FanGraphs Baseball

Comments

RSS feed for comments on this post.

  1. Any home/road splits in % of unearned runs? The home team scorer effect?

    Comment by Nick44 — November 10, 2011 @ 1:20 pm

  2. Good point. Papa Grande (David Murphy is Murphys Law)

    Comment by Blue Jays Fan — November 10, 2011 @ 1:29 pm

  3. Isn’t the league average GB% ~48%? Correct me if I’m wrong, because I’m not exactly sure that’s right, but Buehrle’s career GB% is 45.9%, and he’s been getting ground balls at slightly lower rates in each of the past three seasons…

    Overall point of the article is obviously still valid, but is it right to call Buehrle a ground ball pitcher?

    Comment by Jack Nugent — November 10, 2011 @ 1:30 pm

  4. Two other reasons that weren’t mentioned here: Buehrle doesn’t walk guys and he picks off a lot of guys/limits the running game because of his pickoff move.

    I’d love to see first-to-third against Buehrle vs. other pitchers (Even other similar pitchers). Consider: A reasonably fast runner is on first, but because he’s worried about getting picked off, he gets a bad secondary lead and has to hold up at second on a base hit. The next guy hits into a double play, guy after that makes some other out. That runner is stranded on third when normally he would’ve scored on the double play.

    There’s plenty of permutations here, and I have no idea how many runs Buehrle is saving by the threat of his pickoff move (not to mention the more quantifiable number of runs when you consider actual pickoff outs). Guys don’t steal off him, and I’d bet guys take fewer extra bases, too.

    Comment by JD — November 10, 2011 @ 1:33 pm

  5. This can also be partly attributed to secondary skills. Buehrle helps himself by making plays in the field that a lot of other pitchers don’t make, come backers that get by a lot of pitchers are outs or the start of DP’s with Buehrle on the mound. Guys who use the bunt for hit as a weapon vs. guys who can’t get off the mound might think twice about using it vs. Buehrle.

    The other is that the running game is a non starter vs. Buehrle. Guys simply don’t steal. While Gavin Floyd teamed up with Pierzynski to allow 23 of 25 successful attempts, Buehlre allowed 3 steals all season, with 7 guys caught or picked off. What might be harder to measure with the metrics is how his move impacts base runners. What I mean is that guys are so paranoid about being fooled by his move they usually stick about two feet away from the bag since they can’t tell if he is going home or over to first. As a result, they have awful jumps, which may lead to more DPs and outs at 2B since the fielders can get the runner on account of their horrible leads from first.

    Comment by JK — November 10, 2011 @ 1:34 pm

  6. i think the walks thing is covered with all the mentions of FIP… but you make a very interesting point about holding runners close to the bag.

    Comment by Woodrum's UZR Article — November 10, 2011 @ 1:43 pm

  7. ERA shouldn’t factor into it in the first place, though, right? I thought FIP was a major component of pitcher’s WAR but ERA isn’t. So his FIP feeds into his WAR, and thus $WAR, and thus what he should be paid per win he’s adding…

    Comment by Chris from Bothell — November 10, 2011 @ 1:52 pm

  8. Also, the BABIP factor is dismissed too quickly. Especially for an enormous sample, a .291 BABIP is low for a groundball pitcher. And BABIP has much more of an effect for Buehrle than it does for most pitchers, because he induces so many balls-in-play.

    Comment by Yirmiyahu — November 10, 2011 @ 1:53 pm

  9. I think controlling the running game is the big one. Zambrano too is another one that gets a lot of pickoffs and allows very few steals. And I haven’t watched the other guys mentioned enough but I wouldn’t be surprised if they’re good defensively as well. I think pitcher defense is something that doesn’t get accounted for that at least on the margins can have pretty significant effects on ERA.

    I think it’s pretty reasonable to think that an elite defensive pitcher pitcher could save an extra 5-7 runs per season above average. And five runs over 200 innings is .23 points of ERA, seven is .32. Buehrle’s career FIP-ERA is .3. Definitely something to think about.

    Comment by Jilly — November 10, 2011 @ 2:01 pm

  10. If you use Fielding Bible’s Defensive Runs Saved for the years they are available (starting in 2003), Buehrle has saved 62 runs from his own ledger. His ERA between 2003-2011 is 3.907, his FIP is 4.148, and his xFIP is 4.236. If you assume all runs Buehrle saved during this period would have been earned, and then pretend he saved none of these runs, but broke even in the DRS metric, thus allowing 62 more ER over this period, his ERA would have been 4.191, smack in the middle of his FIP and xFIP. Even if you assume that of Buehrle’s 62 DRS, 10.1% would have been unearned (his career rate), he still saved 56 earned runs. If these 56 runs were added to his ER totals over this time period, his ERA would have been 4.163, again between his FIP and xFIP. Am I forgetting something or would this go a long ways towards explaining his success?

    Comment by FJL — November 10, 2011 @ 2:02 pm

  11. WAR for starting pitchers is simply a combination of FIP and innings, with a park/league adjustment. So, yeah, his ERA is irrelevant to that.

    But this article isn’t about WAR. It doesn’t even mention WAR. The question at hand is why Buehrle’s ERA is always much better than his peripherals. Over such a large sample, it’s more than randomness.

    Comment by Yirmiyahu — November 10, 2011 @ 2:08 pm

  12. Average GB% for pitchers is generally 43-44%.

    Comment by NM — November 10, 2011 @ 2:19 pm

  13. DRS should manifest itself through a low BABIP, but his BABIP isn’t particularly low.

    Comment by kinnerful — November 10, 2011 @ 2:37 pm

  14. 91% of his career runs allowed at home have been earned, 88% of his road runs. Below average at both, but more pronounced on the road. Not sure it’s large enough to be meaningful.

    Comment by Dave Cameron — November 10, 2011 @ 2:38 pm

  15. This would theoretically show up in strand rate, though. If Buehrle is putting an average amount of baserunners on but keeping them from scoring, his LOB% should be among the highest in the league. It’s not.

    Comment by Dave Cameron — November 10, 2011 @ 2:39 pm

  16. What about double plays? Does he get a lot of those?

    The reason the Fielding Bible guys love him is because he completely ruins the running game. I think he allowed something like 3 or 4 steals all of last year. Perhaps he gets a lot of guys on first base, keeps them there, and then gets double plays. That wouldn’t necessarily show up in LOB%, would it?

    Comment by Eminor3rd — November 10, 2011 @ 3:08 pm

  17. But at the very end, he notes, “… if a team is going to pay for Buehrle’s services, they should adjust their expectations of value down slightly from what ERA says he is worth.”

    Which seems odd, since I thought one could use WAR & $/WAR to figure out fair market value, and that was pretty much that.

    Comment by Chris from Bothell — November 10, 2011 @ 3:13 pm

  18. More than half his DRS (36/62) came from the baserunning control component, rSB. Also BABIP would be more heavily weighted by the runs saved/caused by the other fielders on the team than the pitcher on the mound (how many times does a pitcher make a play in the field compared to other fielders?)

    Comment by FJL — November 10, 2011 @ 3:31 pm

  19. A higher % of the hits Buerhle gives up are singles compared to the average MLB pitcher. Over his career, he’s at 68.8% singles, 19.2% 2B, 1.5% 3B, and 10.5% HR.

    The MLB average for singles was 67.2% in 2011, 67.1% in 2010, and 66.2% in 2009. Doubles are basically at 20.0% and about 2.1% for 3B.

    HRs ranged from 11.6% in 2009 to 10.8% in 2011.

    I’m not sure FIP or xBABIP takes into account the distribution of non-HR hits, but this has to help his ERA relative to the average.

    Comment by Steve — November 10, 2011 @ 3:43 pm

  20. Is the league average GB% recorded anywhere at FanGraphs, or somewhere else perhaps?

    Comment by Jack Nugent — November 10, 2011 @ 3:46 pm

  21. So a team with a good defense has a better shot at getting value out of the contract I suppose.

    Comment by t ball — November 10, 2011 @ 4:18 pm

  22. A good bullpen is another factor..having good pitchers come in to mop up any messes he might have left over the years.

    Comment by slash12 — November 10, 2011 @ 4:28 pm

  23. Average GB rate for starters was 44.4% in 2011

    http://www.fangraphs.com/leaders.aspx?pos=all&stats=sta&lg=all&qual=0&type=2&season=2011&month=0&season1=2011&ind=0&team=0,ss&players=0

    Comment by Taylor — November 10, 2011 @ 5:06 pm

  24. Wouldn’t this show up in his LOB%, though?

    Comment by Kevin S. — November 10, 2011 @ 7:48 pm

  25. Very in depth article. Great job putting it together. All great points. His career Babip might be lower then it is if the White Sox had put better defensive players behind him over the course of his career. Jose Valentin, Juan Uribe, Royce Clayton. Without closely looking at the stats I would make the assumption that those White Sox teams have been consistenely below average defensively. He also plays half his games in a decent homerun park so his HR/FB might be a bit better then it looks at first glance.

    Comment by Shane Heathers — November 10, 2011 @ 8:37 pm

  26. Outstanding stuff, Dave. Extremely relevant to hotstove league. Thanks.

    Comment by PepeShady — November 10, 2011 @ 9:19 pm

  27. Couple of problems with the unearned run theory and him being a groundball pitcher and getting a bias.

    1) Shouldn’t his BABIP be lower than normal if hits are being classified as errors?
    2) Presumably this unearned run bias would not be specific to Buerhle, but to all heavy groundball pitchers?

    It would seem if one was to throw out this hypothesis, it would be worth checking the BABIP and ERA-FIP delta of other groundball pitchers (and I suspect that there is not a correlation).

    Several other problems:
    1)If the premise is to be believed… it still is only 30% of the delta, which obviously means there’s another 70% unaccounted for. So even if theory is dead on it is missing the major cause of the gap (which is still an unknown)

    2) The RA/9 comparisons include some pitchers with significant NL time… is that really a fair comparison? How would that list look with a league adjustment? (there’s also potentially a significant park effect with some of those guys too)

    Seems like a lot of handwaving for what amounts to a 0.09 ERA difference, especially when the FIP/ERA gap is 0.30

    Comment by Tom — November 10, 2011 @ 10:18 pm

  28. Dave – I thought LOB% does not factor in runners getting on via error?

    If Buerhle is indeed seeing a greater # of errors behind, his LOB% is actually understating his strand capability (as the runners reaching via error are not being accounted for)…

    So if Buerhle is maintaining a league average or slightly above average LOB% despite these hidden extra runners, it might actually mean his PO/CS/holding runners closer skill is actually helping and simply offsetting these unaccounted for runners.

    The other thing is there is also a generic HR factor in the LOB calculation.not sure which way that’s cutting for Buehrle…

    Comment by Hank — November 10, 2011 @ 11:31 pm

  29. Is Hawk a scorer? I guess he couldn’t be, since Buehrle’s ERA is greater than 0.00

    Comment by Steve — November 11, 2011 @ 10:50 am

  30. Great article.

    After following baseball for so long, it’s rare to learn something new about.

    Comment by Chris — November 11, 2011 @ 1:27 pm

  31. As a big Buehrle fan I have been thinking about this alot lately for some reason and I think the way he controls the running game must enter into it somehow.

    Buehrle had 6 pickoffs last year so that’s two free innings. Not alot. That can;t be the whole story. He also only had 3SB in 10 attempts – even with AJ behind the plate for much of the year. So that’s 2.1 more free innings (up to 1/2 a game now) and a whole lot of runners who stayed at first. I don’t know the ins and outs of strand rate but it is certainly easier for a pitcher to strand a runner at 1st than at 2nd or 3rd. I see what you are saying and that should show up in a better strand rate. I don’t know why it doesn’t.

    Another non standard thing he does to his benefit is play defense. He had 9 DRS last year. Isn’t that nearly 1 full defensive WAR? That would be impressive for any fielder playing only 32 games. I know that should show up in these stats too because it doesn’t matter who makes the play but it is another thing that he brings to the table that most pitchers don’t.

    He’s simply unique these days and he works fast. In my opinion, he’s one of the most fun guys to watch pitch in person because he does all those little things well.

    Comment by MikeS — November 11, 2011 @ 1:37 pm

  32. True of any pitcher. More true of any GB pitcher.

    Comment by MikeS — November 11, 2011 @ 1:39 pm

  33. You know that the perceived hometown scorer effect theoretically results in LESS errors (and therefore higher ERAs), right?

    I don’t think Hawk would be willing to give all those errors to ALEXEI

    Comment by JG — November 11, 2011 @ 2:16 pm

  34. It isn’t low for someone like Weaver, but it is low for someone who has groundball tendencies like Buehrle.

    Comment by JG — November 11, 2011 @ 2:21 pm

  35. Hard to say that isn’t significant when I’m looking at his career numbers vs. league average for the past three seasons, but… I doubt that’s statistically significant.

    Comment by JG — November 11, 2011 @ 2:23 pm

  36. If the White Sox had a lower home town scorer effect (i.e. they gave out more errors that other home town scorers) this could potentially still be an explanation.

    My post made no mention of bias either way in the effect, but I assumed that people would understand my point. Gavin Floyd has a similar discrepancy between ER and R. Look it up.

    http://www.fangraphs.com/leaders.aspx?pos=all&stats=pit&lg=all&qual=0&type=0&season=2011&month=0&season1=2008&ind=0&team=4&players=0

    Comment by Nick44 — November 12, 2011 @ 1:30 pm

  37. Tom,

    1) The reason you see GB pitchers with typically higher BABIP and runners reaching on errors is really two symptoms of the same root cause: groundballs are harder to field for outs than fly balls are (at least those that stay in the park). No matter whether the scorer sees those as hits or errors, in general, GB pitchers give up more of them.

    2) Yes, there is an unearned run bias for groundball pitchers. Most typically give up more errors, leading to a higher number of unearned runs. R/9 is a better metric to use, as it removes this bias.

    You’ll more often see a positive delta in ERA-FIP for groundball pitchers than for flyball pitchers, and the flipside of that as well — more of a negative delta for flyballers. That’s only a very generalized statement, as it’s far from 100%. It’s only a slight trend that you see when looking at a large sampling of pitchers. Look at the above pitchers that Dave lists. Most of them are well known as heavy flyball pitchers, some of the most flyball-happy over the last decade. I’d have to look around for links, but this isn’t something new — this tendency been noted for quite a while. Googling should turn up something for you fairly easily.

    That’s what makes Buehrle so mysterious: it would be one thing if he were a flyball pitcher, and you could explain most of the negative delta with that. Being a groundball pitcher with that career-long difference is pretty unusual. He’s either been the consistent recipient of a lot of luck, or he’s doing something to prevent runs that pitching metrics have difficulty capturing.

    Comment by Nathaniel Dawson — November 12, 2011 @ 4:09 pm

  38. Hawk doesn’t know how to read or use a pencil, that’s another reason Hawk clearly isn’t the scorer.

    Comment by antonio bananas — November 13, 2011 @ 10:04 pm

  39. Like Jamie Moyer, Mark Buehrle defines “crafty”.

    Comment by NEPP — December 7, 2011 @ 7:13 pm

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>


Current day month ye@r *

Close this window.

0.080 Powered by WordPress