Mike Beuoy

3

31

53.3K

Mike Beuoy@inpredict·15h

@dylanmatt “Detlef Schrempf” by Band of Horses

English

3

752

dylan matthews 🔸@dylanmatt·17h

I'd read something on the multi-decade rise, from "Clint Eastwood" to "Demi Moore" to "Drew Barrymore," of songs named after celebrities whose lyrics have nothing to do with those celebrities

English

52

47

2.3K

61K

Mike Beuoy@inpredict·2d

@casualtweeter02 Colege Basketball has wacky results all the time, not just in March. Have you verified that March Madness is wackier? It's possible it is, but I don't know if I've seen the data supporting that.

English

0

243

nah@casualtweeter02·2d

I think it has to be a combination of both. March Madness has wacky results all the time. If you evaluate each game at face value, 95% of the time you should pick the 10 seed over the 7 but statistically speaking the odds 4 65-70% bets all hit is *extremely* low.

Mike Beuoy@inpredict

Probably the most common March Madness modeling error: Overfitting to past tournament results (e.g. "11 seeds have over-performed in the last 7 years") rather than just treating it like another basketball game and evaluating team strength differentials.

English

0

549

Mike Beuoy@inpredict·2d

Probably the most common March Madness modeling error: Overfitting to past tournament results (e.g. "11 seeds have over-performed in the last 7 years") rather than just treating it like another basketball game and evaluating team strength differentials.

English

7

2

265

26.3K

Mike Beuoy@inpredict·3d

@CapnNath Ed Vaguely Jr.

English

90

1.5K

Roy Donk Bet@CapnNath·3d

Ever since this 30 Rock joke 15 years ago I have been trying to come up with a celebrity-impersonator name as good as Edward James Almost. The closest I have come so far is Mary Elizabeth Instead

English

177

334

8.6K

336K

Mike Beuoy@inpredict·3d

Last year, the perfect bracket odds were about 10 billion to 1. When I looked at this way back in 2010-2014, the odds averaged around 50 billion to 1.

Mike Beuoy@inpredict

2.6 billion to 1 The odds of a perfect mens tournament bracket this year, if you picked based on @kenpom ratings and assume that those ratings represent the true probabilities. That's better odds than last year, where I estimated a 10 billion to 1 chance.

English

46

14.3K

Mike Beuoy@inpredict·3d

Survival odds by round: Round 1: 2,800 to 1 Round 2: 2 million to 1 Round 3: 67 million to 1 Round 4: 400 million to 1 Round 5: 1.4 billion to 1 Round 6: 2.6 billion to 1

English

“forklift certified” em@cowlesbian

1

9

4.6K

Mike Beuoy@inpredict·3d

2.6 billion to 1 The odds of a perfect mens tournament bracket this year, if you picked based on @kenpom ratings and assume that those ratings represent the true probabilities. That's better odds than last year, where I estimated a 10 billion to 1 chance.

English

5

8

107

42.1K

Mike Beuoy@inpredict·13 Mar

Beeswarm plot to illustrate the hilarity of Bam's 83 points.

English

2

9

108

7.7K

Mike Beuoy retweetledi

they who walk silently 🔻@meatRelic·11 Mar

evergreen

ai can’t write good because it’s designed to write pleasantly average sentences with all the right words in the right order and imho good writing requires u to strategically deploy a word that is wrong and maybe even evil

English

7

1.8K

12.9K

287.1K

Mike Beuoy@inpredict·7 Mar

@DBrianBlank It’s definitely all subjective but it does align with the Wikihoops user scores from last night. Spurs was +37, Hornets was +12. So, kinda objective in the aggregate?

English

0

1

137

Brian Blank@DBrianBlank·7 Mar

@inpredict I do not really disagree, but it is not an objective truth, is it? It was not necessarily exciting until the comeback, right? Is this just affirming your comeback preference & measuring something different than average competitiveness throughout? Not being critical, just curious!

