it is up to the legislatures; unfortunate because here’s their calculus:

    “If we send a slate of Trump electors despite the popular vote, the media will skewer us for subverting democracy, the 75 million reported to have vote Biden will howl in agony, and the kiddies will take to burning everything down, and we will all surely be out of jobs in the next few years. If we certify a slate of Biden electors, the 75 million who actually voted Trump will grumble about being cheated, maybe hold a rally or two, protest by having Christmas with their families, and then go back to work and continue paying taxes, and, maybe, just maybe, they might throw a few of us out of work, but they won’t take us all out.”

    This is what the Dems in general and Republican state legislators in particular are counting on. There is little upside for legislatures to correct errors or fraud even if they are real.

    But from the Dems perspective, this is a poor plot line for a colour revolution. The better choice would be to have the legislatures flip for Trump, and thereby make the case for an actual coup and sweeping “reforms” like eliminating the Electoral College and even more centralisation of power in DC.

  76. JimB says:
    @prosa123

    Similar to the way that dating sites have a peculiarly large number of 29-year-old women and 5’11” men.

    You mean 6’ men, since every guy rounds up to 6’ from 5’10”

    • Replies: @prosa123
    , @JimB
  77. @International Jew

    We know that smoking kills. Statistically, nobody will argue with that. But proving in a court of law that a particular individual died as a result of a particular tobacco company is much more difficult. Win or lose, the Democrats are now Philip Morris.

    • Agree: Hypnotoad666
  78. Neoconned says:

    Steve isn’t this the “too many kinds of specific numbers in a certain order in a huge data set indicate fraud” thing mentioned by Ben Affleck’s forensic accountant character in “The Accountant”??

    Would an actuary or forensic auditor be able to prove malfeasance?

    Or are you saying “swallow it and lump it Biden won fair and square?”

  79. @Hockamaw

    Except they planned on flipping Ohio too. Biden was here multiple times. They harvested everywhere. They had more money than Croesus. What they didn’t have everywhere was time and access to head off the massive groundswell for Trump (in reality against those behind Biden). Where they did (control of state and/or Dominion software) they went hogwild.

    By they I don’t just mean the Ds of course.

    There’s cope going on alright. If Trump goes down in the face of brazen fraud the GOP and shortly after the Republic is headed the way of Fox News.

    • Agree: Ron Mexico
  80. @Michael S

    Consider the possibility that the people behind this don’t want it to be believable this time.

    • Replies: @Thoughts
  81. Benfords Law applies inversely for CoronaHoax deaths. Viz:

    90 and older: 45%
    80 to 90: 40%
    70 to 80: 10%
    60 to 70: 4%
    50 to 60: 1%
    1 to 50: less than 0%

    Hence why, if every human had the same BULLSHIT-2020 prognosis, average human life expectancy would increase.

    Hoax. And I knew it since February.

    • Agree: Old Prude
  82. The legitimacy crisis is real.

    What isn’t is the assumption is that TPTB want it to be legitimate. If they did they would be doing to usual things to legitimize it. They’re not, ergo they don’t.

    Occam’s razor

    • Replies: @J.Ross
    , @vinteuil
  83. @johnmark7

    Benford’s law is what is used by NGO election observers and our State Department to look at elections overseas to detect fraud. It is cheap and easy and can be done from afar. Part of the unfortunate fall out with all of this is that America will never be looked to again as a fair promotor of democracy. It will not matter what Team Biden says or the do-gooders from now on. They will reap nothing but scorn and ridicule not just from fellow Americans but many of those in the developing countries that used to look to us for leadership.

    • Replies: @Desiderius
    , @Bill
  84. Anon7 says:

    OT: Sidney Powell and claims about Venezuelan software: cue the New York Times from 2006.

    U.S. Investigates Voting Machines’ Venezuela Ties

    “The government should know who owns our voting machines; that is a national security concern,” said Representative Carolyn B. Maloney, Democrat of New York, who asked the Bush administration in May to review the Sequoia takeover.

    “There seems to have been an obvious effort to obscure the ownership of the company,” Ms. Maloney said of Smartmatic…

    But after a municipal primary election in Chicago in March, Sequoia voting machines were blamed for a series of delays and irregularities. Smartmatic’s new president, Jack A. Blaine, acknowledged in a public hearing that Smartmatic workers had been flown up from Venezuela to help with the vote.

    Some problems with the election were later blamed on a software component, which transmits the voting results to a central computer, that was developed in Venezuela.

  85. yeah, not a good statistics argument from Steve. lots of issues here. plus his usual attempt to use baseball numbers as an analogy for some other set of numbers, not always applicable.

    we don’t need BL evidence to know Democrats stole the election – we have lots of other more obvious evidence. but at minimum you’d want to do BL analysis on every election, candidate by candidate, for a background set of data, then compare Biden-Harris 2020 numbers to the last 100 years of data.

    but all that does is give you yet another data point that Democrats went all out on voter fraud in 2020. as i posted before, everybody knows they cheated, including them. what they count on, every time, is that it is very hard to prove. and of course, they work hard to destroy the evidence. and even if you can prove some of it, you can’t prove enough of it to overturn a national election. and then there’s the fact that Democrat allies control lots of the courts anyway.

    the long term, continuous problem is that Republicans don’t work hard enough on election integrity, and are comfortable losing close elections to Democrat voter fraud anyway. meanwhile, Democrats work extremely hard to make sure voter fraud continues. they use power to secure more power.

    • Replies: @Desiderius
  86. jon says:
    @Anonymous

    Could you make the case, to illustrate for us how it might look?

    He is probably referring to the idea of disparate impact. Basically, if you can show the statistical anomaly in the results — i.e. you can show that some protected class is underrepresented — then you do have to show actual discrimination in the process leading to the results.
    So for this election, Trump and his team could just show all of the statistically improbable results — voter turnout, Benford’s law, etc., and then we could presume that their was fraud.

  87. anon[251] • Disclaimer says:
    @Tiny Duck

    Fake Duck! Spot the fakeness!

    Who are you, and what have you done with the real Tiny Duck?

  88. Anonymous[290] • Disclaimer says:
    @Coemgen

    What is the probability of event A, then event B, then event C? If it is low, then it is always true that the process was not random.

    Reasoning like that isn’t valid.

    On the most abstract level, any long sequence of random numbers is, once recorded, wildly unlikely. A 10 digit number (digits 0 to 9) has a probability of 1/10^10. A more concrete example is the license numbers of the cars at parking lot. What is the probability that exactly these numbers would appear at that parking lot on that particular day?

    As a heuristic, a time series is not random when it shows evidence of an externally controlled process – like all of the digits in a 10 digit number being 7, say, or all of the cars in a parking lot having the same third character in their license plate numbers.

    What you describe sounds more like a mix of random events with opportunistic responses to these events.

    One thing about probabilities that is not generally understood is that bad things (and good things) really do tend to happen in strings under some circumstances. You can observe this at filling stations. The inter-arrival time for automobiles follows a negative coefficient exponential distribution. You can see this distribution in the inter-vehicle distribution on a thruway. One gets clumps of cars. The usual distance between cars is small but occasionally you get a long distance. By the same token, the usual inter-arrival time for cars is short, but occasionally is long. Customers come in clusters. The bad thing (cars arriving too often) occurs episodically, and the good thing (cars arriving infrequently) also occurs episodically.

    Abstractly, superposition of normally distributed inter-arrival times from several sources {the cars} at at one measurement place {the gas station} will always produce the negative coefficient exponential distribution. That remains true if the phenomena are things like irregular solar cycles and meteorite strikes and volcanic action and plagues. They will, on occasion, all hit at once.

    So, certainly everything bad hitting at once is unusual, and there is more to the current problems than just cascade failure, but clumps of bad luck are apparently built into the mathematics of the universe.

    Best short story I know that captures the feel of the 1960s (or 2020 for that matter) for most of the US population was an early 1952 story by R. A. Heinlein, “The Year of the Jackpot”, http://www.weylmann.com/The_Year_of_the_Jackpot.pdf The story assumes that everything goes by cycles, but the story works if one just admits the idea of clumped bad luck.

