clock menu more-arrow no yes

Filed under:

Is It That Important to Hold Runners?

New, comments

Over the past week, there's been a lot of talk over at Fangraphs about how horrendous Chris Young has been at holding runners.

Since 2006, his steals / caught stealing numbers are 41/4, 44/0, 15/2, and 14/0 (so far this season).  His ERA+ in that time frame has been 117, 129, 97 and 68 (very small sample size).

Despite his lack of ability to keep runners from stealing, he's been pretty successful.  In the FanShot discussing the latest article from Fangraphs, I asked whether holding runners might limit Young's effectiveness in other ways - perhaps distracting his attention from the batter - similar to how you can't just tell a power hitter to strikeout less and assume the rest of his game will be unaffected.

Sky responded by wondering how important controlling the running game actually was.  I'm quoting myself here (and keep in mind this was back of the envelope type stuff, so not very precise):

Young went 0/44 in 2007.

Using the .240 average CS% for catchers (supplied by Brian Cartwright in a comment to this post) and assuming that runners would still have the same number of attempts, you’d expect him to allow 34 steals.

So he was roughly 10 steals worse than average. The run value for a stolen base is .175, while the win value is .018 (steals tend to happen in slightly higher than average leverage situations).

Either way, ten steals is roughly .18 wins over the course of a season – important but not hugely so.

But we also have to account for the additional 10 outs that would have occurred. A caught stealing is worth .467 runs to the defense, or .043 wins. So that’s another .45 wins or so. We’re now up over .6 wins.

It’s getting bigger, but remember, this is probably the worst season ever.

Now if you were to assume that fewer runners would attempt to steal if Young and his catchers were better at throwing people out, that could increase the effect.

According to Brian Cartwright, Young’s expected steals results in 2007 were 18 SB / 3 CS. That would have been .32 wins against, and .14 wins for – a total of .18 wins against.

He was actually 44 SB / 0, or .79 wins against, and 0 for.

The difference between expected and actual? Roughly the same .6 we got before.

That seems to be a pretty fair upper level estimate of how badly a pitcher can be hurt by not holding runners.


As Sky pointed out, that's a difference of .3 in ERA over 180 innings.

I wasn't sure how big an effect that was compared to some other things, so I decided to take a look.

Using FIP as our starting point, a pitcher can lose .3 runs of FIP in 180 the following ways:

  • 4 more home runs (or .2 HR/9)
  • 18 more walks (or .9 BB/9)
  • 27 fewer strikeouts (or -1.35 K/9)
So the worst season ever holding runners was the equivalent of giving up 4 more home runs.  It's certainly meaningful, but perhaps the doom and gloom scenarios that people are raising for Young are a little overblown.  Of course if teams follow Dave Cameron's advice and run on him continuously, that would be a different story. 

There is a breakeven point where Young would be forced to change his approach and focus more on the runners, but that exact point depends on how that change affects his delivery and results - which makes it pretty much unknowable.