Archive for PITCHf/x
I know it has been a little while since he pitched, but I wanted to take a look at Ivan Nova‘s last outing. It was one of his most impressive performances of the season despite it being his first start back from injury. In total, Nova went six innings, giving up just two runs on four hits (only two of which went for extra bases) and two walks, with eight strikeouts.
Honestly, my expectations were pretty low for Nova in this start. He has been hit hard all year, giving up a ridiculous quantity of extra-base hits en route to a 4.81 ERA (including his most recent outing). While his strikeout rate is at a career high and his walk rate has been pretty low, one would expect him to be having a career year. Instead, Nova has been one of the weak points in the Yankee rotation, and I figured that missing time with a shoulder injury would likely cost him his rotation spot. However, Nova proved the doubters wrong with a strong outing. I was curious to see what was working for him.
Looking at Nova’s outing on Brooks Baseball, the first thing that comes to mind is his fastball. His average velocity on the pitch jumped a good 0.6 MPH, going from 93.46 to 94.07. Just to put that number in context, an average fastball of 94.07 MPH would place Nova 7th among qualified starters in average fastball velocity (behind usual suspects Stephen Strasburg, David Price, Jeff Samardzija, Matt Moore, Max Scherzer and Justin Verlander). With the increased velocity also came better movement, as both the vertical (by about two inches) and horizontal movement by about 2/3 of an inch) on the heater were increased. The extra life on the pitch corresponded to a four-fold increase in whiff rate, 16% in his last outing compared to just over 4% for the rest of the season. This is a massive difference, even though the change in velocity and movement seems to be relatively small. For what it’s worth, Texas Leaguers shows more dramatic differences in velocity and movement.
While the improved fastball is the main thing that jumps out at me, there are some noticeable differences in Nova’s curveball as well. The velocity of the pitched jumped about 2-3 MPH over his season average, though the vertical and horizontal break actually decreased. This could mean that he was throwing a tighter, sharper pitch, and consequently, opposing hitters whiffed at it about twice as often as they did earlier in the season.
The fastball and curve were Nova’s bread and butter in his most recent outing, as they have been throughout the season. He threw them about 83-percent of the time earlier in the season, and threw them nearly 90-percent of the time his last time out. I have no idea if Nova made a mechanical change during is time on the DL — it appears he did — or if the extra rest has simply given his raw stuff a little boost. Regardless, the extra hop on Nova’s fastball and tighter curveball seemed to be very effective in the small sample size of one outing. At this point in the season, Nova is auditioning just to earn a spot on the postseason roster. Unless he is absolutely lights out and another Yankee starter suffers an injury or setback, it is hard to picture Nova earning a spot in the playoff rotation. Regardless, if he continues to show his improved fastball velocity and more effective curveball, the Yankees could have a tough decision on their hands.
Hello readers, I’d like to thank everyone for the warm reception. It is truly an honor and a privilege to write for such a passionate, dedicated group of fans, on a blog that I have been reading since its inception (not to mention reading Mike, Ben, and Joe prior to that). It’s also fantastic to be reunited with my former partners-in-crime Moshe, Larry, Matt, and (briefly) Stephen. I look forward to getting to share my thoughts on my beloved Yankees, and will likely write on a wide variety of topics. My goal while writing here is not only to produce quality content, but also to interact with the RAB commentariat, so feel free to leave comments on this or any other piece I write here. I can’t promise I will get to reply to every one (other commenters can likely answer certain questions better than I could), but I will try to get to as many as I can. Also, feel free to hit me up on Twitter (@Eric_J_S) where I talk baseball, and a variety of other topics. And away we go…
At this point the wonderful news of Andy Pettitte‘s return to the fold has already been covered to death, and so there’s no need to rehash all of the details here. As a Yankee fan I’m thrilled, and as a statistical analyst I’m equally thrilled (I did something of an ode to Andy a little over a year ago, so be sure to have a look at that). Andy has been a pillar of consistency throughout his career. To wit:
1995-2003: 3.94 ERA (86 ERA-)/3.73 FIP (83 FIP-)/3.41 xFIP (77 xFIP-), 6.4 K/9, 2.9 BB/9, 0.7 HR/9, 49.3% GB%
2004-2006: 3.38 ERA (77 ERA-)/3.58 FIP (81 FIP-)/3.41 xFIP (77 xFIP-), 7.4 K/9, 2.5 BB/9, 0.9 HR/9, 50.4% GB%
2007-2010: 4.08 ERA (92 ERA-)/3.89 FIP (88 FIP-)/4.05 xFIP (93 xFIP-), 6.6 K/9, 2.9 BB/9, 0.8 HR/9, 46.9% GB%
However, today I’m primarily concerned with reviewing Pettitte’s stuff, and thankfully with Brooks’ excessively robust and reclassified new PITCHf/x database, we can have a more advanced look at what Pettitte did during the last few seasons of his career than ever before. The following table I’ve compiled details takes a look at each of Pettitte’s five pitches during the last three years he was active across a variety of categories. PitchIQ is ostensibly the equivalent of OPS+/ERA+; 100 is league average, while anything above is above-average and below is below-average. This is outstanding, as it gives us an idea of how well or poorly Andy’s pitches fared in comparison to his peers.
