SoxProspects News
|
|
|
|
Legal
Forum Ground Rules
The views expressed by the members of this Forum do not necessarily reflect the views of SoxProspects, LLC.
© 2003-2024 SoxProspects, LLC
|
|
|
|
|
Forum Home | Search | My Profile | Messages | Members | Help |
Welcome Guest. Please Login or Register.
|
Post by jmei on Dec 8, 2014 13:47:06 GMT -5
My question about these outlandish pitch framing numbers is, can we find the effect in pitcher performance? Do we have an example of a team that went from a poor pitch framer to a good one, and everyone's ERA dropped by a third of a run? Because with some of the numbers being thrown around for elite pitch framers, it seems like it should have an obvious effect on ERAs, strikeout rates, etc. Has anyone actually shown this to be the case? Russell Martin joining the Pirates in 2013 is one example that comes to mind. Compare how the following Pirates starters performed in 2013-14 with how they performed from 2011-12: | 2011-12 | 2013-14 | Francisco Liriano | 8.54 K/9, 4.89 BB/9, 5.05 ERA, 4.45 FIP | 9.41 K/9, 4.01 BB/9, 3.20 ERA, 3.26 FIP | A.J. Burnett | 8.10 K/9, 3.33 BB/9, 4.31 ERA, 4.13 FIP | 9.85 K/9, 3.16 BB/9, 3.30 ERA, 2.80 FIP | Jeff Locke | 7.33 K/9, 4.05 BB/9, 6.36 ERA, 5.43 FIP | 6.47 K/9, 3.75 BB/9, 3.69 ERA, 4.18 FIP | Charlie Morton | 5.47 K/9, 3.57 BB/9, 4.01 ERA, 3.86 FIP | 6.95 K/9, 3.06 BB/9, 3.52 ERA, 3.67 FIP |
|
|
danr
Veteran
Posts: 1,871
|
Post by danr on Dec 8, 2014 13:59:16 GMT -5
I can see how hard it is to value pitch-framing, especially since it is so new as a metric. It occurs to me that a really good pitch-framing catcher is going to cause more problems for the really good hitters who know the strike zone than he is for the average or bad hitters, or the free swingers. Those guys who really know the strike zone and have high OBPs, usually take close pitches that they see as out of the strike zone. With a really good pitch-framer, many of those pitches may be called strikes.
The average or bad hitters, or the free swingers, may not be as affected because the average/bad hitters can't tell whether or not is a strike, and the free swingers probably swing at them.
And the drop-off of those ERAs may be further evidence of this.
|
|
|
Post by James Dunne on Dec 8, 2014 14:06:15 GMT -5
The numbers getting thrown around are a lot higher than 2.5 wins per 1000 innings, though. At 4.0 wins per 1000 innings, he'd be roughly 1/7-1/10 as valuable per inning as the best pitchers in the league, and 1/3-1/4 as valuable per inning as someone useful like John Lackey. That's beyond where I'm comfortable. One post by mgoetze in the trade proposal subforum projected Vazquez to be a 3.5 WAA (not WAR, WAA) player, and called that projection conservative. That's in the realm of the top 20 position players in the game.
|
|
|
Post by James Dunne on Dec 8, 2014 14:11:49 GMT -5
My question about these outlandish pitch framing numbers is, can we find the effect in pitcher performance? Do we have an example of a team that went from a poor pitch framer to a good one, and everyone's ERA dropped by a third of a run? Because with some of the numbers being thrown around for elite pitch framers, it seems like it should have an obvious effect on ERAs, strikeout rates, etc. Has anyone actually shown this to be the case? Russell Martin joining the Pirates in 2013 is one example that comes to mind. Compare how the following Pirates starters performed in 2013-14 with how they performed from 2011-12: Burnett isn't really a great example here though, since he was throwing to Martin in 2011 and 2013 and to other catchers in 2012 and 2014. 2011 was probably the second worst season of his career, and 2013 was, by FIP, his best. EDIT: I tried bolding Burnett in the table and the whole formatting went to heck. Sorry about anyone who tried to read that.
|
|
|
Post by jmei on Dec 8, 2014 14:12:42 GMT -5
The numbers getting thrown around are a lot higher than 2.5 wins per 1000 innings, though. At 4.0 wins per 1000 innings, he'd be roughly 1/7-1/10 as valuable per inning as the best pitchers in the league, and 1/3-1/4 as valuable per inning as someone useful like John Lackey. That's beyond where I'm comfortable. One post by mgoetze in the trade proposal subforum projected Vazquez to be a 3.5 WAA (not WAR, WAA) player, and called that projection conservative. That's in the realm of the top 20 position players in the game. I'm saying the +4 win numbers thrown around by mgoetze look wrong to me. By both BP and Statcorner, Vazquez was roughly a +2.5 to +3 win framer last year if you prorate his half season to a full season's worth of defensive reps. Moreover, per BP's data, Vazquez got more framing opportunities than the average catcher last year. If we adjust that to per 7000 opportunities (roughly the league-average for a starter catcher), they have him as a +2.3 win framer.
|
|
|
Post by James Dunne on Dec 8, 2014 14:19:32 GMT -5
Yeah, that's more in the range I'm comfortable.
Just for my understanding (and apologies if this has been mentioned elsewhere), why did Vazquez get so many more framing opportunities? Was it mostly a product of Red Sox pitchers' inability to get swings-and-misses down the stretch? Or was it just a fluky sample size thing?
|
|
|
Post by Oregon Norm on Dec 8, 2014 14:20:32 GMT -5
We might want Eric to chime in, here. Does he feel the undervaluation is that dramatic? Has pitch-framing been so under-valued that a player that good, one who can get you over 3+ additional wins above an average catcher - and a lot of guys with spots reserved for them on ML rosters are just that - is realistic?
|
|
|
Post by beasleyrockah on Dec 8, 2014 14:29:15 GMT -5
Yeah, that's more in the range I'm comfortable. Just for my understanding (and apologies if this has been mentioned elsewhere), why did Vazquez get so many more framing opportunities? Was it mostly a product of Red Sox pitchers' inability to get swings-and-misses down the stretch? Or was it just a fluky sample size thing? If Vazquez continuously "steals" strikes at an elite level the pitchers will notice and test the boundaries of the strike zone more often than they would with an average/below average framer.
|
|
|
Post by jmei on Dec 8, 2014 14:39:51 GMT -5
Just for my understanding (and apologies if this has been mentioned elsewhere), why did Vazquez get so many more framing opportunities? Was it mostly a product of Red Sox pitchers' inability to get swings-and-misses down the stretch? Or was it just a fluky sample size thing? I think for a sample as small as that one, it's probably mostly noise, but I could also imagine it being partially sustainable skill. The number of chances are based off pitchf/x location, so the only way Vazquez can "create" more chances is if his pitchers threw more borderline pitches. Maybe that's something Vazquez can affect (e.g., by calling for more borderline pitches), but it seems more flukey than anything else. There's also the fact that the pitching was really bad the second half of the year, which meant there were more raw pitches thrown, which means Vazquez got more chances than the average player. The Red Sox were fifth in the league in total batters faced over the second half of the season and eighth in the league in pitches thrown over the entire year.
|
|
|
Post by Oregon Norm on Dec 8, 2014 15:02:27 GMT -5
He also had guys like Webster, for whom the "edge of the plate" had been an abstract concept, at least till Vazquez became his regular catcher. Webster alone probably added a wealth of framing chances to that portfolio. Just my take on it, but I don't think it's a coincidence the guy was as good as he was in the last month. I think having his Pawtucket catching buddy behind the plate was invaluable.
|
|
|
Post by jimed14 on Dec 8, 2014 15:05:15 GMT -5
He also had guys like Webster, for whom the "edge of the plate" had been an abstract concept, at least till Vazquez became his regular catcher. Webster alone probably added a wealth of framing chances to that portfolio. Just my take on it, but I don't think it's a coincidence the guy was as good as he was in the last month. I think having his Pawtucket catching buddy behind the plate was invaluable. And it has to be even more exaggerated in moving from AJP to Vazquez.
|
|
|
Post by mgoetze on Dec 8, 2014 15:34:28 GMT -5
I took my +4 number from Eric: forum.soxprospects.com/thread/1988/vazquez-pitch-framing-logEric has Vazquez pitch-framing at a +3.8 WAR rate. I assume that replacement-level catchers have average pitch framing for now, and in any case I'm pretty sure all that data is measuring Vazquez against league average, not some hypothetical replacement level, which is why I say WAA rather than WAR.
