23 June 2009

Buster's Questionable "Olnalysis" of Rockies' Huston Street

ESPN's Buster Olney writes, at the beginning of a 2,800-word "blog post", that Rockies closer Huston Street has experienced dramatic improvement due to the smallest of changes:

...Bob Apodaca, the Colorado pitching coach, approached him and told him he wanted to make a rather major alteration. "We'd like you to move to a different part of the [pitching] rubber," Apodaca said.

Street didn't buy it. Not then, anyway. But after a mediocre spring training, and after he allowed four runs in his first four relief appearances in the regular season, Street had an open mind. He shifted from the left side to the right side of the rubber, and after he had done it for a short time and saw what the change did for him, he couldn't even imagine moving back to the left. "I don't know why things work out the way that they do," Street mused Sunday, "but they do."

Since those first four outings, Street has excelled, posting a 2.33* ERA, converting saves in 15 of 16 chances. He is 8-for-8 this month as the Rockies have made their push back from deep in the NL West standings to over .500, capped by Street's picking up the save Sunday, closing out the Pirates.

*Editor's note: Actually it was 2.22. I'm just sayin'.



Articles like this one always make me laugh, about how changing from one side of the pitching rubber to the other made someone a better pitcher, or how an offseason training regimen or starting to jog everyday or eat more granola or something made some former star a better player again. Usually it's just normal statistical fluctuation, but of course sportwriters can't write about that because

A) it's boring and

2) ZZZZZzzzzzzz......

So they talk to the pitcher, who probably has never taken a statistics class, or a physiology class or a physics class, for that matter, and he explains to them that this and that is the reason for his sudden success. This is much more interesting, or at least less sleep-inducing, than Chi-square distributions and bell curves and standard deviations and all that rot, and anyway, by next week nobody will remember what he said or what Buster wrote because they'll have moved on.

Olney's explanation for the improvement is as follows:

When throwing on the left side of the rubber, Street could throw strikes against left-handed batters by running a fastball over the outside corner, no matter how flat it was -- and the ball would have to travel a longer distance from his hand to the corner. But once Street was on the right-hand part of the rubber, it forced him to be more disciplined in his mechanics. He'd have to get on top of the ball properly to throw it for strikes to the outside corner to lefties, and inside to right-handers. If he didn't throw the ball correctly, it would drift off the plate.

"I've got more of the sinking action than the running action," Street said.


In case you're wondering, the difference in the distance to one side of the plate from one side of the pitching rubber or the other is, at most, about 0.4".

Four-tenths of one inch.

Or, as they say in France, "almost nothing".

It's a simple geometry problem, with a right triangle, 60.5 feet on one side and two feet (the width of the pitching rubber) on the other. The hypotenuse of that triangle is therefore 60.53 feet, or 60 feet, 6.4 inches. That 0.4" difference represents an increase of 0.7% compared to throwing from the other side of the rubber. No wonder Street's been so fatigued!

Street really was quite terrible in those first four games of the season, or at least in three of them. In mid-April, having pitched only four times in the team's first eight games, Street had an 0-1 record with one Save and a 13.50 ERA. Then he supposedly made this change, and over the next four games he was...

...still pretty lousy.

No Wins or Losses, but no Saves either and only one Hold, to go with a 5.79 ERA in those games, though he struck out six and walked none in 4.2 innings. The real improvement followed that. From April 26th to June 21st, he had 15 Saves, two wins and no losses, 27 K's and eight walks in 23.2 innings, to go with a sparkling 1.52 ERA.

The real improvement was not in the walks, as Olney's "analysis" (or shall we say, "Olnalysis"?) suggests it would be. In the first eight games of the season, Street issued only one walk in 7.1 innings. After that, Street's walk rate more than doubled, from 1.23 per nine innings to 3.08 per nine.

The real problem was that he allowed three homers and three doubles in those first seven innings and change, while in the 23.2 innings after that he allowed three extra base hits total (two homers and a double).