    • Replies: @Coemgen
  89. jon says:
    @Hockamaw

    I think it’s frankly a sad state of affairs that so many on the right seem to prefer living in fantasy world than coming to terms and dealing with the reality of the situation.

    Dems never accepted 2016. Before Trump ever took office, the Dems were already spying on him. Then it was the ridiculous Russiagate, and then the even more ridiculous Ukrainegate after that. Turnabout is fair play.

  90. Wilkey says:

    1) I seriously doubt that Trump lost due to fraud. I mean I wouldn’t be shocked if he did, but I doubt it. Falsifying that many votes and getting away with it can’t be that easy….can it?

    2) However…the Democrats have been pushing for several decades now – at least since Motor Voter in the early 90s – to loosen election laws in every possible way. The party that believes in massive government spending and regulation anywhere and everywhere believes it would be oppressive to require voters to show an ID to cast their ballots.

    If you don’t want us to think you’re winning via fraud, stop pushing laws that make fraud all the easier.

    3) And then there’s their whining about gerrymandering, while supporting the biggest gerrymander of all, by far: the invasion of our country by illegals who they know will eventually vote for Democrats by an overwhelming margin.

    • Agree: AnotherDad
    • Replies: @Desiderius
    , @Reg Cæsar
  91. nothing says the other side feels confident they won fair and square, like threatening the other team’s lawyers, or threatening election official’s children. i mean, who doesn’t feel convinced that the other team won free and clear when they start threatening your family and saying that they know where you live? i know that’s when i’m convinced that things are on the up and up.

    anyway, i’ve still seen no explanation how in 2016 the race can be called in PA, MI, and WI around 11pm on election night with margins under 50,000 votes, deciding the entire election, Democrats accept it, and no votes show up a week later overturning everything. but then in 2020, Trump is winning all those states again, but they ‘stop counting’ for the night, then at 4AM, a million votes show up for Biden. that’s never happened in 200 years of elections.

    it was almost like in 2016, Democrats thought it was in the bag and they didn’t even need to cheat, and were shocked and totally caught off guard when Trump had the votes to win, and they didn’t have the made up, fake votes stashed in the back room to tilt the election to Clinton. but then in 2018, and 2020, they took no chances, and had a million fake votes ready to go in case the Trump thing happened again.

    by the way, did anybody check that work order on the ‘busted pipe’ in Georgia? which pipe? how much to fix it? and when did it get repaired?

  92. @Thoughts

    Sociopaths rule, and everyone else drools.

    That is the history of Homo Sapiens in one sentence.

  93. Polynikes says:
    @Hockamaw

    Except ballot harvesting, in many of its forms, is illegal. Not all the time but getting ballots for unregistered people or without the proper applications or identification is. Do you think these votes, that did not comply with the law, should count?

  94. @Wilkey

    Trump didn’t lose.

    If one controls the software and there’s no one stopping them it would be trivial to manufacture millions of votes, especially if one didn’t really care about legitimacy (say one dominated all the levers of power and had grown tired of democracy, or if one had a very strong motivation to de-legitimize the American example).

    • Replies: @Corvinus
  95. prosa123 says:
    @JimB

    You mean 6’ men, since every guy rounds up to 6’ from 5’10”

    You are right, it was my mistake. Two inches seems to be about the most that a man can round up his height without it being obvious in person. One inch is probably safer.

  96. Math, why did there have to math?

  97. J.Ross says:
    @Desiderius

    This, they want you to feel totally helpless, unable even to criticize them, since the criticism would be based on dead letter law.

  98. J.Ross says:
    @kpkinsunnyphiladelphia

    It’s hilarious to see Ars Technica try to claim that computer scientists doubt fraud because an enormous pop cultural literature has sprouted on voting machines (featuring videos of experts hacking them), over the years since the 2000 mess, and, like a Dominion error, all of it is completely one way. It’s always war with Eastasia: these same people have said the machines were worthless for years but now ask us to believe that they never did that. It used to be Democrats complaining about the machines.

  99. J.Ross says:
    @kpkinsunnyphiladelphia

    Should the legislators act with the ONLY remedy available, on would come the death threats, the antifa demonstrations, the doxxing, the cancelling, as sure as night follows day.

    Does nobody remember the second Obama term? They used not antifa street trash but the full force of the federal government to go after anyone who looked at them sideways. There’s no alternative here but delusion. If we allow Biden to take office then we are immolating all our property, pointing an empty shotgun at a gaggle of angry cops, and shooting ourselves in the face. We might as well go with the option where we live.

  100. What’s going on is that winning a major league baseball game is difficult, so the number of men drops off the higher you count.

    setting the artificial limit of at least 100 makes a huge difference. I’d be surprised if the number scoring in the fifties, sixties and seventies didn’t vastly outnumber those scoring in the tens and single hundreds.

    You talk of naturally occurring sets of numbers and then artificially deform it to be unnatural.

    • Replies: @Bill Jones
    , @Rex Little
  101. JimB says:
    @JimB

    You’re comment reminded me of a funny exchange in Full Metal Jacket.

    HARTMAN
    How tall are you, Private?

    COWBOY
    Sir,
    five foot nine, sir!

    HARTMAN
    Five foot nine? I didn’t
    know they stacked shit
    that high! You trying to squeeze an inch in
    on
    me somewhere, huh?

  102. Carol says:
    @TestAcctforUnz

    No, let’s all look at the source code.

    The algorithms! Let a jury chew on that.

  103. @anonymous

    Dude, it says right there on the tweet that “this claim about election fraud is disputed.” So this must be fake news. (Unlike everything else on Twitter which is not disputed and therefore totally true.)

  104. @Inquiring Mind

    Get software called “Investment Account Manager”. It tracks your securities by lots (purchase date, cost per share, number of shares). When time to sell you can pick lots and it gives a running total of the net gain/loss for your proposed sale. Just balance winners with losers (lots) and get very little taxable gain.

    When I was a young lawyer I bought some great mutual funds with monthly dollar cost averaging over many years. Reinvested dividends and gains also. 240 lots over twenty years in each of three funds. Well over 1,000 lots with distributions reinvested.

    Over the years the hot mutual funds that outperformed grew cold and lagged like zombies. I had 40-60% gains locked in because of the long time invested. I couldn’t get out or reallocate to better investments without a big tax bill and a smaller pile to invest. Then I discovered IAM. IAM downloaded all the data available from the fund websites. I ordered dumps of all earlier paper records and entered them by hand (lots of work).

    When the market swooned (2008?) I liquidated many lots with gains and many recently purchased lots with losses. I got out an embarrassingly large amount of cash with a net tax bill of zero capital gains. I reinvested the wad in new stocks (ETF’s this time so not distributions except dividends). That work entering the old paper records by hand was the most valuable work I ever did 🙂

  105. Barnard says:
    @Hockamaw

    I believe that is what happened, they collected mail in ballots, filled them out and submitted them in the early morning hours Wednesday, when they knew what they needed to do to win. All of that either is illegal or should be. Votes should not be counted from people who did not fill out the ballot and submit it personally.

  106. Roger says:
    @ic1000

    I am a mathematician, and that is correct. We expect Benford’s Law to apply is there is no natural scale to apply to the data. We do not expect it for percentages that are out of a maximum 100%.

  107. @Michael S

    So – Benford’s Law is useful for a first go at numerical anomalies. In case that an important probability for such anomalies can be detected with this tool, a detailed investigation is necessary. – Thanks Michael S., that sounds reasonable.

  108. @Michael S

    So – Shiva Ayyadurai and his team could be at something real?

    • Replies: @Michael S
  109. anon[314] • Disclaimer says:

    I’m having a hard time with the ‘Trump as victim’ narrative. People who cared passionately about his signature issues, like immigration and mideast wars … they were victims as their electorally won aspirations were thwarted.
    Trump was a victim of the deep state. Of the Democrats. Of the main stream media. Of the bureaucracy, the swamp, and on and on.
    It’s like no one noticed that Trump was elected. Not the least of whom was Trump. Who decided to lead like a outsider candidate. for god sake, the guy was president. Part of the job description is to kick some ass, as needed.
    Trump rattled the intelligence agencies in the first hectic weeks. I recall in 2008, it was a big deal when Obama was forced to surrender his cell phone for security reasons. Then Hillary’s email. They were flabbergasted when Trump started conducting affairs of state on Twitter. Along with the rest of the world.
    The one thing he was highly effective at was trolling Democrats to the extent they humiliated themselves with their excesses (not that they noticed). Trump as a victim is preposterous. The deplorables that voted for him…sure.

