iiinova — where your digital self keeps evolving

Mike Beuoy @inpredict.bsky.social · 30/09/2026

On Feb 3, 2024, the Warriors lost to the Hawks, despite +101% in win probability added from Steph Curry. It's the 12th best WPA performance from an NBA player in a losing effort (in the play by play era).

092

Mike Beuoy @inpredict.bsky.social · 30/07/2026

Two 🔥 games in the WNBA last night. stats.inpredictable.com/wnba/preCap....

021

Mike Beuoy @inpredict.bsky.social · 16/06/2026

Here's an alternative look at the biggest shots in modern NBA history, this time sorted by change in title probability (rather than KL divergence). Some my think this is a better list - we get Jordan twice here. But Ray Allen's shot gets downgraded.

0133

Mike Beuoy @inpredict.bsky.social · 15/06/2026

The analysis above utilizes a simple model to assess the importance of each playoff game to a team's chances of winning the championship.

101

Mike Beuoy @inpredict.bsky.social · 15/06/2026

Ranking the biggest shots in modern NBA history (link in thread). As mentioned yesterday, the #1 shot is a put-back, but it wasn't OG's, which ranked 6th. Derrick White #1 Ray Allen #2 Kyrie #3 Big Shot Bob #4 and #5

2197

Mike Beuoy @inpredict.bsky.social · 11/06/2026

Clutch Win Probability Added leaderboard for the 1st four games of the 2026 NBA Finals. This counts all potential clutch plays - field goals, free throws, rebounds, turnovers, steals, blocks, and assists.

131

Mike Beuoy @inpredict.bsky.social · 02/06/2026

Visiting Jacob Bernoulli’s tomb in Basel. He loved the logarithmic spiral because of its self similarity property. No matter how much you zoom out it looks the same. The epitaph reads “EADEM MUTATA RESURGO”, meaning “Although changed, I will rise again the same”.

251

Mike Beuoy @inpredict.bsky.social · 02/06/2026

But the post also contains a chart showing what situations this new measure of excitement considers more "exciting". For example, going from 5% win probability to 15% win probability (see bottom left of chart) is a much bigger deal under the new measure.

000

Mike Beuoy @inpredict.bsky.social · 20/05/2026

The 4th and 5th most improbable comebacks of the NBA playoffs took place almost exactly a year apart and at the same venue.

130

Mike Beuoy @inpredict.bsky.social · 19/05/2026

Using KL-divergence excitement (coming soon to the site), last night was #2, with this Raptors-Cavs game as #1:

020

Mike Beuoy @inpredict.bsky.social · 19/05/2026

By the current definition of excitement index, Spurs-Thunder last night was the most exciting game of the playoffs.

160

Mike Beuoy @inpredict.bsky.social · 29/04/2026

it's fun when your ggplots go off the rails in aesthetically interesting ways

090

Mike Beuoy @inpredict.bsky.social · 29/04/2026

The Sixers were 360 basis points from elimination early in the 3rd quarter.

050

Mike Beuoy @inpredict.bsky.social · 15/04/2026

Charlotte-Miami win probability chart:

030

Mike Beuoy @inpredict.bsky.social · 14/04/2026

To make the case for Jokic, you would need to believe that assists and rebounds count as clutch play. Shai leads by a good margin when it comes to clutch shooting. stats.inpredictable.com/nba/jerryWes...

200

Mike Beuoy @inpredict.bsky.social · 14/04/2026

Here are the final player clutch rankings for the 2025-26 NBA season. Shai and Jokic finished neck and neck.

281

Mike Beuoy @inpredict.bsky.social · 10/04/2026

Inspired by Zak Kram's article on pace in ESPN, here is an updated chart that shows team correlation by season between offensive pace versus efficiency. An downward sloping line indicates a positive correlation between fast pace and offensive efficiency.

150

Mike Beuoy @inpredict.bsky.social · 08/04/2026

You can set your watch to the reliability of the rubber band effect in the NBA. Red = Team overperforms Blue = Team underperforms Teams play better from behind. It's pretty much the opposite of momentum.

183

Mike Beuoy @inpredict.bsky.social · 28/03/2026

You can check a box on the inpredictable win probability graphs to display real time, rather than game time, on the horizontal axis. That can be helpful for a game like LAC-IND where a game's worth of insanity was packed into the final second.

130

Mike Beuoy @inpredict.bsky.social · 27/03/2026

Update after last night's games:

020

Mike Beuoy @inpredict.bsky.social · 26/03/2026

New Substack post: March Madness Forecasting Thunderdome I apply my Kelly-based forecast evaluation method to models predicting the NCAA Men's Basketball national champion. Link: open.substack.com/pub/inpredic...

010

Mike Beuoy @inpredict.bsky.social · 26/03/2026

Just an absurd win probability chart

0143

Mike Beuoy @inpredict.bsky.social · 24/03/2026

A lot of conseqeuntial games in the NHL today. Here's how playoff chances for each team are impacted based on the outcomes today: win: 2 points overtime loss: 1 point regulation loss: 0 points

010

Mike Beuoy @inpredict.bsky.social · 21/03/2026

Meta-bracketology for the NCAA Women's Tournament. Of the 3,511 brackets analyzed, there were 2,817 unique brackets (80%). The most common number of seed upsets picked was 0. Next most common was 8.

