Prior to the 2020 US elections, I ran a predictions tournament. Given that the Electoral College has now formally voted, putting the matter to rest – at least so far as non-ideologues are concerned – it is now time to run the results.
Best performance in terms of deviation of EC votes from Biden’s 306 (see bottom of the post for a full list of 37 predictions):
- Not Raul
- Blinky Bill (archive)
- Gorgeous George
- Anatoly Karlin
- Supply and Demand
- E. Harding
Annatar made his prediction after Not Raul, so Not Raul has priority. Both only got Georgia wrong. Congratulations to both of them.
Blinky Bill‘s prediction no longer appears because the Tweet it linked to has been deleted.
However, Audacious Epigone – who ran a parallel predictions tournament to mine – says that he recalls Blinky also only got Georgia wrong. I will take Blinky at his word if he decides to claim second place here, since he posted his prediction ahead of both Not Raul and Annatar. Preserved comment on archive.org confirms he is in joint second place with Annatar and Not Raul, and overall 2nd place if sorting by precedence. UPDATE – However, as was pointed out, Blinky Bill didn’t call the congressional districts in Nebraska and Maine, so despite getting the EC vote total correct, he did so less well than Annatar and Not Raul; as such, he drops to fourth place.
People who predicted <270 didn’t call the outcome. As such, close as some of them may have been at 268 and the like, there’s no point in extending it further.
Likewise, a pair of rose glasses to Nemets for the blackpill OD. Though Nemets at least made big bucks bettin’ on Biden on PredictIt. Good strategy, that, if you gamble, bet on the side you’re not emotionally invested in, that way you won’t feel too sad about the outcome either which way.
Given the narrow margin that Trump won by in several states, perhaps a superior way of identifying the winner would be seeing who got the fewest number of winning voters wrong in states that they didn’t call:
- Not Raul – Georgia (+11,779) = 11,779
- Annatar – Georgia (+11,779) = 11,779
- Anatoly Karlin – Arizona (+10,457), Georgia (+11,779) = 22,236
- Blinky Bill – Georgia (+11,779), NE-2 (+22,091) / ME-2 (+27,996) = 61,866
- Jayce – Arizona (+10,457), Georgia (+11,779) / North Carolina (+74,481) = 96,717
- Gorgeous George – Minnesota (+233,012), Pennsylvania (+81,660), Wisconsin (+20,608) / North Carolina (+74,481) = 409,761
- E. Harding – Florida (+371,686), North Carolina (+74,481) = 446,167
- Supply and Demand – Arizona (+10,457), Georgia (+10,457), Colorado (~+440k), NE-2, New Mexico (~+100k) / Iowa (~138k), ME-2 = ~700k
Joint congratulations to Not Raul and Annatar on joint first place out of 37 contestants! (Blinky Bill didn’t call the congressional districts in Nebraska and Maine, as was just pointed out to me).
Also congrats to E. Harding for the impressive Georgia call.
All things considered, the people who predicted around 268 generally didn’t do badly either, it was mostly tight states that they were getting wrong.
This is a 0.42% point deviation from the actual current results of 51.31% for Biden, and 46.86% for Trump, as of today’s Wikipedia page on the elections. This placed me 2nd out of 64 people (see right) who participated in such a predictions tournament at the blog of Alexander Kireev, a Russian elections blogger. (Or outright first, had he gone with the current results from The Green Papers where Biden is now at 51.27% and Trump is at 46.82%).
On that note, I should point out that Kireev’s commentariat – just as during the 2016 elections – was extremely accurate on the ultimate result, as could be expected of informed Russian commenters not particularly beset by TDS and MAGA hopium. They predicted an average of 51.3% for Biden (spot on) and 45.8% for Trump (wrong by 1% point).
Seva Bashirov also compiled a “consensus map” of theor predictions from the first 42 participants in the states-level competition – everything correct except the Georgia upset:
Here are your predictions of how many EC votes Biden would get (let me know if I mistyped anything):
- nickels – 173
- Matt Forney – 207
- Marco de Wit – 209
- A123 – 218
- LondonBob – 218 (2)
- Raphael – 226
- botafeza – 229 (2)
- Dreadilk – 229
- e. york – 232
- Paulina – 233
- AltSerrice – 236
- Svevlad – 238
- Thorfinnsson – 249
- davidgmillsatty – 256
- AP – 258
- Excal – 258 (3)
- jeppo – 258 (2)
- Sean9292 – 259
- Tyler Durden – 259 (3)
- Vendetta – 259 (2)
- The Alarmist – 268 (2)
- Auferstanden – 268 (3)
- Patto – 268 (4)
- Tusk – 268
- 270 TO WIN
- Supply and Demand – 271 [-35]
- Anatoly Karlin – 279 [-27]
- Gorgeous George – 280 [-26]
- Annatar – 290 (3) [-16]
- Blinky Bill – 290 [-16]
- Not Raul – 290 (2) [-16]
- Jayce – 294 [-12]
- BIDEN ACTUAL RESULT – 306
- E. Harding – 350 [+44]
- protoneutype – 375
- Nemets – 376
- In Catilinam – 407
- zepplin – 413