While he’s never had blow-you-away stuff, Andy’s been an incredibly successful Major Leaguer (and perhaps borderline Hall-of-Famer) due in part to his ability to hit his spots and change his speeds with a variety of secondary pitches that play off his 90mph fastball. According to Lucas’ and Harry’s reclassified PITCHf/x data, during his last three seasons in pinstripes Pettitte threw a fastball, sinker, cutter, curveball and changeup. Interestingly, ESPN’s Stats & Info blog put up a post last week detailing how one of the keys to Pettitte’s success in 2010 was his slider; however, according to this data Pettitte throws no such thing. I don’t know what data ESPN is being supplied with, but I’m inclined to go with the guys who manually reclassified more than 3 million pitches.
Per our data, Andy’s bread-and-butter — at least on a whiff/swing basis — has been his cutter, with a whiff/swing% an impressive 37% better than league average in 2010. None of his other pitches generated an above-average percentage of whiffs/swing. Part of the reason Andy’s able to get away with not having overpowering stuff is that his sinker and changeup each got him ground balls more than 50% of the time in 2010.
The one aspect of the PitchIQ Scores I’m still trying to get a firm read on is how to interpret them when it comes to LD/BIP and FB/BIP. I have an e-mail into Dan Brooks about this, and I’m pretty sure that we need to be looking at the PitchIQ Scores for these two categories as if they were “minus stats,” (i.e., below 100 is above-average and vice versa), given that general baseball convention holds that lower flyball and line drive percentages are thought of as a good thing. If my interpretation is indeed correct, both Pettitte’s curve and cutter have helped him limit the percentage of line drives, although the cutter is his only pitch that yields a FB% higher than 30%.
I also compiled Pettitte’s platoon splits from 2007-2010, although I won’t make your eyes glaze over by also posting a JPEG of that chart; feel free to download it here if you’d like. The gist of it is, Pettitte, as one would expect, handles righties and lefties with equal aplomb, although he’s really really good against same-side batters, with a PitchIQ whiff/swing of 148 on his cutter against lefties. For comparison’s sake, Jon Lester’s cutter against lefties during the same time period was 16% above average; while Cliff Lee’s is, rather surprisingly, 4% below league average. That doesn’t seem like it could be right, although then again for as good as Lee is I guess he’s always been a bit more about generating weak contact than outright overpowering hitters with strikeouts (though it’s not as if a 7.6 K/9 since 2007 is anything to sneeze at).
Conclusion and projections
So how well will Andy fare? Clearly if he can come close to throwing the way he’d been throwing during 2010, the Yankees will be adding a bona fide #2/#3 lefty starter at some point in May, which is just awesome to think about. Of course, only Andy knows how his soon-to-be 40-year-old body will react to returning to the grind of retiring Major League hitters and whether he still has the craftiness he’ll need to succeed.
Mike covered Pettitte’s ZiPS projection earlier this week, which sees a 4.45 ERA and 1.5 WAR for Pettitte in 125.1 IP. Marcel has Pettitte at 73 innings of non-adjusted 4.07 ERA/4.06 FIP ball; while SG’s baseline forecast (which is park- and league-adjusted) calls for 127 innings of 4.01 ERA/4.00 FIP ball. The 65% CAIRO forecast is even sunnier, with 140 innings of 3.69 ERA/3.64 FIP ball. All things considered, those are some pretty robust projections for an older player who skipped an entire season of work, and if he’s able to approximate some sort of amalgam of those numbers the Yankees will be in very good shape.
A couple of Fridays ago a bomb was dropped on the analytical baseball community. However, in this case, it was perhaps the greatest bomb ever deployed. You see my friends, Dan Brooks of the renowned Brooks Baseball announced with zero fanfare that Brooks — a terrific asset as far as individual game data goes, but lagging behind TexasLeaguers.com and JoeLefkowitz.com in multi-seasonal data — would not only now be carrying Player Cards featuring seasonal data, but that the PITCHf/x data dating back to 2007 (the first year data became available) had been manually reclassified by PITCHf/x gods Lucas Apostoleris and Harry Pavlidis.
That’s right; somehow, someway, Lucas and Harry sifted through three-and-a-half-million pitches worth of PITCHf/x data, so that amateur analysts like myself would have the most accurate data possible to play with.
Why is this important? Well, for starters, pretty much any time I’ve talked about Ivan Nova over the last six months, it came with the caveat that we knew his second-half success was due in part to increased deployment of his slider, but I didn’t have the data to back this assertion up, as the PITCHf/x system stubbornly insisted that Nova only threw a slider 3.9% of the time. Now we know the truth.
Check out the following table, showing Nova’s non-reclassified 2011 PITCHf/x data, against Lucas and Harry’s reclassified 2011 PITCHf/x data:
The four-seam classification was pretty much on the money, as was the curveball, but the rest of Nova’s repertoire was pretty butchered by PITCHf/x. As you can see, Nova actually threw his slider 13% of the time instead of 3.9%, while the reclassification also determined that Nova threw a sinker, not a two-seamer. He also threw about half as many changeups as the un-reclassified data said he did, and he doesn’t actually have a cutter at all.
However, Lucas and Harry could’ve called it a project and we would’ve been plenty happy simply having accurate PITCHf/x data. But no, they decided to go even further, providing pitch and sabermetric outcome breakdowns by pitch type, and while some of these categories have been available at T-Leaguers and Lefkowitz, never before has all of this data been available in one place. In particular, the Whiff/Swing% on an individual pitch level is simply astounding, and something that’s never been freely available. Check out the remainder of Nova’s 2011 stats:
Now, we had a pretty good idea that Nova’s new-and-improved slider was nasty, but I don’t think anyone realized it was 43.1% Whiff-per-Swings-Taken nasty! As a frame of reference, CC Sabathia, who boasts one of the top sliders in the game, recorded a Whiff/Swing of 40.9% last season (though in fairness, he also threw it 27% of the time).
In the aftermath of this insane treasure trove of new data, I couldn’t help but wonder whether they’d be adding league average data (helpful as an additional reference point), and also if we could expect to have manually reclassified data for the upcoming 2012 season, as it’d be quite helpful to have the full spectrum of accurate data when looking at a given pitcher’s offerings across multiple seasons. Incredibly, both Lucas and Harry confirmed via e-mail that they do indeed plan to reclassify pitches on an ongoing basis throughout the season.
This is probably one of the most important sabermetric projects undertaken in the last 10 years. It’s incredible that not only have they devoted their time and energy into delivering a product any of us can access free of charge, but that they’ve also committed to maintaining an accurate set of data on a go-forward basis is just mind-blowingly awesome.
Several of you asked for a bullpen version of the “best pitches in the rotation” post, and so here you go. Instead of just the 2011 season I’ve gone back and corralled the last two seasons worth of data for this post. The columns headed by “w” and “w/100″ are the pitch type’s linear weights (representing the total runs that a pitcher has saved using that pitch) and linear weights per 100 pitches (the amount of runs that pitcher saved with that pitch type for every 100 thrown), which provide some level of insight into a pitch’s relative level of effectiveness but should not be analyzed in isolation, as they are subject to the whims of both sequencing and BABIP. I’ve ranked the hurlers by their respective Whiff rates, as the ability to generate a swing-and-miss is probably the most transparent indication of pure stuff.
(Note: This post was researched and written prior to the release of the reclassified PITCHf/x data at Brooks Baseball — which I’ll be chiming in on next week — and the numbers are from TexasLeaguers.com and Fangraphs. Given that relievers typically have less variation in their repertoires than starters, I feel comfortable that the data presented below is mostly accurate.)