062

Mike Beuoy @inpredict.bsky.social · 20/03/2026

Here is the distribution of pre-tournament perfect bracket probabilities, charted on a log scale. Pre-tournament, there was about a 3 million to 1 chance that a perfect bracket would emerge from one of these 18,724 distinct submissions.

020

Mike Beuoy @inpredict.bsky.social · 20/03/2026

Some meta-bracketology, using 20,000+ brackets submitted to the NCAA's bracket challenge. Of the 20,751 brackets analyzed, there were 18,724 unique brackets (90%). The most common number of upsets picked: 12

230

Mike Beuoy @inpredict.bsky.social · 13/03/2026

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

0203

Mike Beuoy @inpredict.bsky.social · 07/03/2026

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 ✅

050

Mike Beuoy @inpredict.bsky.social · 02/03/2026

Using a key concept from information theory, KL-divergence, I create a new version of the Excitement Index that better aligns with how actual humans evaluate the games.

110

Mike Beuoy @inpredict.bsky.social · 28/02/2026

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.

100

Mike Beuoy @inpredict.bsky.social · 21/02/2026

Favorite random science-related point of interest in Pasadena: There’s a mural dedicated to Richard Feynman on the back wall of a Ross Dress for Less.

040

Mike Beuoy @inpredict.bsky.social · 05/02/2026

The Knicks and the Celtics used to be outliers too, but have now merged with the pack.

010

Mike Beuoy @inpredict.bsky.social · 05/02/2026

Venue effects. At this point, it's the Rockets, and then there's everyone else. Rebounds are grabbed a second earlier in Houston, compared to the rest of the league.

110

Mike Beuoy @inpredict.bsky.social · 05/02/2026

After dark Wednesday minutiae: The average time elapsed between a missed shot and a rebound* in the NBA is holding steady at around 2.5 seconds. *as recorded in the play by play logs

110

Mike Beuoy @inpredict.bsky.social · 16/01/2026

Saw Marty Supreme last weekend, and was inspired to do a comp of serving advantage across multiple sports. Tennis 🎾: it's an advantage to serve Volleyball 🏐: it's a disadvantage to serve Table Tennis 🏓: nearly 50/50, based on what limited data I was able to find

172

Mike Beuoy @inpredict.bsky.social · 31/12/2025

It’s almost the 65th anniversary of The Great Rose Bowl Hoax, in which a group of CalTech students switched the Washington cheer team’s card system so that their cards spelled “CALTECH” instead of “HUSKIES”.

040

Mike Beuoy @inpredict.bsky.social · 26/12/2025

That was, quantifiably, one of the best slate of Christmas games since the NBA standardized the schedule to 5 games in 2008:

051

Mike Beuoy @inpredict.bsky.social · 17/12/2025

And finally, here are the stats for Wisconsin, and their setter Charlie Fuerbringer. Note this does not include the early season games when she was out due to injury.

000

Mike Beuoy @inpredict.bsky.social · 17/12/2025

Kentucky's setter, Kassie O'Brien, was recently named the AVCA Freshman of the Year. Here is how the Wildcats have fared with O'Brien at each rotation:

100

Mike Beuoy @inpredict.bsky.social · 17/12/2025

Here's the same chart for their opponent tomorrow, Pitt, and their setter Brooke Mosher. On serve, they've had the most success with Mosher in the 5 position, and struggled the most in the 2 position.

100

Mike Beuoy @inpredict.bsky.social · 17/12/2025

Playing around with rotation data prior to tomorrow's semi-final games, and thinking about new stats to add to @evollve next season. Here is Texas A&M's success rate by rotation, where rotation is specified by the rotation position of their setter Maddie Waak:

110

Mike Beuoy @inpredict.bsky.social · 14/12/2025

Latest data on 15+ point comebacks in the NBA (this includes last night's Spurs comeback over the Thunder). When a team has fallen behind by 15+ points this season, they've come back to win 13% of the time.

071

Mike Beuoy @inpredict.bsky.social · 23/11/2025

The Colts decision making when playing the Chiefs on the road:

040

Mike Beuoy @inpredict.bsky.social · 18/11/2025

So far, offensive efficiency roughly in line with the post-2022 norm.

051

Mike Beuoy @inpredict.bsky.social · 18/11/2025

Pace and efficiency trends in the NBA through the 1st 207 games. Pace slows down as the season progresses, but even accounting for that we're on track for the fastest pace in the play by play era.

3207

Mike Beuoy @inpredict.bsky.social · 15/11/2025

In addition to halting an opponent run, timeouts can be used to "ice the server" in high leverage situations. As with the results on opponent runs, calling a TO in a high pressure situation does not appear to impact the server's performance on the following point. 3/

100