MLB Data Warehouse

MLB Data Warehouse

Beware the Late Season Steals Surge

Stolen bases increase in September. I examine why this is and find some potential steals decliners in 2026.

Jon A's avatar
Jon A
Nov 12, 2025
∙ Paid

I’ve noticed something over the last few years. It seems to me that there are situations where players will just go crazy stealing bases in September while they chase a milestone.

The first example of this was Shohei Ohtani in 2024. All eyes were on his chase for the first-ever 50-50 season, and no doubt that was in his head as well. He went for around 17% attempt rate from March-June, and then once he got a whiff of history, he started going nuts. He stole 31 bases in the final two months of the season with an attempt rate well north of 50%.

Unsurprisingly, he cooled it down in 2025. Ohtani stole just 20 bags last year on a 13.4% attempt rate.

Juan Soto is another example. He stole 37 bags last year, with 22 of them coming in the final month of the season. His 40-40 potential showed itself late in the year, and he went for it.

It seems to me that, in certain situations, we should throw out late-season steals. The first thing I did to test this was look at league-wide stolen base attempt rates by month dating back to 2021. If my theory is true, then the portion of the players in the league trying to reach a certain SB milestone late in the year should result in a higher-than-average league attempt rate in September.

And that’s exactly what we find:

It is similarly high in April. I’m not as sure what that’s about, but my guess would be that players are healthier and maybe pitchers are still kind of getting their bearings and not paying as close (or as effective) attention to base runners. That’s just speculation, though.

Stats matter quite a bit in baseball for contract purposes. They present stats in arbitration hearings. It makes sense that players think that adding ten steals to their tally could earn them some extra money on their next contract.

We saw this with Aaron Judge when he stole 16 bases in his contract year in 2022, adding five steals to his previous career high and ten from the prior year.

Stolen bases is the biggest will stat in professional sports, as far as I can tell. By that, I mean that the biggest predictor of whether a player will steal bases is if the player wants to steal bases. This isn’t true with home runs. Everybody would love to hit 70 bombs. I promise you, every player would hit 70 homers if they could. That’s not true with steals. Some guys don’t want to steal. Some teams don’t even care much about their players stealing bases.

So external factors (such as chasing milestones and trying to earn more money in the next contract) have an influence on steals.


To further prove this, I went back to the data. Here’s what I got for each year between 2021 and 2024:

  • Non-September SB Attempt% by Year

  • September SB Attempt% by Year

  • SB Attempt% the Following Year

I isolated the players who went +10 points in SB Attempt% in September above what they had done previously, and then I looked at what they did the next year.

In almost every case, next season’s value looks a lot like the non-September number. These late-season September gains rarely stick around for the next year.

CAVEAT! You could do this with any single month. If you picked out each player’s max SB Attempt% month and compared it with the next year, you’d surely find the same thing.

I did a correlation test to see if throwing out September improves the predicted SB Attempt% for the next year. So I compared these two:

→ Year N vs. Year N+1
→ Year N (ignoring September) vs. Year N+1

The correlation coefficient was exactly the same at 0.76. So it’s no better to throw out September, but it’s no worse either. You can take it or leave it. And you would usually expect throwing out one-sixth of the season to make the prediction less accurate.


The Takeaways

You can see where we’re headed here! We want to find the late-season steals surgers from 2025. Who are the guys who stole a bunch of bags late on elevated attempt rates?

But I’m going to leave that as a teaser. The paywall hits now, become a paid sub today to get the rest of this post and the multitude of other things going on here at MLB DW!

This post is for paid subscribers

Already a paid subscriber? Sign in
© 2025 Jon A
Privacy ∙ Terms ∙ Collection notice
Start your SubstackGet the app
Substack is the home for great culture