A Shohei Ohtani 50/50 Odds Update

Last week, I modeled Shohei Ohtani’s chase for 50 home runs and 50 steals to predict when he might reach that historic dual milestone. That prediction isn’t static, though. Every time Ohtani plays a game, the likelihood of his getting to 50/50 changes. Good news, though: Updating the model is as easy as hitting a few keys and listening to my computer hum for a bit.

This isn’t going to be a long article. It is, however, an updated set of probabilities, which is the whole point of this exercise. Ohtani hit two homers in his weekend series against the Guardians, which leaves him only four home runs and four steals short of a half century in each statistic. His odds of reaching 50/50 are up to 61.3% in my simulations – they were 55.6% before this series.

As a quick refresher, I’m simulating the likelihood of his hitting 50 of each statistic with a Monte Carlo simulation that takes his talent, his opponents, and the stadiums where he plays into account. I also introduce a random fluctuation in his home run talent: Sometimes he’s hot, sometimes he’s not, and sometimes he’s in between. I then simulate the season a million times and note whether he hits 50/50, and if so, in which game he does it.

The two homers in the weekend series have slightly moved up the most likely date for when he’ll reach the 50/50 threshold. Before his series against Cleveland, my simulation suggested that the single game most likely to see Ohtani either steal the base or hit the homer that pushes him over the line was the Dodgers’ September 27 game in Colorado. That’s still the case – but it’s now dead even with the previous game, September 26 in Los Angeles against the Padres. Furthermore, the Padres series has overtaken the final Rockies series as the three-game set in which he is most likely to set the mark.

Here’s the complete set of game-by-game probabilities:

I think these projections do a good job of handling a tricky problem. But I do want to make one point about their limitations: Steals aren’t quite as easy to model as home runs. Pretty much every time that Ohtani comes to the plate, his ideal outcome is a home run. He swings to hit home runs, and pitchers do their best to prevent them. The past does a great job of predicting the future here, because intent doesn’t change from one plate appearance to the next. Steals don’t work quite like that. Sure, Ohtani’s speed is a consistent and important input; the same is true for his baserunning instincts, the opposing pitcher’s ability to hold him on, the catcher’s throwing arm, and so on. But how much he wants to steal is also crucially important. He’s attempting to steal more frequently in the second half of the season than he was in the first, and his desire to run presumably will only accelerate if he’s sitting on, say, 50 home runs and 49 steals. I’m modeling a steady-state true-talent world, but I think it would be reasonable to tilt the distribution slightly earlier if Ohtani hits the homer plateau before the stolen base one, which looks more likely today than it did last week.

In any case, some takeaways: The last six games of the season are the most likely time to see history. The series against the Padres is now the best bet despite San Diego’s excellent pitching staff. The last series of the season, at elevation against a bad pitching staff, is the next most likely. The likelihood of Ohtani’s getting to 50 during both series is higher now than it was on Thursday, and I might even be underestimating it given that he might decide to attempt more steals as he nears the border of history.