  110. anon[257] • Disclaimer says:

    Apparently it’s ‘ableist’ for a non autistic actor to portray an autistic person.

    Ok, so what about all of those actors in their early twenties who portray teens, is that ageist? Is Benedict Cumberbatch , a non mathematician, anti-maths because he portrayed a mathematician? Also, since Turing was gay, does that mean he should have been portrayed by a gay actor?

  111. vinteuil says:
    @Desiderius

    The legitimacy crisis is real.

    What isn’t is the assumption that TPTB want it to be legitimate. If they did they would be doing the usual things to legitimize it. They’re not, ergo they don’t.

    I.e., this is an assertion of raw power by the patricians against the plebs.

    Yes, we didn’t leave anything to chance, this time. So what are you going to do about it?

  112. @prime noticer

    Given the behavior of Kemp and Co. in Georgia what makes you think it was Democrats who stole it?

    If they were going to steal it they would have taken the House and Senate too.

    • Replies: @anon
    , @prime noticer
  113. @Prof. Woland

    My sense is that this exact fallout was a strong motivator for at least some of the people behind this, not all of whom are foreign. If I’m not mistaken the domestic side may not even be primarily Democrat.

    If this result is not one that satisfies you I’d strongly encourage you to ride things out with Trump and several of the smartest/savviest people on the interwebs who are researching the fraud as we speak. A system that can self-correct backed by an engaged populace will be a great example for the rest of the world, or at least a non-embarrassing one for those of us who have little stomach for shining cities.

    • Replies: @Prof. Woland
  114. M_Young says:
    @Hockamaw

    I believe ‘ballot harvesting’ is illegal in most states (though not California).

  115. @Bill Jones

    I guess you could fix it by having single digits start with 00. Double digits with 0.

  116. M_Young says:

    Seems to me its an empirical matter whether election returns follow Benfords law. You could look at the returns in states/regions that are more known for electoral honesty, during less meaningful elections, and see if they are ‘Benfordy’.

    • Replies: @Reg Cæsar
  117. Corvinus says:
    @Desiderius

    Yes, Trump lost.

    “If one controls the software and there’s no one stopping them it would be trivial to manufacture millions of votes”.

    Assuming something occurred is not evidence, it’s accusation.

  118. @jimmyriddle

    All the votes ought to counted at the same time.

    Republicans prevented mail-in ballots from being counted as they came in Pennsylvania and some other states, delaying the count until Election Day and thus creating more room for election night chicanery and counting last in whatever numbers were needed. And of course Trump stupidly kneecapped the US Postal Service, openly stating his goal of hampering mail delivery. They removed sorting machines and cut all overtime, significantly increasing mail transit times and opening up plausible opportunities for lots of delayed ballots to be “found” on election night during sweeps- once it was known how many needed to be found. It didn’t help that Trump’s war on the post office created a lot of enemies among USPS employees who were then motivated to look the other way at or to aid efforts to “find” the needed absentee ballots.

  119. @Wilkey

    1) I seriously doubt that Trump lost due to fraud. I mean I wouldn’t be shocked if he did, but I doubt it. Falsifying that many votes and getting away with it can’t be that easy….can it?

    Did a Dominion Voting Systems Employee Brag About Rigging the Election Against Trump?

    Patents by Inventor Eric Coomer

  120. Corvinus says:

    Exactly what I thought, and confirmed in Mr. Sailer’s cagey manner–massive voter fraud and ballot harvesting alleged by Trump and company occurred is other than deserving of a full blown pattern recognizing. Why? iSteve would have made the case in part using Benfords Law.

  121. @M_Young

    Seems to me its an empirical matter whether election returns follow Benfords law. You could look at the returns in states/regions that are more known for electoral honesty, during less meaningful elections, and see if they are ‘Benfordy’.

  122. Bill says:
    @Tiny Duck

    The fear and resentment that inspires in many white people cannot be overstated.

    Yeah, its white people who are full of fear and resentment. I mean, who has ever met a resentful black or dot indian or feather indian or jew . . .

  123. As other’s have noted Sailer’s baseball example doesn’t really explain Benford’s Law.

    I will try to give an intuitive example of the sort of situation that can produce this distribution. Suppose you are an investor with a portfolio and you want to double your money. You can do this by achieving an annual return of 6% for about 12 years. This applies whether you are starting with $1 million or $2 million or $5 million. In a certain sense the distance between $1 million and $2 million is the same as the distance between $2 million and $4 million and the distance between $5 million and $10 million. So if you are earning a steady 6% as your portfolio increases in value it will spend as much time with the first digit equal to 1 as with the first digit equal to either 2 or 3 or with the first digit equal to 5, 6, 7, 8 or 9. So if you look at the portfolio value at a random point in time the first digit is most likely to be 1 (and the distribution among digits will obey Benford’s Law).

  124. Corvinus says:
    @Reg Cæsar

    “I Know Sidney Powell. She Is Telling the Truth”

    The author of this piece said…

    I don’t have any evidence that Democrats used software invented in Venezuela to help socialist dictator Hugo Chavez steal elections. Or that the Soros-linked company Dominion was used by Democrats across multiple states to steal votes from Donald Trump to give to Joe Biden in the middle of the night, in just the strategic states where Trump was leading. I don’t have that evidence, but the fact that Sidney Powell claims she does is enough to convince me. This is a woman who gave up years of income she could have earned in her profitable practice to write, print, and promote a book no publisher would touch. Its charges were too explosive, I imagine.”

    It’s settled then! After all, Sidney Powell told radio host Glenn Beck she heard that our forces confiscated the server of an electronic voting system company in Germany with alleged ties to the president’s voter fraud accusations. “The servers at Scytl in Germany were confiscated the other day,” Powell told Beck during a 15-minute conversation related to voter fraud claims made by President Trump. “I’m hearing it was our forces that got those servers, so I think the government is now working on an investigation of what really happened.” And, of course, Trump won in a landslide by earning 410 electoral college votes, and even winning California and Minnesota!

    Wait, what?

    https://www.snopes.com/fact-check/dominion-servers-germany-seized/

    This rumor referenced a claim that computer servers belonging to the Dominion Voting Systems and/or Scytl Secure Electronic Voting companies had supposedly been seized by the U.S. Army in Frankfurt, Germany, and the served data showed that Trump had actually won a landslide victory in the Nov. 3 election. The rumor was one no reliable news outlet gave any credence to, but nonetheless, the far-right, pro-Donald Trump OANN cable channel devoted some airtime to it, as narrated by U.S. Rep. Louie Gohmert of Texas…Indeed, the claim echoed by Gohmert was a completely fabricated one. In response to that rumor, Scytl noted that they had no servers or offices in Frankfurt, nor had anything of theirs been seized from them by the U.S. military:

    • Replies: @newrouter
  125. Bill says:
    @Prof. Woland

    Every cloud has a silver lining!

  126. Michael S says:
    @Dieter Kief

    So – Shiva Ayyadurai and his team could be at something real?

    Afraid I don’t know who that is, sorry.

    Separately, to the people saying it was plain old ballot harvesting: sorry, but no. This is checkable, and it doesn’t check out. Enormous numbers of ballots do not have signatures, or do not have matching signatures, or the envelopes were “thrown out” in violation of election law, or some unconstitutional decree does not allow the signatures to be verified, or in an appallingly high number of cases, the ballots do not even correspond to registered voters.

    In Detroit, for example, there look to have been more untraceable counts (AVCBs reporting totals out of zero, yes zero registered voters, look it up, it’s right there in the official election data) than there were traceable votes. Nearly 175,000 votes apparently conjured out of thin air, roughly 75% more than the recorded turnout. You can’t harvest that much. You just can’t.

    And if they were harvested ballots, it would not have been necessary to stop the counting, then resume hours later with a sudden dump. Those ballots would have been mixed in with the good ones, and done much earlier in the election cycle, and they would have had the downballot selections filled in.

