Page 2 of 2 FirstFirst 12
Results 21 to 24 of 24

Thread: Why I think SB is underrated

  1. #21
    Quote Originally Posted by brett View Post
    That is the effect of SB, which actually is going to be in the WPA anyway since the steal value is determined by what happens afterwards. The effect of having a potential base stealer on first is also quantifiable, but not currently in WAR.
    WPA is determined by the run values in an RE24 table, so the so-called intangible effects of base-stealing are not going to be included. WPA is context-dependent, so e.g., a stolen base with a tie game in the bottom of the ninth is worth more than a stolen base while leading by five runs earlier in the game. But the stolen base has the same run value, .2 (at FG), in either situation. It's just that adding that run value has a greater effect on the outcome of the game in the ninth inning scenario than it does in the earlier situation.

    Run values are usually determined from large amounts of data from past games, which allow one to calculate the average number of runs scored from any particular base-out state. These data are context-independent not only in terms of the WPA-relevant factors--inning and score--but also in terms of the actual player. I.e., the run value of having someone at first with no out is not affected by whether the runner is a threat to steal second or not. As I said before, it might be possible to address this with enough data, but it isn't addressed in the RE24 table, and therefore not in WPA, either.

  2. #22
    Join Date
    Jan 2007
    Blog Entries
    Quote Originally Posted by Stolensingle View Post
    …The one conclusion that is supported by a large body of evidence is that on average most batters most of the time hit the same whether runners are on base or not--this is the rationale for making the value of hits context-independent….
    Could you please point out some of that large body of evidence? I did a quick search and found this.

    That article sure SEEMS to show there is a significant improvement for hitters with runners on base. But I suspect that hitter performance improvement is because pitcher performance decreases.
    The pitcher who’s afraid to throw strikes, will soon be standing in the shower with the hitter who's afraid to swing.

  3. #23
    A good question might be whether singles and walks are worth more for base stealers (independent of their hitting numbers or spot in the lineup). In that case, the. Alien is already there, but showing up in following batter's stats and we would have to dock them.

  4. #24
    Quote Originally Posted by scorekeeper View Post
    Could you please point out some of that large body of evidence? I did a quick search and found this.

    That article sure SEEMS to show there is a significant improvement for hitters with runners on base. But I suspect that hitter performance improvement is because pitcher performance decreases.
    All right, this is a good point that you bring up. But note that:

    a) Most of the difference in OPS—the metric used to draw the conclusion—is in OBP. There is some increase in BA and SLG, but it’s much less. IOW, batters tend to walk more with runners on base. This isn’t surprising, as there are several base-out situations—runner on second, runners on second and third, even occasionally runner on third, or runners on first and third—in which a pitcher is more willing to walk a hitter, to face a less dangerous hitter or to improve the chances of a DP. I’m talking about what is nominally an unintentional walk, but even intentional walks matter here, because that is an improvement over what statistically a batter would be expected to do if allowed to hit.

    b) As the author notes, there’s probably a selection bias in these data. I.e., relatively poor pitchers are more likely to have men on base than better pitchers. This is one of the problems I was getting at when I said a thorough exploration of this issue would require an approach somewhat analogous to that used for pitch-framing. The author apparently believes that while this selection bias matters, it wouldn’t entirely remove the effect, but in the absence of a rigorous study, we don’t really know. Another factor, that the author alludes to, is problems with covering bases. I suspect this has a lot to do with the BA increase, and it might be addressed by comparing BABIP in these situations.

    c) Even taking the stats at face value, they say nothing about base stealers vs. slow baserunners. There may be an improvement in hitting when runners are on base, but as far as the stats cited in this article show, that is independent of the runners’ threat to steal. Indeed, since the numbers aren’t broken down to specific base-out situations, a situation where stealing isn’t even possible may be as good as one where it is.

    d) Finally, based on the data cited in this article, we can estimate very roughly the maximum amount of runs above average baserunners might add. The article finds an average of 31 points increase in OPS; league average OPS is roughly .700, so that is about 4.5%. How many extra runs is that worth? In the wOBA discussion at FG, it’s noted that a 20 point increase in wOBA is worth about 10 extra runs per 600 PA. League average wOBA is about .320, so a 20 point increase represents about 6%. From these numbers, we might estimate that the OPS increase is worth roughly 7.5 runs per year.

    However, it’s a little more complicated than that. Situations in which baserunners exist represent a little less than 50% of all base-out situations, so the increase must be cut in about half, to about 3.5 runs per year. Against that, though, the average run value of an OPS increase is greater when we consider just situations where baserunners exist. I don't want to get into the math, but very roughly 40% of the increase of run value for a walk or a hit is due to getting on base and scoring, while the other 60% is due to advancing runners already on base. This means that roughly we increase the extra runs to about 8, or a little less than 1 WAR. This is very roughly what the reported OPS increase would be worth, with no evidence as yet that any of it is due to the threat of base stealing.
    Last edited by Stolensingle; 05-21-2017 at 05:06 PM.

Page 2 of 2 FirstFirst 12


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts