In baseball, numbers never unveil the complete narrative yet they reveal a significant part of it, or so we believe. Equipped with a keen interest in AI and a plethora of baseball statistics, we developed a chatbot capable of forecasting the World Series result.
On Monday, defying all odds, it predicted that the Los Angeles Dodgers would triumph in Game 3 and eventually clinch the World Series. As observed last night, the chatbot rightly predicted the first part.
Today, it is predicting that not only will the Dodgers clinch the World Series, but they will also win tonight’s game, becoming the first team to sweep 4-0 since 2012 when the Detroit Tigers were defeated by the San Francisco Giants.
This outcome is somewhat different from what the betting markets were forecasting yesterday and today.
The chatbot we devised was essentially a GPT loaded with performance data obtained from Baseball Savant and Statmuse. We then incorporated a brief introduction to sabermetrics—baseball’s mathematical backbone—to aid it in comprehending those numbers.
Creating the model was straightforward (we even developed a guide to assist you in building your bot on virtually anything). We were uncertain about the most crucial statistics, so we essentially fed it with everything we discovered: raw data covering team performance and player statistics through 2024; recent game data, including play-by-play breakdowns to keep the model updated; and a host of unusual things such as exit velocity, pitchers’ arsenals, arm strength, and hits against left- and right-handed pitchers.
In the end, we provided it with a framework for analysis—a “chain of thought” process that juxtaposes historical patterns with current probabilities. We tested the model and tweaked the prompt until we were content with the outcomes. And when we were about to click the “save” button…
It threw an error. It was likely a temporary server glitch that prevented us from saving our changes.
Regardless, we managed to capture some intriguing responses before it excused itself. The model predicted a 60% victory for the Dodgers in last night’s game, with a slight chance for the Yankees to win by 1-2 runs.
For the final outcome, our bot zeroed in on a stark statistic: Teams that win the first two games in a best-of-seven series end up winning 80% of the time. This appears to be fairly accurate, and you probably already knew that from listening to the commentators during last night’s game.
Our AI assimilated years of World Series data along with current-season stats. For the Yankees-Dodgers showdown, the Dodgers exhibited a slight edge in overall pitching stats, while the Yankees’ relief pitchers dominated throughout the playoffs.
The results diverged from popular prediction markets. Polymarket bettors, for instance, gave the Yankees a 56% chance of winning Game 3, deeming three straight losses as unlikely. Meanwhile, our chatbot concentrated on broader patterns.
The game ultimately concluded with a 4-2 victory for the Dodgers. So our model was correct, as Los Angeles won with two more runs than New York. At least this time, DI (degen intelligence) didn’t outperform AI.
What about tonight—and the Series?
Things looked bleak for the Yankees yesterday, but the situation is considerably worse today. We asked our chatbot for their probability to recuperate and win the next four games. If that happens, it would be a historic moment in American baseball.
The chatbot gave the Dodgers a 55% chance of winning tonight and sweeping the series.
“In MLB postseason history, only one team has successfully come back from a 0-3 deficit in a best-of-seven series: the Boston Red Sox in the 2004 American League Championship Series against the New York Yankees,” our chatbot said. It provided us with a statistical calculation for a four-win streak under similar circumstances, giving the Yankees a theoretical probability of around 6% to make history, come from behind, and win the Series.
Things become a lot more pessimistic when we consider elements that influence the teams’ performances.
“If we assume a lower win probability per game (due to the Dodgers’ strength this season), the probability will decrease accordingly. For example, if we believe the Yankees have a 40% chance per game, the probability of four straight wins would be 2.56%,” our chatbot said. According to our chatbot, the Dodgers are in a much better position—so instead of a 50-50 scenario, it believes a 60-40 chance for the Dodgers to win is more realistic.
The Polymarket folks are stubborn. For tonight’s game, the odds are 58% in favor of the Yankees at the time of writing this article. This is not what our chatbot thinks. However, it reminded us that some factors could favor the Yankees—like the game being played in New York and the sense of urgency that could affect the players’ physique. This was enough for our bot to slightly increase the Yankees’ odds from 40% to 45%.
That’s not enough to term it an even 50-50, and it still doesn’t align with our favorite prediction market.
A 90% chance of losing the World Series is hard to overlook, but in baseball, one loss can rewrite the narrative and undo even the most rigorous predictions. Perhaps our chatbot needs more heart, but the Polymarket degens could use a dash of data too.
Edited by Andrew Hayward.
