The 2020 ZiPS Playoff Probabilities Are Live!on September 29, 2020 at 1:00 pm

The general methodology remains the same this year, though the mathematics-based shenanigans are slightly more robust. To aid in modeling these probabilities, the full ZiPS model is used, which includes updated platoon projections to (hopefully!) enhance accuracy. In situations where the starting pitcher has yet to be named, you will see the starter listed as “TBD,” which represents a weighting of different pitchers who could theoretically start the game in question. In addition, I’ve modified the model for bullpen quality somewhat to better simulate the mix of pitchers used in high- and low-leverage situations.

One final variable to consider is the effect of the playoff bubble. Home field advantage was larger in 2020 than is typical (teams are usually between .510 and .530 at home, but were .557 this year), but it’s hard to tell how much of that was due to the comfort of playing at home in what was arguably the most stressful season of players’ careers. So, how do we estimate home field advantage in a situation where the comfort and familiarity are no longer present, with no team having a true home park? It would be unsatisfying to simply say that home field advantage is completely gone since there does appear to be some small tactical advantage to hitting second. To estimate this, I looked at historical player home/road splits compared to actual record in a rough attempt to try to isolate decision-making rather than ability as an advantage. It’s unfortunate that teams have never alternated between hitting first and hitting second for a season so that we have data to go through! In the end, my estimate is that home field advantage in this situation is worth about 12 points of winning percentage: