The Stolen Base Matters More Now

Much of the deliberation over whether or not Russell Martin deserves the two-year, $17.5 million contract he pulled in from the Pittsburgh Pirates yesterday will hinge on how his defense is viewed. Martin has been measured as a tremendous pitch framer, and for many that will be enough to put a notch in the team-friendly side for this contract.

But what about his arm Martin threw out just 24 percent of basestealers last season, down from 30 percent last year. The league rate also fell, from 28 percent to 25 percent — 2012 was the best year for basestealers since 2007, when they also accomplished a 75 percent success rate. Still, it was a disappointing year for the 29-year-old backstop, his first below average since 2008.

In today’s pitcher-friendly environment, the equation on the basepaths has shifted towards more and more stolen bases. Observe:

On a rate basis, there are now far more stolen base attempts (SBA) per stolen base opportunity (SBO, from Baseball-Reference: plate appearances through which a runner was on first or second with the next base open). The relationship is clear: the league took a year to catch up, but as run production goes down, stolen bases go up (For the record, the coefficient of determination between OPS and SBA/SBO is 0.78.).

It might seem counter-intuitive — the more run scoring drops, the more we hear about the sanctity of the out. Each of the 27 is hugely valuable, yes, but as run scoring drops each plate appearance is also more likely to produce an out.

The league on-base percentage has fallen from .336 in 2007 to .319 in 2012, and therefore the marginal out — the out risked by the stolen base — is less valuable. That’s why the “runCS” value in the Guts section — the cost in runs of a caught stealing — has lessened from minus-.433 in 2007 to minus-.398 in 2012.

Put another way, the stolen base becomes a less risky proposition because there is less to lose. The hitter at the plate is now less likely to get on base or hit a run-scoring extra base hit, and the chances of two hitters singling in an inning to knock a runner home drops in a compound fashion. Conversely, making the hitter’s job easier becomes more valuable now that the “wait for a three-run homer” strategy isn’t as viable.

The break-even rate on steals has fallen from 68 percent to 66 percent, down from 70 percent at the height of the steroids era in 2000. A player that stole 75 bases and was caught 25 times would have gained 4.2 runs of value in 2007. In 2012, that number rises to 5.1, and the player likely would have been running 10-20 percent more often; a routinely successful basestealer’s value in the running game therefore increases by some 30-40 percent.

The Pirates allowed 154 of 173 basestealers (89 percent) to successfully steal last season. The mark falls partly on the arms of Rod Barajas and Mike McKenry, but Pirates pitchers were also painfully slow to the plate. Opposing baserunners racked up over 23 runs of value against Pirates’ catchers; the average MLB team (27 percent caught stealing) would have given up just six runs.

The Pirates will still have A.J. Burnett, James McDonald and Wandy Rodriguez in the rotation as holdovers, and Jeff Karstens may be back as well. If Russell Martin really did get worse at throwing out baserunners in 2012, it’s an issue that could be exacerbated by the starting rotation’s problems with holding runners on. And in this brave new pitcher-dominated era, the attempts will be fast and furious, and each successful one allowed more costly.

Print This Post

Jack Moore's work can be seen at VICE Sports and anywhere else you're willing to pay him to write. Buy his e-book.

21 Responses to “The Stolen Base Matters More Now”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. TBH says:

    Terrific article.

    +33 Vote -1 Vote +1

  2. Tom says:

    Kind of amusing how the composition of the Jays team reflects this as well. From the good old days of swing for the fences to having 4-5 guys that can nab 20+ bases

    Vote -1 Vote +1

    • Jake says:

      My thoughts exactly.

      Vote -1 Vote +1

      • Jake says:

        Just to build on this, not only is the Jays’ new lineup built for speed, but they’re also built for contact and getting on base. The top of the Jays’ lineup, including Bonifacio in the 9th, probably lead the league in terms of the mix of contact rate, stolen bases, and wOBA. Should really make for a boost in runs.

        Bonfiacio (9)

        Reyes (1)
        Cabrera (2)
        Bautista (3)
        Encarnacion (4)
        Lawrie (5)

        Vote -1 Vote +1

  3. philosofool says:

    I think this is a corollary of the general rule that the value of slugging is inversely proportionate to OBP. This is because, as OBP falls, the value of moving runners over increases.

    Vote -1 Vote +1

  4. SKob says:

    All the better to stick-it to Billy Hamilton haters! Love it!

    Vote -1 Vote +1

  5. David says:

    Your statement – the SB is less risky because there is less to lose. Shouldn’t you also consider the run environment? If we’re in a lower run environment than 2007, then the loss of a run should be considered an even greater loss.