Rafael Soriano‘s generated the highest whiff percentage on the four-seamer out of the six primary members of the Yankee bullpen, though that is probably partially propped up by his excellent 2010. As far as pitch type linear weights go, David Robertson‘s four-seamer has been the most effective at 12 runs above average, while Cory Wade‘s was most effective on a per-100-pitch basis, at 2.17 runs above average.
Without looking at the numbers I’d have assumed that Mariano Rivera would easily lead in cutter Whiff%, but he actually lags both Soriano and D-Rob. Of course, having even an 8.1% whiff rate on a pitch you throw 86% of the time is still absurd.
For all the crap Boone Logan gets, his slider’s actually pretty outstanding, generating a whiff nearly one out of every four times he throws it. Joba Chamberlain also has a big-boy slider, though at times (cough cough full count cough) he’s fallen a bit too in love with it, occasionally making it painfully predictable.
David Robertson has the best curveball in the ‘pen by a pretty substantial margin, though Cory Wade’s isn’t terrible. Joba’s had a decent amount of success with his curve though he throws it pretty infrequently.
It’s Cory Wade by a landslide here, though he wins by default as no one else in the ‘pen really throws a changeup. It hasn’t been an outstanding pitch by linear weights, but it was his bread-and-butter in a terrific season for the Yankees in 2011.
Inspired by the excellent Red Sox blog Over the Monster, today I’m going to take a look at which Yankees starting pitchers throws the “best” pitch among each pitch category. As there are a variety of factors involved in determining a given pitch’s overall effectiveness, “best” in this instance is going to be subjective. In the interest of simplicity, I’m ranking the hurlers by their respective Whiff rates, as the ability to generate a swing-and-miss is probably the most transparent indication of pure stuff.
All of the data in the tables you’ll see below is from the 2011 season, and should be mostly self-explanatory. I’ll be the first to admit that a one-year sample is less-than-ideal, but I tried to run a three-year search and TexasLeaguers.com didn’t take to that request too kindly. The columns headed by “w” and “w/100″ are the pitch type’s linear weights (representing the total runs that a pitcher has saved using that pitch) and linear weights per 100 pitches (the amount of runs that pitcher saved with their fastball over the course of 100 fastballs thrown), which provide some level of insight into a pitch’s relative level of effectiveness but should not be analyzed in isolation, as they are subject to the whims of both sequencing and BABIP.
And right off the bat we have a prime example of the problems one can encounter with pitch type linear weights. If you sorted this table by wFF, Phil Hughes would come out on top. How on earth is that possible, you are likely asking yourself. I’m not entirely sure myself, as I don’t think anyone that saw Hughes pitch last year thought much of his fastball. However, he did get some people out, and presumably the vast majority of those outs came via his four-seamer, because, as you’ll see later on in this post, everything else he threw last season was pretty awful, at least by pitch type linear weights. Lending further credence to this notion is the fact that Hughes yielded a .282 BABIP on ground balls on his heater, compared to a .360 BABIP on ground balls on the curve, .444 on the cutter and .556 on his changeup.
As far as Whiff% goes, it should be quite heartening to see that the Yankees’ two newest rotation acquisitions outperformed everyone else in the rotation by a rather substantial margin. While both will likely see a decrease in their Whiff rates with the move to the AL East, at least they’re starting from a high baseline.
We know Ivan Nova threw a slider more than 3.9% of the time last season and so this table is a bit misleading. However, the pitch did become one of the keys to his improved second-half performance, and so there may be a case to be made for Nova having one of the better sliders on the team. Of course, Michael Pineda and CC Sabathia might have something to say about that. In any event, the Yankees’ front four in the rotation all boast pretty big-time sliders; bad news for opposing lineups.
While Pineda probably threw some two-seamers last season, I’d surmise that some of his four-seamers may have been misclassified, as a 10.6% Whiff% rate on a two-seamer/sinker is pretty damn high when you consider league average is 5.0%-5.4%. Not to mention the fact that the player with the best wFT/100 in MLB last season (Doug Fister), had a 5.4% Whiff% on his two-seamer. Sabathia probably has the best sinker on the team, although Kuroda is in that conversation as well if he can get his GB% back above 45%.
