Matching Up, Streaming for Stolen Bases

In a comment on the post of a recent ep of “The Sleeper and the Bust,” quinceleather asked if there’s a way to play matchups for stolen bases for players like Rajai Davis and Ben Revere. Presumably, this would be for the purpose of increasing the efficiency of a roster spot by maximizing the rates of thievery of that brand of one-category contributor in games, series or weeks in which he has advantageous matchups and then using a different type of player (power hitter, an all-around player, a fringe regular) in place of them at other times.

I don’t think I’d take that approach for a full season. But for a particular period, especially at this point, when there are only about four and a half weeks remaining in the regular season, it could certainly, conceivably be helpful. You want to make up ground in steals, and you can afford to use this kind of method to do it. Someone else is more qualified than me to study whether it would prove to be advantageous for an entire campaign.

Eno and I talked about the subject for a few minutes in the latest podcast. I stated the obvious, and Eno elaborated and directed listeners: You’re looking for stolen base and caught stealing data versus both catchers and pitchers.

Schedules would then become more of a factor in your lineup decisions. Eno declared the importance of putting players such as Davis in beneficial platoon situations in order to take advantage of something like matchups in the first place. Davis has historically been a poorer hitter against right-handed pitchers, for those who’ve missed that.

That could backfire, sometimes. If your one-category rabbit is facing a battery that invites an excellent chance for success and the pitcher isn’t very good anyway, then there’s some temptation to plug him in. That seems like a kind of boom-or-bust play. Setting your lineup based on this kind of info, given the number of variables involved, probably isn’t very scientific.

I’d also wondered aloud in the show if it’d be possible (without the use of SQL queries, so that I wouldn’t have to bug someone else to run them frequently and instead could use my extremely limited knowledge of database management) to compile a simple set of tables that would be easy to refresh. That’d be convenient for a variety of obvious reasons. So, here we are.

I used Baseball Reference because it has some columns I couldn’t find on FanGraphs. I added the stolen base attempts, stolen base attempt percentage and stolen base percentage. They’re the same in the team tables of both tabs, obviously, but it’s handy to have a team table available while viewing individual players on either side. Each tab allows you to compare for individual matchups. You can check the schedule and then filter by team on both tabs. Sort or filter anything as you please.

I set individual catcher and pitcher tables to display only those who have had a minimum of 10 stolen bases attempted against them. It’s far from a perfect sample size, but anything greater than that, and a good number of notable catchers and pitchers would’ve been excluded. You can adjust by clicking on the filter for SBA and selecting Number Filters > Custom Filter…

I added SBA% because success isn’t always determined by caught stealing percentage. The threat of a catcher’s prowess at throwing out runners and/or a pitcher’s ability to keep runners close can reduce the pool of willing thieves to those who are best or most reckless at it.

Along the lines of the impact of the schedule on your lineup, keep in mind that getting on base, period, is part of the equation. A pilfered base appears to be easier to net versus the Oakland Athletics than most other major league clubs. (. Reaching against hurlers like Jon Lester, Jeff Samardzija and Sonny Gray is a different task altogether. The pitcher still largely referees opportunities to steal against both him and his battery mate. (The just acquired Geovany Soto, who’s been about league-average in his career, might help a tad, besides.)

I (and, I think, some others around here) have received a number of suggestions to incorporate or questions about schedule analysis, particularly for waiver wire columns. I think that it’s a good idea, provided that there’s some context. Take situations such as the one in which you’re aiming to game the player pool in order to boost your SB count relatively quickly.

Eno also mentioned in the pod that he typically does a piece on streaming for steals at about this time every season. He said that he’d aim to do one soon. I asked him after the show if a workbook like I’d described would be handy as a starter or prompt for a blog such as that. I ended up with this thing, so he encouraged me to run with it. I’ll look to highlight some players who look good for a particular stretch next time around.

This is clearly not groundbreaking research. Somebody else may have already come up with something like this. But it could be a nice tool. I invite folks to make suggestions or improvements to the process.