    It wasn’t harvesting. They took fresh ballots, or in some cases “copied” existing bad ballots (no signatures, no birthdates, etc.), put them into batches, and scanned them up to four or five times each. That was the only way to make up such an enormous gap in such a short time, and the only reason to board up the windows so observers couldn’t see them – why would they do that with legitimately harvested ballots which would look and essentially be authentic to any observer?

    Ballot harvesting is illegal in most states – they actually arrested a candidate in California for harvesting from homeless people, right here in this very election – but even if it was totally legal or gray-area, that’s a total shill narrative, and frankly one of the stupider ones I’ve heard.

    • Agree: Ron Mexico
    • Replies: @Dieter Kief
  127. anon[415] • Disclaimer says:
    @Desiderius

    If they were going to steal it they would have taken the House and Senate too.

    Meh. Dumb argument.

    Dude, one doesn’t fill out 10,000 ballots in a few hours by marking every office, one does it by marking the only office that matters. 90% of the effort into the 10% that matters; Pareto, right? The problem worsens when 100,000 ballots are needed by sunrise.

    Plus apparently you were in a coma during 2018 and somehow missed all the interesting late results in House races — that means there wasn’t supposed to be a need to clean up the House this time, given the polling….

    …which leads to the self-drinking Fla Vor Ade problem, where the DNCe believed its own polling and did not have adequate supplies of pre-marked provisional ballots stashed in car trunks, because they didn’t expect to need them.

    • Thanks: Dieter Kief
    • Replies: @Steve Sailer
  128. @anon

    I thought the current theory was that the stealing was done by software, not by filling in paper ballots. So why not program the software to elect more downballot Democrats as well as Biden?

  129. Thoughts says:
    @Desiderius

    PRecisely

    And if Trump ends up pulling a win off…or a recount goes to Trump

    Then this was all done on purpose, according to a Script that was written a long time ago

    We’re just watching a play

  130. Anon[280] • Disclaimer says:
    @Steve Sailer

    Apparently the guy behind the software theory is a serial con artist:

    • Replies: @Desiderius
  131. @Buzz Mohawk

    I am rather disinclined to this point of view. This is why Steve-ites always lose. The Democrats immediately organize for a globe-spanning ballot harvesting and election theft effort, and meanwhile the Steve-osphere consoles itself by discussing mathematical abstractions in an internet virtual conventicle.

    This is not going to make any difference. I like mathematical abstractions as much as anyone, but I take a rather dim view of their utility in the field of practical politics.

    • Replies: @Desiderius
  132. @Steve Sailer

    The software the Ds were complaining about just last year. Kemp’s afraid if they look too close they might find out that Stacey Abrams isn’t as crazy as everyone assumed.

    My ongoing assumption is that besides distaste for Trump the Rs have been strangely passive on everything because there’s been a deal in place to stop Bernie since spring of ‘16 and that it involved voting machines as a failsafe that of course got pressed into other service as such failsafes do.

    One thing I’m curious about is both McCain and Biden coming back from the dead in their respective primaries.

    • Replies: @Inquiring Mind
  133. @Anon

    That’s Hammer and Scorecard.

    Separate issue from Dominion, which has issues all over the place (mostly identified by Democrats/MSM).

    https://www.theamericanconservative.com/articles/the-extremist-at-dominion-voting-systems/

    It’s software, paper ballots, and everything else they could come up with. Completely high on their own supply and terrified of Trump.

  134. @Intelligent Dasein

    We don’t always lose.

    Quite the contrary.

  135. @Steve Sailer

    Plan A was to juice the totals across the board with a combination of ballot harvesting (which is illegal in most cases) and tabulation “errors” where the numbers weren’t quite working out. They were grooming us to accept these results with planted stories about the “Red Mirage” and the constant barrage of fake polls. A major tell for this fraud was the weird debate questions about accepting the results of the election.

    Plan A didn’t work because it was overwhelmed by the surprisingly huge Trump turnout. Trump was ahead by a huge number of votes with 85% of precints counted in Pennsylvania, Wisconsin, Michigan and Georgia.

    Plan B then became necessary. Stop the count in all of those states so they could airlift in additional ballots to comingle with the remaining ballots in rotten boroughs. These had to be rushed so did not have time to put all the downballot races. Resume the count. Ram it home. Under no circumstances let observers watch the count. Rely on friendly/compromised local election officials and handpicked Secretaries of State to certify the count and make sure that the actual ballots are never audited. Rely on total media and tech censorship to suppress the obvious statistical proof of fraud. Use the simmering threat of collective violence to cow decent people trying to do the right thing.

    Ultimately it doesn’t matter what you, or I, or Tucker Carlson thinks about the evidence, it only matters what the Supreme Court thinks. Or more accurately it probably only matters what Neil Gorsuch thinks. Assuming he does the right (and heroic) thing and sides with Trump and the will of the actual voters, it will bring us to Plan C.

    Plan C is mass collective violence on an unseen scale and a full color revolution. The tech companies will be complicit and will throw the kill switch on dissenting thought, to include the President himself.

    Trump’s plan to check Plan C is to invoke the insurrection act. Hence his firing of Esper and the reorganization of the Pentagon. He will need the people to rally behind him.

    Whatever emerges on the other side of this will mark the beginning of a completely new era. We will either be a Chinese colony or a free people.

    • Replies: @Dieter Kief
  136. Correlation Between [party] Support Rate and Fertility Rate

    • Agree: Desiderius
  137. Michael S says:
    @Steve Sailer

    The question is – whose theory? A lot of people are pushing a lot of different theories, some of them totally nuts (Hammer/Scorecard) and some totally mundane.

    The theories worth looking at are the ones in the Trump suits in Michigan (now withdrawn, since the county officials refused to certify) and Pennsylvania. And these suits allege a wide variety of different types of fraud. It wasn’t just one thing.

    There’s also a nice compendium, sorted by importance, or at least someone’s opinion of importance, at hereistheevidence.com.

    What it all points to is that there wasn’t really a centrally-coordinated grand design for fraud, but there were marching orders given – formally or informally – to all of the lower-level staff, saying, in effect, “don’t let Trump win, use any means at your disposal”. Or maybe there was no marching order and it’s just terminal TDS. Either way, that’s how you explain all of the different factors including the incredibly wide reach (between 4 and 8 states depending on your degree of skepticism), the disparate yet oddly similar types of “errors” all somehow pointing in the same direction, the need to stop counting for several hours before closing the gap, the incredibly sloppy and obvious manipulation in many cases, and the fact that it only took place in the Democrat-controlled counties of the states that were leaning but not overwhelmingly Trump.

    Central orders but local planning and execution. Trying to pinpoint it to just one type of fraud is wasted effort, but it’s how Democrat shills want to frame it – i.e. “don’t try to add up all the little errors, it’s impossible for those to swing anything, you have to show us one single instance that changed the outcome!” And they know this is impossible because that’s not how it was done. Although statistically speaking their outcome was literally impossible, they want to rig the evidence game just like the election itself, by not allowing anyone to put together all the individual pieces of the puzzle.

    There may or may not have been software-based rigging; I’m skeptical. However, if Team Trump actually did get their hands on the raw data somehow (i.e. if this bizarre stuff about server seizure is legit), the bombshell might not be in some obvious “rig the vote” backdoor, but in the ability to identify the actual fraud from that data – i.e. all of the dead voters, all of the missing or incorrect signatures, all of the presidential-only ballots, all of the duplicate and triplicate tabulations, all of the votes that came in 3 days late, etc. Dems have been working very hard to destroy or block access to a lot of that evidence, but if someone got access to the database itself..
    well, that changes the game.

    • Replies: @Dieter Kief
  138. @Thoughts

    Agree.

    7 Ways To Tell If You Have Too Much Plot In Your Story

    Populist president loved by the people but hated by the elites. Elites mobilise the Media against him. Russiagate. Global pandemic shows up on election year. Anarchists riot for months on end. Quarantine rules require change in time tested voting procedures. Prez seemingly wins as people go to bed. In the middle of the night votes show up and he loses. Vote fraud is investigated. Media ridicules that idea.