|
|
|
Post by jmei on Dec 8, 2014 15:48:17 GMT -5
I took my +4 number from Eric: forum.soxprospects.com/thread/1988/vazquez-pitch-framing-logEric has Vazquez pitch-framing at a +3.8 WAR rate. I assume that replacement-level catchers have average pitch framing for now, and in any case I'm pretty sure all that data is measuring Vazquez against league average, not some hypothetical replacement level, which is why I say WAA rather than WAR. He's either calculating it wrong or I'm misreading something. On his BP player card, BP has Vazquez as +13.1 framing runs, and per Statcorner, he's +12.2 RAA. Prorate that to a full season's worth of reps and you get the 25-30 run range I discussed earlier. There's no replacement level adjustment needed to the framing RAA numbers before you add it to the rest of his WAR. The WAR equation already includes a replacement level adjustment, so there's no need to add another one for framing. For example, the WAR equation for position players is [Batting Runs Above Average + Baserunning Runs Above Average + Fielding Runs Above Average + Positional Adjustment + League Adjustment + Replacement-level Adjustment]. If you want to adjust that for pitch framing, just add in Framing Runs Above Average. So there's no issue here (though I'm not sure if I'm explaining it right).
|
|
|
Post by Chris Hatfield on Dec 8, 2014 16:47:26 GMT -5
I took my +4 number from Eric: forum.soxprospects.com/thread/1988/vazquez-pitch-framing-logEric has Vazquez pitch-framing at a +3.8 WAR rate. I assume that replacement-level catchers have average pitch framing for now, and in any case I'm pretty sure all that data is measuring Vazquez against league average, not some hypothetical replacement level, which is why I say WAA rather than WAR. He's either calculating it wrong or I'm misreading something. On his BP player card, BP has Vazquez as +13.1 framing runs, and per Statcorner, he's +12.2 RAA. Prorate that to a full season's worth of reps and you get the 25-30 run range I discussed earlier. There's no replacement level adjustment needed to the framing RAA numbers before you add it to the rest of his WAR. The WAR equation already includes a replacement level adjustment, so there's no need to add another one for framing. For example, the WAR equation for position players is [Batting Runs Above Average + Baserunning Runs Above Average + Fielding Runs Above Average + Positional Adjustment + League Adjustment + Replacement-level Adjustment]. If you want to adjust that for pitch framing, just add in Framing Runs Above Average. So there's no issue here (though I'm not sure if I'm explaining it right). Eric has always mentioned that he takes the BP data and tweaks it somehow. That's the number I've always said didn't make sense. But hey, I'm not the one who has been photographed for ESPN the Magazine kneeling with a broadsword, so what do I know...
|
|
|
Post by sarasoxer on Dec 8, 2014 17:04:08 GMT -5
With knowledge of a particular catcher's pitch framing prowess, at what point do 'the umpires strike back' ?...that is take that into account when calling pitches thereby negating or partially negating that attribute. That's really a rhetorical question but I would expect some counter-effect. Pitch framing, as someone above mentioned, is really employed to fool or manipulate the plate umpire's decision. Nobody likes being intentionally mislead.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,933
|
Post by ericmvan on Dec 8, 2014 18:10:08 GMT -5
I have to update the pitch-framing log thread. I think I have Vazquez at 3.4 right now.
The possible flaw in these metrics is that they might be assigning some value to the catcher that belongs to the pitcher. In fact, there is probably an interaction that should be assigned to the front office, for matching a pitcher (or staff) who could benefit from plus pitch framing to a catcher who can provide it, or to getting a subpar framer with good other skills to handle a staff that mostly has guys who benefit little from it.
Note that Lester with Oakland had plus pitch framing numbers, pitching to catchers who were otherwise negative.
Alas, I think I'm too busy to do much more work on this, for at least a year. Right now, you can start by taking half the measured value, which is conservative, and see what conclusions that leads you to. Sometimes, it's the same conclusion: Vazquez is terrific either way, and Jaso is below replacement level either way.
|
|
|
Post by larrycook on Dec 8, 2014 20:59:46 GMT -5
At the end of the day, having Martin behind the dish, will make the jays pitchers better.
I am hoping Vazquez learns how to handle a staff as well as Martin does.
Plus it would make signing lariano a good move for us.
|
|
|
Post by pedroelgrande on Dec 8, 2014 21:52:04 GMT -5
Would Justin Masterson be one helped by an elite pitch framer?
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,933
|
Post by ericmvan on Dec 8, 2014 22:56:15 GMT -5
Would Justin Masterson be one helped by an elite pitch framer? I have no sense of how to tell that without doing a study of some sort. (IOW, sorry, don't have time to figure it out!) Eric has always mentioned that he takes the BP data and tweaks it somehow. That's the number I've always said didn't make sense. But hey, I'm not the one who has been photographed for ESPN the Magazine kneeling with a broadsword, so what do I know... It's not a broadsword, you heathen! It's Andúril, Flame of the West, the sword that cut the Ring off of Sauron's finger. (Or a replica of same that my friends and I found in its box, set out with the trash, with a "Free" post-it note attached. We share custody. I'm geeky enough to think such a thing is cool, but not so geeky to buy one! The photographer suggested the pose, and I enthusiastically went along, in part because it gives a somewhat misleading impression; I doubt that too many people who own such a replica prop have also seen their favorite postpunk band 268 times. Mission of Burma, of course.)
|
|
nomar
Veteran
Posts: 10,828
Member is Online
|
Post by nomar on Dec 8, 2014 23:10:45 GMT -5
Would Justin Masterson be one helped by an elite pitch framer? Judging by his walk rate I bet he would.
|
|
|
Post by jmei on Dec 8, 2014 23:21:58 GMT -5
I have to update the pitch-framing log thread. I think I have Vazquez at 3.4 right now. How did you get this number? I briefly read through that thread and you talked about adjusting for framing chances, but you suggested that this would adjust Vazquez's framing numbers down, not up, yet your valuation is a half run higher than BP's prorated seasonal numbers.
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,933
|
Post by ericmvan on Dec 9, 2014 4:44:56 GMT -5
I have to update the pitch-framing log thread. I think I have Vazquez at 3.4 right now. How did you get this number? I briefly read through that thread and you talked about adjusting for framing chances, but you suggested that this would adjust Vazquez's framing numbers down, not up, yet your valuation is a half run higher than BP's prorated seasonal numbers. I'm not sure where you're getting your number. This one's very straightforward. 1080 innings for 120 games / 458.3 innings * 13.1 runs saved by count * .985 for his chances per inning compared to average / 8.95 runs per win = 3.4 WAR. He saved 15.4 runs by call, and my earlier study showed that the best predictor for future framing would be something like 85% that and 15% runs by count, so using runs saved by count is conservative.
|
|
|
Post by jmei on Dec 9, 2014 7:19:41 GMT -5
Ah, you're using a different runs/win number. Carry on.
|
|
|
Post by mgoetze on Dec 9, 2014 7:31:23 GMT -5
FanGraphs runs/win by year since 2012: 9.544, 9.264, 9.117. Not unreasonable to project 8.95 as the next number in that series. Then again, longer term it's been somewhat cyclical and 9.117 is the lowest number since 1981...
|
|
ericmvan
Veteran
Supposed to be working on something more important
Posts: 8,933
|
Post by ericmvan on Dec 10, 2014 7:17:06 GMT -5
Just for my understanding (and apologies if this has been mentioned elsewhere), why did Vazquez get so many more framing opportunities? Was it mostly a product of Red Sox pitchers' inability to get swings-and-misses down the stretch? Or was it just a fluky sample size thing? I think for a sample as small as that one, it's probably mostly noise, but I could also imagine it being partially sustainable skill. The number of chances are based off pitchf/x location, so the only way Vazquez can "create" more chances is if his pitchers threw more borderline pitches. Maybe that's something Vazquez can affect (e.g., by calling for more borderline pitches), but it seems more flukey than anything else. There's also the fact that the pitching was really bad the second half of the year, which meant there were more raw pitches thrown, which means Vazquez got more chances than the average player. The Red Sox were fifth in the league in total batters faced over the second half of the season and eighth in the league in pitches thrown over the entire year. Actually, not true, to my surprise. A "framing chance" is a pitch taken for a strike or ball, regardless of location. They really ought to calculate not only the predicted number of strikes based on the location, but some measure of variance. The simplest would be SUM(MIN(strike % for location, 1-strike %))/Chances, which would be close to 0 if all pitches were either down the middle or in the dirt, and 0.5 if all were perfetly on the effective black.
|
|
|