Technically, if he's getting more sink on the ball, this would be one result, and so maybe it is helping. But if so, it's because he's being more consistent with his mechanics, not because of which side of the rubber he stands on before he winds up to throw. He could have done this from either side of the rubber if he'd just been diligent about his mechanics.

Looking at this from the other perspective, if Street now has to force himself to "get on top of the ball" more because he can't throw a strike to the outside corner otherwise, shouldn't he now be susceptible to having the ball run back over the middle of the plate when he's trying to throw inside to lefties? If he gets lazy or fatigued and doesn't get the proper sink on the ball, now he's susceptible to allowing doubles and homers, rather than walks. And yet, just the opposite has happened, he's allowed fewer extra base hits, but more walks.

It still comes down to making sure his mechanics are maintained properly, which he can (theoretically) do from either side of the rubber. The real reason for his marked improvement, I think, is that the Rockies' level of competition has changed drastically from that first three weeks of the season. Street faced the Dodgers and Phillies, ranked #1 and #2 in the National League in Run scoring, in six of his first nine games, and his ERA took a beating for it.

Of the 24 games since, 12 have come against teams in the bottom half of the majors in run scoring, and among those, San Diego, Houston, and Seattle are three of the five worst teams in baseball at scoring runs. To me that makes a lot more sense than four tenths of an inch difference causing Huston Street to finally get his mechanics straightened out, and this after having been pretty darned successful in the first 247 games of his major league career.

Olney closes that portion of the blog post this way:
Street cited something that Troy Tulowitzki said recently -- that the Rockies are beginning to expect good things to happen. "You don't know whether success creates that mentality, or whether that mentality creates success," Street said. "I think it's a little bit of both."
Look, pal, that's just plain lazy.

When I was in college, I spent a lot of time studying the Bible with friends from my InterVarsity Christian Fellowship group, using something they called the "inductive method" of study, which essentially seeks to determine what the original author meant to convey to his or her original audience. In a group of 10 or 15 students or more, we would discuss one small section at a time, say, 10-15 verses of the Gospel of Mark, and inevitably, at least once per session, we would come to some kind of impasse.

One student would suggest that Mark was trying to say this, and another would suggest he was trying to say that, the interpretations being mutually exclusive, of course. And after a while, someone would suggest that, "Maybe it's a little bit of both!" This happened so often and so consistently that we coined a term for it: "The Relativistic Third Option." Sometimes, it might have been the best way to interpret something, as certain passages can be interpreted in several ways, none of which is inherently inconsistent with the rest of scripture.

But usually we were just being lazy. It's often hard work to figure out which of two mutually exclusive options is the correct one, and in this world of post-modern ideas and tolerance and relativism, people are too often satisfied with wuss-out explanations like, "It's a little bit of both."

Either your "winning mentality" put you in a position to help make good things happen, or you got lucky, started to win and then tried to think about what else you could do to help the team win more. There is no such thing as "a little bit of both" in this case.

Why doesn't anybody ever just say, "I don't have any idea"?

Stumble Upon Toolbar

12 June 2009

What's Wrong With Wang?

On the heels of yet another loss to the hated Red Sox, and entering a crucial series with the hated Mets, the New York Yankees have some 'splainin to do. Namely, they have a pretty tall order explaining the continued presence of one Chien Ming Wang in their starting rotation, given that he seems, statistically, at least, to be no closer to returning to the form that twice amassed 19 wins for them.

Those storied* times, the salad days of 2006 and 2007, seemed much farther than two years away as I watched Wang unravel yet again on Wednesday night against the Red Sox, allowing four runs, including six hits and three walks in less than three innings of work. Believe it or not, that start actually constituted an improvement for him, lowering his ERA ever so slightly from 14.46 to 14.34. Oh goody.

*Not steroid.