    Will the people’s Prez’s efforts to prove a stolen election by the elites that is rightfully his be successful? Stay tuned for the next exciting episode…

  139. @Desiderius

    “If they were going to steal it they would have taken the House and Senate too.”

    this is exactly what they did in 2018. as i’ve posted a hundred times now. that was their test run for their new voter fraud initiative. within a few weeks, they were able to flip nearly every close election.

    in 2020, they were focused on getting rid of Trump. not the down ballot races. which, again, is just MORE evidence that they did it. what happened, Trump getting beat by a lot, but the down ballot Democrats losing lots of ground, is something that never happens. Democrats had a coordinated effort to just submit tons of ballots with votes only for Biden, and no down ticket races, because it had to be done quickly. say, at 4AM?

    all the circumstantial evidence supports the hypothesis that the Democrats pulled off one last, huge, coordinated voting fraud effort, to defeat the Republicans at the national level for the final time. now, they’ll never have to ever worry about Republicans ever again, and nobody will ever investigate them, ever again.

    even their behavior post-election, doing anything they can to avoid a fair look at the numbers, is in line with cheaters knowing they cheated, and threatening you if you try to check their work. people who are confident they won fair and square don’t threaten your children if you simply ask to check their work.

    • Agree: Desiderius
    • Troll: Corvinus
    • Replies: @Desiderius
  140. @prime noticer

    If you look at how the GOP has acted in the aftermath, including their official mouthpieces like NRO and Fox, I find it hard to believe that this was the Ds alone this time, or even primarily. Certainly the early, unjustified Fox call followed by a complete reversal of previous momentum is exactly like 2018 and a big tell that something strange is going on, but the fact that this time the late House races are breaking R while last time they went all D is a strange one.

    • Troll: Corvinus
    • Replies: @Dieter Kief
  141. Eagle Eye says:
    @jimmyriddle

    All the votes ought to counted at the same time.

    Seven Simple Rules for REAL election integrity:

    (1) Voting IN PERSON only, at polling stations only.

    (2) VOTING AGE 40-70. Must have passed 2 out of 3 independent reading and knowledge tests each designed to screen out the bottom 5%.

    (3) VOTER ID. Voters must present picture ID obtained 2+ months before election. Strict and ongoing checks to ensure that all holders are U.S. citizens. Felons, bankrupts and welfare recipients don’t vote.

    (3) PAPER BALLOTS ONLY. NO VOTING MACHINES, NO COMPUTERS.

    (4) BALLOTS STAY at the polling station. No moving/touching/burning/churning of ballots by SEIU thugs.

    Ballots are are COUNTED BY HAND at the polling station. Results for each race are tallied on hand-written sheets maintained by multiple independent tally managers. Statewide canvassing of tallies by live phone call or transmission of handwritten tally sheets.

    (5) Ballot boxes made of PLEXIGLASS, ballots remain visible throughout the day. For extra security, single-use, see-through ballot boxes with multiple security markings. Boxes are broken open at the end of the day.

    (6) INDEPENDENT ELECTION AUTHORITY. Regular election work is handled by one-time, well-paid citizen administrators. Only a skeleton permanent organization which is independent in terms of staffing, budget etc., similar to the judiciary but with REAL independence from the executive.

    Elections Authority staff must be between the ages of 40 – 60. Anyone who was affiliated with any other branch of government within the past seven (7) years is ineligible.

    (7) SPECIAL ELECTION JUDICIARY. The Election Authority also includes an Election Judiciary. Regular judges do not get to hear elections cases.

    • Agree: Corvinus
    • Replies: @James Braxton
    , @TheJester
  142. @Michael S

    Interesting, Michael S. – thanks! – What you point out is how things like that could be done – and that is very revealing. You’re obviously able to – see, kinda, what might have happened. – That’s what’s badly needed here – and this task is traditionally fulfilled by – journalism. But it looks as if this was then, and now ‘d be rather different. I’m baffled – again and again.

    Here is the mathematician Shiva Ayyadurai who says he has detected statistical artifacts which clearly hint at election-fraud – – – :

    ttps://www.youtube.com/watch?v=Ztu5Y5obWPk

  143. @Michael S

    Although statistically speaking their outcome was literally impossible, they want to rig the evidence game just like the election itself, by not allowing anyone to put together all the individual pieces of the puzzle.

    That’s – excuse me – Shiva Ayyadurai’s team’s claim (s. my post No. 146).

    Dems have been working very hard to destroy or block access to a lot of that evidence, but if someone got access to the database itself..
    well, that changes the game.

    Would you want to say a bit more about this?

    (Great posts, btw.)

    • Replies: @Michael S
  144. @Desiderius

    Hi Desiderius – Ron Unz gave an interview about the election, and said, that he thinks that 95% of the US billionaires (the vast majority of the media elite) had a consensus that Trump must be history – the sooner the better. What you say hints in this direction, doesn’t it?

    Btw. – Ron Unz says that Trump in his eyes is no big loss, because he acted most of the time “laughably incompetent”.

    One thing springs from my head at this moment in discourse: The rather – may I say: rough – way Trump handled CO-19. – Lots of them billionaires are old (= at a higher than average risk). – If this idea has something going for it, it would explain the – concerted actions – we see going on. – Deeply rooted interests like fighting off a nasty disease might well make people tick in the same way – synchronized like clocks – or an orchestra (you don’t need a conspiracy if a public sphere has reached this state of mind – that’s what I want to point out here).

  145. ziel says:
    @kpkinsunnyphiladelphia

    That was a very organized approach to receipts, far beyond anything my wife might have been inclined to do! But well done!

  146. @Steve Sailer

    Joe Mailed-it-in.

    The total number of votes Biden received should raise some suspicions. Was Biden more motivating than Hillary? Than Obama? Why didn’t any of these people show up at rallys?

    A single cheater would minimize the footprint to hide the trail. I’m agnostic on the electronic stealing, but also don’t believe in vast conspiracies. Most likely is multiple independent efforts with the same objective overshot the mark, leaving Joe with an improbable total. Cheating may not be trivial, but I’d bet there are multiple ways, depending up on the venue.

    I would find unusually low rates of mail-in ballot rejection compelling evidence of government malfeasance, but am having trouble finding an authoritative statistic. I also have zero trust in the electronic touch screen systems. Mail in, in general is fraught with issues of chain of possession, coercion, harvesting, and outright fraud. Where are the observers for Mail-in?

    PA for example, eliminated signatures, claiming the last 4 digits of the social were better. (I guess they hadn’t heard that the credit bureau hacks exposed more than 1/3 of the US population socials) PA requires signature at the actual voting booth, but does not require an ID

    The inability to find clear stats without being an expert data miner sets off alarms.

    The fundamental problem is lack of transparency.

    IMHO, an ideal system starts in person voting with strong ID and a physical token, and a physical vote trail for auditing.

  147. @Dieter Kief

    I can’t buy into Trump “rough handling” of CO-19.

    Operation warp speed was a moon shot that required moving mountains of government roadblocks. It worked! and was likely the only winning play. I didn’t think it would work, but he guessed right.

    The main argument was that Trump failed to act more strongly, instead deferring to the states.
    1) I’m not sure stronger centralized action made any sense since no one knew what would work
    2) anything he did would be resisted by the left suffering TDS
    3) he more or less proved he wasn’t a power hungry tyrant (unlike PA governor Wolf, or MI governor Witmer who decided to legislate without the legislature)

  148. Log scales are approximately Benford distributed, but not quite. Nonetheless, they are very close and provide an intuitive notion of how the law works.
    The best summary of the generative process was by Hill American Scientist. 86 (4): 358. Bibcode:1998AmSci..86..358H. doi:10.1511/1998.4.358.

    Steve’s example with pitchers shows how to make Benford fail,
    “who have won at least 100 games”
    This artificially narrows the range. The scale invariant distances model requires at least several decades of data to work. If we took all pitcher who ever pitched, we would have a lot more 5s,
    5
    5x
    5xx

    and 1 would almost certainly become very common
    1
    1x
    1xx

    Processes that artificially limit the range will make Benford irrelevant, and this can easily happen with voting districts. Nonetheless, I’d bet there are numerous ways to examine the data for fraud. The problem is they require a baseline to validate, and 2020 was an anomaly.

