World Series: A Statistical Prediction

Photo by Domingo Saez

The 2024 World Series will officially start tonight, and fans have been treated to a great match-up: the New York Yankees vs. the LA Dodgers. Not only does this pair up the two teams with the best records in their respective leagues, but it is also one of the most historic rivalries in the sport's history. Although the Yankees and Dodgers haven’t faced off in a World Series since 1981, this will still mark their 12th meeting, which is the most of any two teams in the MLB. When looking back at those previous eleven encounters, the Yankees have the edge, having won 8 times against the Dodgers. However, the Dodgers won their most recent confrontation and aim to do so once more. 

They will rely on their star player, Shohei Ohtani, who is making his World Series debut. The Japanese native has had a stellar year, becoming the first player ever to reach 50 home runs and 50 stolen bases in a single season. On the other side, New York hopes Aaron Judge will be able to deny LA the glory. Judge scored the most home runs in the regular season with 58, although his performance in the playoffs has left more to be desired. Thankfully, he won’t be alone in the task of stopping the Dodgers, having help from the likes of Juan Soto and Giancarlo Stanton. Stanton was impeccable in the ALCS against the Guardians and looks to carry that form over into the World Series.

With all of that said, which team do fans believe is favored to win it all? Most seem to believe that the Dodgers have a better chance, with the bookies listing LA as a slight favorite at around -125 betting odds. Given how close of a match-up most anticipate this to be, I decided to go ahead and do some research of my own. For starters, let’s take a look at all the previous World Series, and see if there are any sort of clues in helping to predict who has a higher chance of winning.

I compiled the list of all 119 World Series winners and looked at their regular season winning percentage compared to the winning percentage of their opponents (excluding ties). I then obtained the difference between both values for every season to determine the average difference in winning percentage that World Series winners have.

The result is listed below:

Graphic by Andy Rodriguez

As you can see, on average, the World Series winners have finished the regular season with a win percentage that is 0.007 better than the losing team. As expected, the teams that have a higher winning percentage than their opponents tend to win the World Series more often. If this were to be true this year, then the Dodgers would be the clear favorites, as they finished the season with a winning percentage that was 0.025 higher than the Yankees.

Nonetheless, this is very clearly not the best way to determine a favorite, as you can also see that the team with a lower winning percentage still managed to win the World Series 54 times. There are also many other variables to consider, such as runs scored, their form heading into the World Series, and the potential difference in difficulty between the American League and the National League.

With all of this in mind, I continued my analysis by looking at the years when playoff games leading up to the World Series were added. These began in 1969, so only 54 seasons will be included in the next step of the analysis. While the number of seasons was now less than half of the original 119, I was able to add the winning percentage of the teams in the playoffs, as well as their run differential (number of runs allowed subtracted from the number of runs scored).

The addition of these new variables allowed for the creation of a decision tree, which generated these probabilities to win:

Graphic by Andy Rodriguez

The result is quite intriguing, as we are shown that teams that have a playoff win percentage that is over 0.19 higher than their opponents are favored to win, with around a 70% success rate. However, if the team’s playoff win percentage is superior, but not by at least 0.19, they tend to lose around 64% of the time. In this year’s case, the Yankees find themselves in this curious scenario, with their playoff winning percentage being 0.141 greater than the Dodgers.

Therefore, although the Yankees boast of a superior winning percentage in this year’s playoffs, they still seem to be slightly more likely to lose to the Dodgers. Clearly, there are still some missing factors to consider, since this pattern does not make complete sense at first glance. How could teams that had a better win percentage during playoffs still be seen as more likely to lose?

To gain a better understanding of this seemingly random pattern, I decided to look at the MLB seasons that have interleague play during the regular season. This way, the records of the interleague games could potentially uncover which league was considered the stronger one of the two during that specific season. If we can understand the disparity between the AL and the NL during each season, we could explain why some teams struggle in the playoffs, yet still manage to win the World Series once they face an opponent from the other league.

Adding this new element meant that we could only look back until the 1997 season when interleague play first began. Crunching the numbers into the program once more led to the creation of an improved decision tree, which as expected, explained some of the possible scenarios where having a better win percentage during playoffs could still lead to losing the World Series.

Note: chances of winning increased from 70% to 100% compared to the earlier graphic due to smaller sample size. This was a result of only looking at seasons since 1997 when interleague play began.

So what does this mean in the case of the 2024 Yankees and Dodgers? Well, as we mentioned earlier, the Yankees did have a better win percentage during the playoffs, ending with a win rate that was superior by 0.141. Additionally, the Yankees play in the American League, which this season only had a win percentage of 0.465 when playing against teams from the National League. Evidently, the National League had the better teams this year and was able to get the better of the American League teams more often. As a result, the Yankees are placed in the third scenario derived from the decision tree. While they did have a better win percentage in the playoffs, they played in a league that was clearly not at the level of the National League this year. To make matters worse, the Yankees had a run differential of 11 in the playoffs, which pales in comparison to the run differential of 23 that the Dodgers had. Combining all of these factors, all roads seem to lead to the same conclusion: The Dodgers are more likely to win the 2024 World Series against the Yankees.

Be that as it may, Yankee fans have not lost hope and remain very excited about their team being in the World Series. Speaking to Yankee fan Luke Gaccione, he remains hopeful despite the advantage to the Dodgers: “All things considered, this is relatively a very close matchup that almost nobody can predict… at that point all bets are off, it’s a really hard series to predict because they both truly are really good baseball teams.” To Luke and his family, the opportunity to watch their favorite team in the World Series is hard to describe: “I feel like this has just been a moment that we’ve been waiting for my whole life. It just gives me such an excited feeling of anticipation and a little extra reason to be happy throughout my day.”

Many fans believe this series will be a very close one, yet New York City fans will hope that Luke’s prediction comes to fruition: “Yankees in seven.”