Cardinals news from a Sabermetric point of view

Should a manager pull a reliever who issues two walks in an inning?

When Jason Motte walked his second batter of the inning Monday night, it didn’t require a soothsayer to figure out how the game would end. When a reliever walks two batters in one inning, it seems like certain trouble. But what do the numbers say? Does walking two batters in an inning really entail imminent tragedy?

In 2011, 90 pitchers who were in save situations in the eighth inning or later yielded two or more walks. Here’s how the outcomes break down:

  • Saves: 36
  • Blown saves: 54

That’s a save rate of only 40%. In other words, not good. All right, you say, including eighth-inning save opportunities isn’t really fair, since few pitchers earn saves starting in the eighth inning (though they get debited with blown saves). True enough. So all 64 ninth-inning save opps in which a pitcher threw two or more walks:

  • Saves: 32
  • Blown saves: 32

An even 50% — still really bad. And keep in mind that these are save opportunities, which means that it’s not even the most crucial relief situations (e.g., a reliever can get one out with a three-run lead and the bases empty). And it makes sense that bad outcomes follow multiple walks: Not only does the pitcher give additional baserunners to the opponent, he undoubtedly doesn’t have his best stuff. And as we saw with Motte the other night, that means he’s more likely to get beaten on his secondary pitches or simply groove a hittable pitch.

Given then the reality of what happens when relievers allow two walks in an inning, why wouldn’t managers pull their pitchers? After all, if all you have is a 50% chance of winning (at best) and quite possibly a better chance of losing the game, why risk it? It would seem that the worst thing would be to have to use another pitcher. But isn’t that better than a likely loss? This looks like an easy call for us: Any time a reliever allows two walks in an inning, it should entail another pair of walks — a walk to the mound by the manager, and a walk in from the bullpen from another reliever.

7 Responses to “Should a manager pull a reliever who issues two walks in an inning?”

  1. osbornesmith Says:

    I think the analysis is a little misleading here. Walking two batters in an inning is not a good sign, but just because 50% of those situations result in an eventual blown save (not necessarily a loss) does NOT mean pulling the pitcher is a good move.

    Sure relievers who create these situations blow 50% of their save opportunities, but to come to a conclusion about what to do you’d have to compare that figure to the success rate after bringing in another (presumably lesser quality) reliever in a situation with 2 men already on base. If that success rate is > 50%, then you might have an argument for pulling the closer.

  2. osbornesmith Says:

    I forgot one point. You mention how the two walks could be a sign your closer is “off” and that does complicate my point about how the next reliever in line likely is of lesser quality. Still, I’m not sure I’m sold on creating a “rule of thumb” based on the 50% success rate stat. Just my opinion.

  3. EFitz Says:

    Do the stats above cover if one of the walks was intentional? If a closer walks one batter on say a seven or eight pitch at-bat, then issues an intentional walk, I don’t think that necessitates a move to the bullpen.

    That wasn’t the case with Motte the other night and with relievers available, I think it’s safe to say Matheny should have gone to the ‘pen.

    Then again, it could be him pushing Motte to learn how to battle through these situations.

  4. MuleRider342 Says:

    I agree with osbornesmith that there shoujld probably be a little deeper analysis to determine if yanking the guy is truly the best course of action, but I really liked and agreed with much of this post in challenging the conventional wisdom of “live or die with the (anointed) closser),” especially in situations he’s noticeably off.

  5. Pip Says:

    Thanks for the comments, everyone. Let me try to respond to them all:

    just because 50% of those situations result in an eventual blown save (not necessarily a loss) does NOT mean pulling the pitcher is a good move.

    True. The manager must consider the replacement value, as it were, of any subsequent pitcher.

    I’m sold on creating a “rule of thumb” based on the 50% success rate stat.

    The numbers show that this situation is at best a break-even deal. But keep in mind that this includes Robb Nen saves, so it’s likely that in close games, it’s worse. A rule of thumb isn’t ironclad; some extenuating circumstances may apply. But it’s enough to make me want to change pitchers.

    Do the stats above cover if one of the walks was intentional?

    Good point, and I didn’t (and don’t know if I can) distinguish IBBs.

    there shoujld probably be a little deeper analysis to determine if yanking the guy is truly the best course of action

    Certainly. I don’t claim to have proven anything here. But based on some quick numbers, I think we’re looking at something that probably has some truth behind it. Maybe this November I’ll spend a little more time researching (unless one of you wants to do it!).

  6. hopandgator Says:

    Pip-

    My apologies for harping on fine details here, but for the reasons emphasized by osbornesmith I’m still not convinced that we’re in a position to believe the claim you make at the end of your last comment that “based on some quick numbers…we’re looking at something that probably has some truth behind it.” The crucial relevant stat here is how often a relief pitcher who inherits two runners in the ninth inning of what was a save situation even before those runners reached base protects the lead. And as osbornesmith suggests, I think many of us suspect (without having seen the numbers on it) that it is not likely that such leads are protected more than half the time – at least not significantly more. And if that’s right, then we don’t yet have a sound case for pulling the closer.

    To be clear, the position you hold on pulling the closer may well, in the end, be the right one. It’s just that I don’t think we can make an informed judgment either way based solely on the numbers you’ve given — we’ve got to see the numbers on the other side.

  7. osbornesmith Says:

    Hate to pile on because I do think managers often blindly stick with closers too long just because they’re closers, but consider this analogy.

    Yadi is batting .100 (1 for 10) this year after falling behind 0-2 (per ESPN). Would anyone take that stat and argue Yadi should be yanked for a pinch hitter with a .250 average whenever he falls behind 0-2? No, because we all recognize that. 250 hitter probably doesn’t get a hit 25% of the time when the county’s already 0-2.

    I think I’m with hopandgator here that while intriguing, there needs to be a lot more analysis before anyone considers adopting this policy.

Leave a Reply

You must be logged in to post a comment.