    When auditing is hard, careful execution becomes more imperative. The actions of PA governor and SoS are consistent with deliberate attempts to hide massive mail-in fraud.

  149. @Dieter Kief

    You’re being far too generous.

    The saving grace is that it is neither conspiracy (they’re entirely open about it due to their obliviousness that anything could be wrong with it) nor theory (but rather fact for the same reason) so the good old CIA magic words won’t work here.

    The white hats need to recognize that the burden of proof is on them not Trump because the way this played out checks all the boxes for illegitimacy. If you want legitimacy going forward you’ve got to come up with a plausible narrative or get things fixed through the established processes (I.e. revotes/real audits).

    • Replies: @Corvinus
  150. @Dieter Kief

    The only rough thing about it was the comms and personnel (like everything else). No other president would have done anything differently aside from far less alacrity than Trump demonstrated on the key vector – treatment and vaccine.

  151. @Eagle Eye

    Good list of suggestions.

    I would add bar codes or some sort of unique identifier on each ballot. Not tied to a particular voter in order to maintain secrecy. But this would allow us to know how many ballots were issued to voters and to know whether any precints were coming up short (or long) in their counts.

  152. • Replies: @Michael S
  153. Michael S says:
    @Dieter Kief

    I see your link now. It is an hour long video, however; maybe you or someone else wants to summarize the key claims.

    Anyway, sure I can say more, although because Steve posts a lot of content, the shelf-life of a comment on a 3-day-old post is pretty short, so I’m going to forego all the nitty gritty and citations.

    In a nutshell, absentee ballots normally have a lot of safeguards, which is how the media can claim that mail-in voting is fine:
    1. They still require voter registration; can audit absentee requests against registered voters.
    2. They need to be explicitly requested, to protect against “harvesting” and other scams. This also acts as a check on their physical address.
    3. There need to be three signatures that match: ballot, envelope and voter registration.
    4. The signing must be witnessed; this is in theory a check against intimidation and of course forgery.
    5. They have to be received by election day and counted on election day – for the same obvious reason that physical polls are only open one day, if you can still act after the outcome is known then it’s very easy to cheat.
    6. Rejected ballots are rejected, period. The system prefers a few false negatives over many false positives.
    7. Observers are supposed to observe the counting (not talking about poll-watchers here) so that the counting itself is above-board.
    8. County and state officials must certify the vote.

    So, what have we seen the Democrat and cuck-controlled states and counties doing? Well…
    – They implemented (unconstitutionally, without a legislative procedure) same-day registration, eliminating safeguard #1 and weakening #3.
    – They sent ballots to every registered voter, living or dead, eliminating safeguard #2. Some Republicans comically reported receiving 7 or 8 ballots.
    – They banned signature checks, or destroyed envelopes, eliminating safeguard #3 and by extension #4.
    – They implemented absurd “curing” schemes, also killing safeguards #3, #4 and #6 and seriously compromising #5. (Voters who didn’t actually vote, but hate Trump, can “cure”, and we can’t even prove that the real voter was contacted)
    – They (unconstitutionally, again) extended deadlines and counting by several days, eliminating safeguard #5.
    – They kicked out observers or even boarded up the windows, obviously to eliminate safeguard #7.
    – When it looked like #8 might kick in, they started a massive intimidation and railroading campaign to force certification. (Fortunately, they lost, at least in Wayne County).

    All told, they systematically dismantled all of the safeguards for absentee ballots in this election and only this election (“because covid!”) and destroyed any possible audit trail. As I mentioned, those “absentee” ballots in Detroit didn’t even have names, never mind signatures – how can you audit that?

    There’s much more to it in terms of the mechanics of the steal. But tampering with a security system is usually a felony on its own – for mere subjects like us. For Democratic officials, apparently it’s a divine right, which no one is permitted to question.

    • Replies: @Corvinus
  154. Michael S says:
    @Desiderius

    There is another explanation, aside from algorithmic manipulation, for which there is real evidence: they were scanning the same batches of ballots over and over again, racking up the tally, and “committing” the batch only after reaching hundreds or thousands of votes (it’s supposed to be 50).

    Algorithmic manipulation would actually be smarter than that, in my opinion; no reason to use a fixed ratio like that. It would be harder to detect. But stupid apes duplicating ballots, that comes through loud and clear.

  155. Coemgen says:
    @Anonymous

    Are you trying to obscure the facts that germ warfare exists and that the SARS-CoV-2 pandemic took root at the same time the impeachment began to go seriously awry?

  156. Corvinus says:
    @Michael S

    “So, what have we seen the Democrat and cuck-controlled states and counties doing? Well…”

    Your laundry list are accusations, not evidence. You have to bring in hard data in each specific case to prove each point. You are basically stating your opinion that safeguards were circumvented.

    • Replies: @Michael S
    , @vinteuil
  157. Corvinus says:
    @Desiderius

    “The white hats need to recognize that the burden of proof is on them not Trump because the way this played out checks all the boxes for illegitimacy.”

    No, the burden is on Trump’s legal team. They have affidavits. They have allegedly the servers in Germany. Perhaps you, ic1000, res, Dr. GloomNDoom, Michael S, etc. lend expertise to the Trump Team next week when it comes to your citizen statistical analysis. It’s your duty! Tick, tock, time is wasting!

    • Troll: YetAnotherAnon
  158. @James Braxton

    Your post sounds not too unplausible, unfortunately. Hopefully, I don’t know enough about this stuff to see the holes in your argument. Well, I do not know much, that’s for sure. Whether my ignorance is just big enough to shield me from a bitter insight or too small to withhold me from being deceived? – Time might tell.
    What a story.

  159. @Desiderius

    A system that can self-correct backed by an engaged populace will be a great example for the rest of the world

    The most horrific part of this train wreck election is that we were warned in advance and almost all of it could have been avoided. And the worst of it is that all of this was by design.

    This is a color revolution folks. My first thoughts were that they just wanted to delegitimize the election so they could weaken Trump. We knew they wanted to cast aspersion but I did not imagine that millions of votes could be stolen or flipped like this. The Democrats have been spending all of the last 4 years and who knows how many before using this in other countries and getting it ready for here. I think they are so desperate at this point that they could not keep their fingerprints off of murder weapon.

    All of the hacks that worked for the CIA / State Department cabal are now working for Lawfare and the Transition Integrity Project (there is an Orwellian name if there ever was) lead by a scum named Norm Eisen. Learn who this guy is and it will tell you everything you need to know. Mail in ballots and Hammer and Scorecard were their ultimate ace up their sleeve.

    https://www.revolver.news/2020/09/meet-norm-eisen-legal-hatchet-man-and-central-operative-in-the-color-revolution-against-president-trump

    The changes that are needed are obvious and well within our grasp:

    1 No Mail in ballots
    2 Block Chain ballots
    3 Secure voter ID
    4 Chain of custody – no USPS handling, city workers, etc.
    5 Quick counts
    6 Open source software on everything that touches it
    7 No Ballot harvesting
    8 cleaning up voter roles so dead people cannot vote
    9 etc.

    I agree with you that America must rebuild our democracy or else we are finished but we could and should lead the way.

    • Replies: @Jack D
  160. TheJester says:
    @Eagle Eye

    I find it interesting that we landed on the moon in 1969 … yet 51 years later we can’t accurately count ballots in democratic elections. My suspicion is that certain interests in the United States don’t want to accurately count ballots. Obfuscation allows them to routinely manipulate results.

    This is personal. In 1970, my wife and I participated as canvassers in a congressional election. We were given the voter registration rolls for the Democratic Party. We were AMAZED! Voters registered at addresses that don’t exist; voters registered in houses located in the middle of alleys; voters registered in vacant lots. How widespread is this? We assumed it was commonplace.

    Yet, we hear pronouncements by public officials, the MSM, friends, and family that there is no such thing as voter fraud in the United States. Trump, as it were, REALLY lost the election. No question about it! Yet, but we saw voter fraud first hand. Yes, it was years ago, but what would have changed?

    If the above degree of voter fraud existed in a moderate-sized city in 1970, imagine what happens today in, let’s say, Detroit, Chicago, and Philadelphia in the era of electronic voting machines where recounts are impossible … or the new phenomena of mysterious bounties of mail-in ballots arriving just-in-time to turn elections.