It should surprise no one that Sweaty Freddy had the best changeup on the team given his slow-slower-slowest approach, although Sabathia’s is also pretty great. No one else in the rotation has a particularly effective one, although Burnett’s did generate a slightly above-average Whiff% last year. Surprisingly, despite a rather diverse arsenal, Hiroki Kuroda is the only starter on the team that doesn’t throw a change at all. However, in his case he presumably partially makes up for it with his splitter, which can function like a hard change.
No surprises here; Burnett’s curve is the only thing keeping him away from the glue factory, but as everyone knows you can’t get very far with one working pitch. Nova’s curve is probably best described as a work-in-progress; while there were times in the second half that Phil Hughes looked like he was employing a harder (and more effective) curve and other times where his curve looked terrible. Stop me if you’ve heard the one about Hughes needing to improve his curveball to become an effective Major League starter.
The splitter is a fun pitch that Yankee fans don’t get to see too often, and this coming season we may have two members of the rotation featuring one (albeit in very different forms). Prior to Freddy Garcia, the last Yankee starter I can think of off the top of my head that threw one is Roger Clemens (Ed. Note: Jose Contreras threw a forkball, which is kinda like a splitter but slower). Per linear weights, neither Freddy nor Kuroda fared all that well with their splitters last season, but they still generated plenty of whiffs with the pitch.
So who boasts the best pitch in the Yankee rotation? Probably either Sabathia, with his heater or slider, or Pineda and his heater. I certainly wouldn’t argue against any of those three.
There has been no lack of Michael Pineda PITCHf/x analysis in the aftermath of the Big Trade, and if you haven’t already done so, be sure to check out Lucas Apostoleris here, Whelk at DRays Bay here and our pal Matt Imbrogno here. With these fine fellows having already done some of the legwork I was planning on doing, I thought I’d shift my focus to a compelling comp:
Feel free to guess in the comments, or find out what we’re looking at after the jump.
During the past year, one of the most frustrating aspects of conducting advanced baseball analysis has been the widening gulf between the reliability of the pitching data supplied by Baseball Info Solutions (and which FanGraphs uses) and Sportvision’s PITCHf/x, the latter of which is near-universally acknowledged as the superior data set (though that’s certainly not to say PITCHf/x is without its own flaws). To the delight of baseball nerds like me, FanGraphs recently went a long way toward rectifying this situation, adding PITCHf/x data not only to the individual player pages, but even more importantly, to the sortable leaderboards, enabling us to make comparisons that were previously impossible unless one wanted to input every individual name into TexasLeaguers.com manually and compile the data themselves in as painstaking a manner as possible.
In honor of the newfound ability to see where our favorite pitchers ranked in relation to their peers, I’ve taken a first pass through last year’s data to see where the members of the Yankees’ starting staff ranked among the 42 qualified AL starters across several categories. Neither Freddy Garcia nor Phil Hughes made the innings cut, which is why they’re not present. I also didn’t evaluate horizontal movement (pfx_x) or vertical movement (pfx_z), as you can’t really rank H-break and V-break in descending/ascending order (though you can try), at least not without first binning by pitching arm, which FanGraphs does not have the ability to do.
It may surprise you to learn that Ivan Nova threw the 7th-most four-seamers in the American League last season, although despite the fact that it’s frequently been derided as a lesser pitch for Nova, he also wound up having the most relative success with it on the Yankee pitching staff, with a 0.46 FF/C, good for 12th-best in the league. It will not surprise you that A.J. Burnett had by far the worst four-seamer in the league, at -2.21 per 100 thrown.
Here’s another interesting Nova nugget: despite the fact that PITCHf/x only has him having thrown a slider 3.7% of the time — a percentage that appears to be incorrect — he had tremendous success with it (which is, of course, something we already knew), putting up a 0.90 wSL/C, good for 11th-best in the AL and second on the team after CC Sabathia. As I and many other shave noted throughout this offseason, Nova’s slider will likely be the deciding factor behind whether he can continue to pitch as effectively as he did in 2011.
CC Sabathia’s sinker was one of the best in the game, ranking second-fastest (93mph), most valuable on an overall basis (2.5 runs above average) and second-most valuable on a per-100 basis (0.41). The only thing that worked for A.J. Burnett last season was his curve, which was worth 10.7 runs above average. And Bartolo Colon, as we saw all season, had one of the best two-seamers in the business, worth 4.7 runs above average, good for 7th-best in the AL.