English

0

1

42

Mike Beuoy@inpredict·7 Mar

Tonight's Spurs comeback is a good example for why KL-divergence is a better measure of game quality. Current Excitement Index (total change in win prob): MIA @ CHA: 8.6 ✅ LAC @ SAS: 8.2 KL-Divergence: MIA @ CHA: 2.2 LAC @ SAS: 4.4 ✅

Mike Beuoy@inpredict

New (and 1st) post up on Substack. Creating a better way of ranking games with the help of Claude.......Shannon. Borrowing a concept from information theory, KL-divergence, I create a new version of the Excitement Index that better aligns with how people evaluate the games.

English

0

28

4.8K

Mike Beuoy@inpredict·7 Mar

@amalbrecht10 I’m hoping by the playoffs. And yep, the Charlotte game was the second most exciting game by KL divergence.

English

1

77

Alexander Albrecht@amalbrecht10·7 Mar

@inpredict I love this adjustment. When do you think the new excitement index will be live on the site? And does the new excitement index still give the Charlotte game credit for being exciting for 3 quarters? Blowouts early on would be lower?

English

0

1

41

Mike Beuoy retweetledi

Kostya Medvedovsky@kmedved·5 Mar

Soft launching a new DARKO site/format, as I've decided to embrace the vibe-coded future. This is still in Beta and being built out rapidly, and as you can see, I'm testing out design elements in production here. This will also solve the issues with the site being out of date (breakage between the Google Sheets and the Shiny app). Please provide comments/ideas.

English

22

15

268

18.3K

Mike Beuoy@inpredict·3 Mar

@ngreenberg Claiming 91% accuracy but then picking 7 teams is quite the sleight of hand.

English

Sports Betting Tips & Insights@vsininsights

0

4

585

Neil Greenberg@ngreenberg·3 Mar

This is literally the top seven of eight shortest odds to win the title. tl;dr the oddsmakers know the best teams win the men’s tournament. News at 11.

Teams currently qualifying under the formula: • Michigan (+300) • Duke (+350) • Arizona (+500) • Florida (+750) • Houston (+1100) • Iowa State (+1800) • UConn (+1800) (All odds via FanDuel) 5/7

English

0

8

8K

Mike Beuoy@inpredict·3 Mar

@dougliebe that would definitely be a useful feature, but also a tall order. i think it's really hard to say which games have thge likelihood to *become* exciting. aside from just high level heuristics like "a win probability near 0.50".

English

1

20

dr. doug liebe@dougliebe·2 Mar

@inpredict I.e., if I were following that game on your site, I would not have been compelled to tune in throughout Q4 using your new method. Granted, there was a lot of excitement left in that game!

English

0

122

Mike Beuoy@inpredict·2 Mar

New (and 1st) post up on Substack. Creating a better way of ranking games with the help of Claude.......Shannon. Borrowing a concept from information theory, KL-divergence, I create a new version of the Excitement Index that better aligns with how people evaluate the games.

English

2

62

11.7K

Mike Beuoy@inpredict·2 Mar

A huge thank you to wikihoops.com for providing game level data on user ratings, allowing me to deomnstrate that using KL-divergence in this way wasn't just a fun theoretical exercise, but an objectively better measure of "excitement".

English

7

1.6K

Mike Beuoy@inpredict·2 Mar

Post: inpredictable.substack.com/p/quantifying-…

English

9

2.8K

Mike Beuoy@inpredict·1 Mar

@PepeFel58185408 That's using the NBA's definition of clutch, which counts a 2 point shot when up 4 with 4:30 left the same as a 2 point shot down 1 with 0:10 left. This is based on the actual win probability impact of each player's plays.

English

0

2

48

Oliver Jake@PepeFel58185408·28 Şub

@inpredict You can check everything here: nba.com/stats/players/…

English

0

44

Mike Beuoy@inpredict·28 Şub

NBA season leaders in Clutch Win Probability Added. The market favorite is Ant, but by this metric Desmond Bane deserves to be in the conversation for the Jerry West Clutch Player of the Year Award.

English

5

27

3.3K

Mike Beuoy@inpredict·28 Şub

@PepeFel58185408 Here’s more background. inpredictable.com/2023/01/the-je…

English

0

85

Oliver Jake@PepeFel58185408·28 Şub

@inpredict I don't understand where that metric comes from. Bane has scored almost half as many points as Edwards, and his clutch efficiency is terrible, with a 37% FG and 21% 3P.

English