    One definition of insanity might be to deny the obvious. What do sociologists call this when this involves large demographics … mass hysteria? If so, the current legitimacy of the Government of the United States might be described as the result of a mass psychogenic disorder. There are precedents:

    Dancing manias in the Middle Ages …
    The Dutch tulip bubble hysteria in 1637 …
    The Salem witch trials in 1692 …
    Russiagate in 2016 …
    The Presidential election in 2020 …

  161. @Bill Jones

    I’d be surprised if the number scoring in the fifties, sixties and seventies didn’t vastly outnumber those scoring in the tens and single hundreds.

    (Reference is to number of wins by a major league pitcher.)

    If you put together a complete list of every pitcher who ever won a game in the major leagues, I’m quite sure you would indeed be surprised. Thousands and thousands of pitchers have had brief MLB careers, winning far fewer than 50 games. If there was an easy way to check, I’d seriously bet $10000 that over half of all pitchers who’ve ever appeared in the majors have single-digit career win totals.

    The closest thing I could find to a comprehensive list shows that less than 1000 pitchers have career win totals of 70 or more.

  162. Michael S says:
    @Corvinus

    You ridiculous moronic shill, all of these things have been documented multiple times not only on the right-wing blogs but in the mainstream media, and several times captured on video and posted to YouTube. I don’t know why Steve or Ron keep letting you post here, I guess it’s amusing for some people to watch.

    I’m not going to effortpost for a shill who will just ignore or dismiss any evidence anyway, and doesn’t even understand the concept of evidence. Head on over to hereistheevidence.com and see for yourself.

    • Replies: @Corvinus
  163. Hence, there’s little reason to expect the number of votes recorded in precincts to follow Benford’s Law.

    The theoretical basis for Benford’s Law is a bit fuzzy,but it makes sense in a hand-waving sort of a way. What’s more interesting is Powell’s statement that the same pair of six-digit numbers (Biden votes and Trump votes) appeared in two separate states. That proves fraud, badly done, but fraud nonetheless.

    As for Benford’s Law, aren’t there other suppositions that phony numbers have a distribution that is too random and insufficiently chaotic?

    • Replies: @Jack D
  164. @Desiderius

    What does stopping Bernie do for anybody?

    Isn’t Joe going full bore on Green New Deal, debt forgiveness on college loans and a bunch of other things.

    How is Joe different than Bernie, or is the “fix” in that with Bernie it would be for real but with Joe, Wall Street knows he is only paying lip service?

  165. Jack D says:
    @Prof. Woland

    The problem is that to people who are interest in power, democracy is (as Erdogan said in Turkey) like a trolley car – you ride it to your destination (power) and then you get off. In the US, Democrats are not going discard the formalities of democracy (even in the Soviet Union they had “elections”) but they are going to corrupt the process in any place that they gain power in order to insure their permanent hold on it. Not only will they not implement your 9 steps, they will take steps in the OPPOSITE direction to make voting (and vote fraud) easier and more available to everyone. And in any place that they have manipulated the majorities to entrench themselves, they are NEVER going to allow reform. Democrat big cities have been corrupt for at least a century. The last Republican mayor of Chicago was elected in 1927.

    • Replies: @Prof. Woland
  166. Jack D says:
    @James Speaks

    the same pair of six-digit numbers (Biden votes and Trump votes) appeared in two separate states. That proves fraud,

    It proves nothing of the sort. In fact if you were going to do vote fraud you would be careful NOT to use the same numbers more than one.

    It has happened to me more than once that my total at the grocery store comes out to an even # of dollars with no cents. Doesn’t mean that the grocery store is cheating, it’s just coincidence – one out of every 100 times the last 2 digits will be 00.

    • Replies: @James Speaks
  167. vinteuil says:
    @Corvinus

    You are basically stating your opinion that safeguards were circumvented.

    You go, girl! In the immortal words of The Dude, “that’s just like, your opinion, man”

    You have to bring in hard data in each specific case to prove each point.

    Oh, yeah – just like the “intelligence community” brought in hard data to prove that Donald Trump paid Russian hookers to piss on him on a bed once slept in by Barack Obama.

    A story that many American “liberals” still seem to believe.

  168. @Steve Sailer

    I thought the current theory was that the stealing was done by software, not by filling in paper ballots. So why not program the software to elect more downballot Democrats as well as Biden?

    The downballots in PA were lower stake, and not really in danger.

    I was wrong about the coronavirus vaccine, so I could be wrong here too.

    Old school cheating is safe and effective and the government enabling, such as don’t check mail in signatures, don’t require ID at the poll in PA, and so forth, has “murder in the cathedral” type plausible deniability. Going wholesale would be a treason level offense. It’s possible, but strong claims require strong proof.

    Nonetheless, the assurances are not credible because of lack of transparency.

  169. @Jack D

    It proves nothing of the sort. In fact if you were going to do vote fraud you would be careful NOT to use the same numbers more than one.

    True, if I were to commit fraud I would be careful, but we’re talkng about Democrats here and like Nancy Pelosi, they’re either too arrogant or too stupid, or both, to be careful.

    The combination of one six digit number, lets call it 886,664, with a 2nd six digit number, say 113,336 in one state would be noteworthy, but for the same pair of six digit numbers to appear in two different states, in the same election, after vote counting had been stopped and then restarted, has odds of 1 to 10x10x10x10x10x10 or one in a million against for one of the six digit numbers to be repeated, and 1 to 100x100x100x100x100x100 or one in a trillion odds against both six digit numbers to be repeated. It’s not just the numbers, it’s how they came about (sudden ballot dumps in the middle of the night) combined with presidential overvotes (more voters than residents) and down-ticket undervotes (vote for president but not for senator or representative) leading to Republican victories in downticket races but not for president. All this combined proves fraud.

    It has happened to me more than once that my total at the grocery store comes out to an even # of dollars with no cents. Doesn’t mean that the grocery store is cheating, it’s just coincidence – one out of every 100 times the last 2 digits will be 00.

    Some people, known as stupid people, are capable of understanding “unlikely” but not varying degress of improbability. It’s like the innumerate newscaster who said, “Damage is estimated from $10,000 to $20 billion.”

    Go to the store 100 times and you’ll get an tally of even $1.oo about once every hundred visits.

    Play Powerball once per week and you’re expected one win every six million years.

    Have an election every four years and you’re expected to have one matchup of a 12 digit number once every 250 billion years. That’s 17 times the age of the universe. Can you hold your breathe that long? Please try.

    • Replies: @Dieter Kief
  170. @Jack D

    In the South, only white people could vote up until the 1960s. That is why we have the VRA. We need one now for the rest of us, not just for blacks that live in one of the 14 former Confederate States of America.

  171. res says:
    @Rex Little

    I haven’t looked it up, but I’ll bet if you looked at the career win totals (or loss totals) of all pitchers in major league history, they would follow Benford’s Law. (You don’t have to restrict it to those with 3-digit totals like Steve did in his first example; the Law applies to the first digit, regardless of how many follow it.)

    Agreed. And notice how those numbers span 3 (10 to the 0, 1, 2 power) orders of magnitude. The number of orders of magnitude spanned seems to be important.

    The number of votes in each precinct should be expected to follow the Law. Precinct sizes do get adjusted as Steve explained, but not in such a way as to keep them within a narrow range like baseball roster sizes.

    Disagree. I looked at a few cities and the precinct SIZES were very much out of line with Benford’s law. A range of something like 400-2000 people seemed common which skews the first digits badly.

    According to what I’ve read, Trump’s reported precinct-by-precinct vote totals did indeed follow Benford’s Law. Biden’s did not, at least in the swing states.

    The examples I have seen discussed the most are in areas which voted heavily for Biden. So his vote totals track closely with precinct size while Trump’s are a small fraction of precinct size and seem to end up better randomized as a result. So Trump’s votes might span three orders of magnitude (say 5-500) while Biden’s votes are only roughly in the 300-1800 range.

    I think there was fraud (how much is the question of the moment, as even the Lugenpresse seems to be acknowledging with the transition to “widespread fraud” as the talking point), but I don’t think a naive application of Benford’s law is the way to discover it here.

