Last week, we released a baseball simulator in the FanGraphs Lab. This week, we’re adding home field advantage to the simulator. You can toggle HFA on and off using a new menu option:
The chosen home field advantage will then be applied to whatever simulation you run. But how do we calculate home field advantage in this simulated environment? Let’s go over it.
You’re probably familiar with home field advantage being expressed in terms of winning percentage. From 2000-2010, the home team’s winning percentage hovered around 54%. In the next decade, it declined to around 53%. In recent years, it has fallen to the 52-53% range. Since our simulation works at a plate appearance-level, however, we couldn’t look at game outcomes to measure home field advantage. Instead, we used PA-level data to infer how much playing at home affects the rate of each outcome in our simulator.
We took data from 2022-2025, the universal DH era, and used it to fit three different models of home field advantage. First, we simply compared the rate of each outcome achieved at home and on the road, without considering batter identity. Next, we fitted a logistic regression that considered batter identity plus a categorical home/away variable. Finally, we fitted another logistic regression that used batter identity, pitcher identity, and home/away. The three models produced similar values, within a margin of error, which gives us confidence that the methodology is sound.
In our sample, home field advantage appears to act chiefly on strikeouts, walks, and home runs. Home batters strike out less frequently, (unintentionally) walk more frequently, and hit home runs more frequently. They also hit doubles more frequently, though that effect size was meaningfully smaller. We found no statistically significant variation in hit-by-pitch, single, triple, or “in play, out” rates, though it’s worth noting that for triples in particular, the sample size was too small for much precision.
You Aren’t a FanGraphs Member
It looks like you aren’t yet a FanGraphs Member (or aren’t logged in). We aren’t mad, just disappointed.
We get it. You want to read this article. But before we let you get back to it, we’d like to point out a few of the good reasons why you should become a Member.
1. Ad Free viewing! We won’t bug you with this ad, or any other.
2. Unlimited articles! Non-Members only get to read 10 free articles a month. Members never get cut off.
3. Dark mode and Classic mode!
4. Custom player page dashboards! Choose the player cards you want, in the order you want them.
5. One-click data exports! Export our projections and leaderboards for your personal projects.
6. Remove the photos on the home page! (Honestly, this doesn’t sound so great to us, but some people wanted it, and we like to give our Members what they want.)
7. Even more Steamer projections! We have handedness, percentile, and context neutral projections available for Members only.
8. Get FanGraphs Walk-Off, a customized year end review! Find out exactly how you used FanGraphs this year, and how that compares to other Members. Don’t be a victim of FOMO.
9. A weekly mailbag column, exclusively for Members.
10. Help support FanGraphs and our entire staff! Our Members provide us with critical resources to improve the site and deliver new features!
We hope you’ll consider a Membership today, for yourself or as a gift! And we realize this has been an awfully long sales pitch, so we’ve also removed all the other ads in this article. We didn’t want to overdo it.
We use the home field advantage factors that we calculated in the above step to modify the baseball simulation engine. We make these modifications in odds ratio space by taking the initially calculated probabilities of each outcome, decomposing them into odds ratios, modifying them by the home field factor, and then re-casting them back into probabilities. This method accounts for varying effect sizes based on expected matchup outcomes. Luis Arraez doesn’t get the same raw boost in home runs as Aaron Judge, in other words.
We tested this method of accounting for home field advantage by having identical teams face off against each other, both with and without the home field modifiers. Home team winning percentages averaged between 52% and 52.5%, exactly what we expected given the relationship between the observed change in individual outcome frequencies and actual winning percentage. In other words, we feel that our model is well-calibrated on real-world data in recent years.
By default, we split our calculated home field adjustments in half and then assign a bonus to home batter outcomes and a penalty to away batter outcomes. If you’d like to run your own version of home field, though, you can change the calibration scalar away from that default 0.5 weighting:
Moving that slider will linearly increase or decrease the home and away modifiers added to the simulation. You can also use your own custom weights for each event type, though frankly you shouldn’t need to; we just thought it was pretty cool that you can, so we left the feature in. As a word of warning, moving these sliders by a large amount can produce nonsensical results; if your multipliers mean that the home team hits 50% more singles than the away team on average, the simulated game won’t look much like baseball.
The rest of the simulation is unchanged; you can see a description of its current features here. As before, this is still very much a beta product. If you see any issues or have any feature requests, just let us know, either in the comments here or by using the Lab’s feedback button.