    Shouldn’t you consider the score when counting stolen base opportunities? Also, maybe the inning?

    I do love these articles that show how the game changes over the years. Any chance of showing the data over a longer period of time? Keep up the good work!

    Vote -1 Vote +1

    • Krog says:

      You can’t ever lose a run.

      Vote -1 Vote +1

    • DCN says:

      …and the gain of a run a bigger gain.

      The question is whether the SB increases or decreases your chance of getting runs.

      But, yes, context does matter of course. If the inning is late and the score is tied, you’re more likely to accept an increased chance of scoring one run at the cost of a decreased chance of scoring two or more runs, for example. A lower run environment usually favors the stolen base because it increases the number of situations in which a single run will be more valuable.

      Vote -1 Vote +1

      • CircleChange11 says:

        If I recall correctly from Tango’s book, that the SB is most important if it comes in front of a batter that is basically a “singles hitter”.

        There’s a strong aspect of common sense in that as well.

        Vote -1 Vote +1

  6. Joshua says:

    Interesting. What would this do to run expectancy tables and bunting? Is the 1st to 2nd sac bunt now less offensive than before?

    Vote -1 Vote +1

    • Joshua says:

      (offensive meaning bad, not offensive/defensive)

      Vote -1 Vote +1

    • Bip says:

      It is probably less offensive in the sense you’re using it, but it is probably still too offensive to be used as frequently as it is. I would think bunting would be most valuable in high OBP/low SLG eras, where the value of the almost certain out is at its lowest since getting on base is easier and the value of the advance is highest.

      Vote -1 Vote +1

      • Ryan says:

        I agree that it is probably less offensive than in, say 2000 and probably still too offensive to be used as frequently as it is. However, I would think that bunting would be least valuable in high OBP/low SLG eras (and granted this is a gut-level belief, very open to changing). In high OBP eras, the cost of an out is greater than in low OBP eras, right? So that out you’re punting costs more runs (or a greater fraction of a run, I guess) than in a low OBP era as there is a better chance that the fellow you’ve just sacrificed will move the runner over on his own.

        Granted, I’m not doing much with the SLG component here, but am I missing something?

        Vote -1 Vote +1

  7. Jason says:

    While it makes sense that as OBP gets lower, it’s less likely for the hitter to be able to get that runner over, since each of the 27 outs is just that much more valuable (since the batter is less likely to reach a base throughout the game) isn’t it more of a risk to lose an out on a runner that was one of the few that actually managed to reach base?

    Vote -1 Vote +1

    • Krog says:

      No, because in low run environments it becomes less likely that any baserunner will score. Moving a runner into scoring position becomes more important because the next batters have a lower chance of getting on-base and sustaining the inning.

      Vote -1 Vote +1

  8. Brian L says:

    I think you could look at the HR / K tradeoff in a similar light. Obviously there is a lot more noise because there are more possible outcomes than in the SB / CS scenario, but there is definitely a population of hitters who to some extent are trading HRs for Ks. In general, when less runs are being scored overall, the lost value from striking out is lower than when more runs are being scored. And conversely, when less runs are being scored, the value of hitting a HR is greater than when less runs overall are being scored. So I think you could assume that in 2012 a guy who Ks 180 times in order to hit 35 HRs is more valuable than the same guy in 2007. Moreover, a guy who Ks 200 times to hit 30 HRs in 2012 might be more valuable than that 180 K / 35 HR guy in 2007 – obviously a theoretical example but you get my point.

    Of course, the curse of all fangraphs fans, I cant help but think of Billy Beane and the success of his lineup construction this year – that team set an AL team strikeout record… on the way to the playoffs

    Vote -1 Vote +1

  9. OldScout says:

    Nicely presented evidence. Good to see some degree of balance presented within the usual FG dialogue that historically over glorified the Roidball era style of baseball.

    There’s hope still for this site to offer something other than shrill commentary on stats r us over common sense.

    Vote -1 Vote +1

    • DCN says:

      This is about finding the statistical evidence to confirm that common sense is correct. It’s not a battle of either/or. There should be a positive interaction between statistical and non-statistical insight.

      Vote -1 Vote +1

    • joser says:

      the usual FG dialogue that historically over glorified the Roidball era style of baseball.

      What? That doesn’t describe the usual dialog on this site whatsoever.

      shrill commentary on stats r us over common sense.

      Again, I’m sorry you have such trouble with basic reading comprehension, but it is a common affliction of the elderly (I’ve been dealing with an increasingly senile aging parent, so I’m sympathetic.)

      Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>