    • Agree: James Speaks
    • Replies: @ic1000
  172. @James Speaks

    The whole post is fresh, James speaks. But your last paragraph cracks it. A beautiful, playful and insightful post. As Kant would have said: A man of wits!

    Oh, now comes the sour part: I have a factual question too: “presidential overvotes (more voters than residents)” – could you give an example where that happened?

    • Replies: @James Speaks
  173. Corvinus says:
    @Michael S

    “all of these things have been documented multiple times not only on the right-wing blogs but in the mainstream media, and several times captured on video and posted to YouTube”

    And a good deal of them have been debunked.

    “Head on over to hereistheevidence.com and see for yourself.”

    Here is one source found there. It is a news report about a woman who was concerned about “ballot envelopes visibly identify the voter’s party affiliation”. Except it is a code to make sure the people get their ballots for the proper primary according to Florida State Law. There was no proof here that the ballots themselves were compromised or discarded by election officials. In other words, it is “suggested” that “possibly” there was some sort of malfeasance. So, while a concern about this labeling is legitimate, it does NOT constitute evidence in a court of law of voter suppression, tampering, or removal.

    Here is another source. It is an affidavit. The person says “I directly observed, on a daily basis, City of Detroit election workers and employees coaching and trying to coach voters to vote for Joe Biden and the Democrat party. I witnessed these workers and employees encouraging voters to do a straight Democrat ballot. I witnessed these election workers and employees going over to the voting booths with voters in order to watch them vote and coach them for whom to vote.”

    https://justthenews.com/sites/default/files/2020-11/JessyJacobAffidavit.pdf

    This individual has made an accusation. That is NOT evidence. Had this person supplied audio or video recordings, then that would be proof. So, this person would be subject to cross examination in a court of law to determine her credibility.

    There are dozens of similar instances by which a person says that they saw something illegal. But the collection of these examples does not mean without question that voter fraud occurred. Rather, these cases have to go before a court of law to determine if the claims can be clearly substantiated by other people who witnessed the same thing and/or physical evidence. So in the case of this person, what other co-workers witnessed it? These allegations have to be substance, not style.

    • Replies: @res
  174. res says:
    @Corvinus

    does not mean without question

    So now you are applying an even stronger standard than “beyond a reasonable doubt”?!

    I love the way everyone (well, perhaps not EVERYONE, but the MSM, Corvinus, and many others) is playing No True Voter Fraud.
    https://en.wikipedia.org/wiki/No_true_Scotsman

    Do they not teach logical fallacies in school anymore? It is nice of Corvinus to provide us all with a refresher course by using logical fallacies so frequently. That and his clarification of the current talking points are valuable services.

    • Replies: @James Braxton
  175. @Dieter Kief

    We’ll have to wait a week or two for more details.

    • Replies: @Dieter Kief
  176. @res

    The standard wouldn’t even be beyond a reasonable doubt. That would only necessary down the road if criminal charges were brought against individuals.

    Right now Trump just needs an injunction to stop everything from being certified. To get that he has to make an initial showing that it was probable that he was deprived of his rightful victory by fraud and would be irreparably harmed if it went forward.

    The statistical red flags combined with the specific instances of fraud already documented (e.g. dead people voting, late ballots being counted, etc.) make a compelling case (from what I have seen) to clear that hurdle. I think the egregious conduct that the courts will be able to hang their hat on will be the refusal to let observers watch the count. There is no way to fix a bad count once ballots are separated from their envelopes.

    A preliminary injunction would allow Trump time to pursue a proper lawsuit with the benefit of discovery to prove that the fraud did indeed happen by a preponderance of the evidence in order to get the relief sought.

    Assuming compelling evidence of fraud is presented to hopelessly taint the results in the key states (and I think it will be) I think the Supreme Court will be able to avoid simply giving the election to one side or the other by ordering the one form of relief that will be at least palatable to everyone: Revote.

    • Disagree: Corvinus
    • Replies: @Reg Cæsar
  177. ic1000 says:
    @res

    This guy on twitter, nsram, discusses statistical reports of voting irregularities. He seems willing to consider that fraud may have occurred while maintaining skepticism, demanding that analysts make strong cases to support their contentions.

    He discusses a report of that claims to document evidence of large-scale irregularities in Pennsylvania vote tallies by county in this Nov. 17 thread. It links to this Nov. 17th article at “The Natonal Pulse” website, “EXC: New Stats Analysis Reveals Potential ‘Stuffing The Tail’ Voter Fraud Scheme Similar to 2008 Sub-Prime Mortgage Meltdown. The anonymously authored 40-page PDF is at scribd, “Pennsylvania 2020 Voting Analysis Report”.

    These analysts seem to make several cases for widespread fraud. Unfortunately, they do some controls, but not others. They don’t seem to consider that they should have to show every reasonable comparable case, to make the point that the highlighted Biden-2020 oddities are truly one-of-a-kind (and thus likely the result of manipulation). I.e. they show a very strange distribution of Biden votes by precinct in Montgomery County, PA (page 19). But — not knowing either advanced statistics or the vagaries of Pennsylvania voting practices, I would want to see the equivalent graphs for Clinton 2016 and Obama 2012 and Obama 2008. As well as for Trump 2020 and Romney 2016 and McCain 2008. It would be even better to supplement this picture with controls from precinct-level returns in a “clean” state — perhaps Ohio or Indiana.

    The few technically- and mathematically- capable folks who are willing to do this work don’t seem to appreciate that people who are new to their world will generally be skeptical of getting on board with allegations of massive fraud, based on analyses that they can’t fully understand and couldn’t begin to replicate.

    The consensus media’s consensus that this was the World’s Best Election with No Evidence of Fraud doesn’t help outsiders get comfortable with their claims.

  178. @James Braxton

    Assuming compelling evidence of fraud is presented to hopelessly taint the results in the key states (and I think it will be) I think the Supreme Court will be able to avoid simply giving the election to one side or the other by ordering the one form of relief that will be at least palatable to everyone: Revote.

    On the first Monday after the second Wednesday in December, the electors meet in their respective States to cast their votes for President and Vice President of the United States. Read more about the qualifications and selection of the electors and restrictions, if any, on how they may vote.

    What happens if no presidential candidate gets 270 electoral votes?

    • Replies: @James Braxton
  179. @James Speaks

    Dr. Shiva Ayyadurai seems to have found something quite astounding already: It’s on his Twitter account the video How to Expose Election Fraud for Dummies – 20:00 min in – more votes than voters – – hm?

    Dr.SHIVA LIVE: How To Expose Election Fraud In Your Own Town/City for Dummies. https://t.co/MDuPDMS6yV— Dr.SHIVA Ayyadurai, MIT PhD. Inventor of Email (@va_shiva) November 23, 2020

  180. @Reg Cæsar

    That deadline is statutory, not Constitutional. Presumably the Supreme Court could alter the statutory timeline if it would be necessary to remedy the violation. The hard deadline is January 20.

    But these are uncharted waters.

    If no one gets a majority of the electors it goes to the house of representatives to vote by state delegation.

  181. arfoo says:
    @Rex Little

    This is just one piece of circumstantial evidence . . . . and that’s okay. with everything else u get to point where u ask that Q question: how many “coincidences” b4 it’s impossible to deny? This Benford analysis can’t “prove” that there was election fraud. It can, however, raise red flags.

    But i do have a question for the statisticians w/ confusers: might the inverse be true? if there is fraud . . must Benord’s law suggest that fraud exists? Assuming they’re not smart enough to focus on avoiding it . . . . is the law a necessary but not sufficient condition?

    • Replies: @vinteuil
  182. vinteuil says:
    @arfoo

    what’s up with guys like you who show up after a thread is dead, posting random bs?

Current Commenter
says:

Leave a Reply - Comments are moderated by iSteve, at whim.


 Remember My InformationWhy?
 Email Replies to my Comment
Submitted comments have been licensed to The Unz Review and may be republished elsewhere at the sole discretion of the latter
Subscribe to This Comment Thread via RSS Subscribe to All Steve Sailer Comments via RSS
PastClassics
How America was neoconned into World War IV
The “war hero” candidate buried information about POWs left behind in Vietnam.