Much of the talk about Wang has centered around his heavy sinking fastball, and the idea that he somehow needs to either get his velocity back or his mechanics straightened out so this pitch can again be the grounder-inducing menace that so frustrated the American League in 2006 and 2007. The trouble with this, however, is that he's got all his velocity back, averaging 91-92 mph and often hitting 95 mph on the radar gun with his 4-seam fastball.

The problem isn't his fastball. It's the lack of anything else.



According to FanGraphs.com, between 2005 and 2008, Wang threw his fastballs (including the 4-seam and the sinker) 76.5% of the time, with a velocity averaging between 91.8 and 93.1 mph. This year, while his velocity is just as good (91.7 mph average), he's throwing one or the other of his fastballs 84.7% of the time, a significant difference from his usual modus operandi.

Moreover, most of the extra fastballs are coming at the expense of his slider. Previously he threw the slider about 15% of the time on average, ranging from 12.9% in 2005 to 17.1% last season. This year he's used it only 11% of the time, meaning that there are about half a dozen pitches or more per start that used to be sliders but are now fastballs. And of course, Wang has been getting tattooed all year, so we have to wonder if this is somehow related, right?

This in itself may not be significant, but it got me to wondering why Wang (or his catchers) would be so reluctant to use the slider this season, when he seemed to use it more often and with greater success in the past. Looking at the MLB Gameday data for his last two starts and comparing them to a good start from last year gave me a possible answer:

The slider isn't, well, sliding.

Last year, Wang pitched a complete game, 1-run 2-hitter against the Red Sox in April, no small accomplishment given that those Red Sox finished second in the AL in Runs Scored in 2008 and eventually won the Wild Card. During that game he threw 93 pitches, and according to MLB Gameday, 20 of them were sliders. In addition to the speed of each pitch at release, Gameday provides two measures of the pitch movement, "Break" and "Pitch F/X".

According to MLB.com, Break is

"a measurement of the greatest distance between the trajectory of the pitch at any point between the release point and the front of home plate, and the straight line path from the release point and the front of home plate."
That is, I think, the Break is a measure of the difference between where the ball actually ends up and where the batter might think it would end up if gravity and/or spin were not factors.

By contrast, Pitch F/X "is the measurement of the distance between the location of the actual pitch thrown over the plate, and the calculated location of a ball thrown by the pitcher in the same way, with no spin..."

That leaves the method of that calculation as an open question, of course, but assuming that these guys have some idea what they're doing, this seems the more relevant number for our purposes. The batter will assume that the pitch is going to "break" down, if only due to gravity. Even Daniel Bard's fastball, clocked between 98 and 100 mph on Tuesday night, showed a "Break" of three to five inches.

For the record, Wang's fastball/sinker seems largely unchanged, showing a Break of 5-8" and a Pitch F/X of 10-14 " in that complete game against the Red Sox last April. This year, in his most recent start, the fastball was just as fast, showed a typical Break of 5-8" and a typical PFX of 10-13 inches.

But Wang's slider? Last year its PFX averaged 4.05" (with a range of 2-7), but in his two most recent starts, it's averaged just 2.3 inches, almost half of what it once was, and often only zero or one inch. No wonder Posada doesn't want to call for the slider. It isn't fooling anyone because it doesn't do anything, having almost the same trajectory as a pitch thrown without any spin at all, according to MLB Gameday and Pitch F/X. For batters, this is a win-win situation. Either they swing at the occasional slider, which has hardly any spin on it, or they wait on the fastball, which is Wang's only other quality pitch.

The slider is a subtle pitch, so much so that Pitch F/X often has trouble distinguishing it from a cut-fastball and/or even a changeup. It's thrown with a sideways spin that causes it to drift laterally, across the strikzone, in the opposite direction of the pitcher's throwing arm. Because it gets no assist from gravity, the slider doesn't break as much as a curveball does, but it does ehough that it ends up several inches from where you'd expect, either out on the end of the bat or in on your hands, depending on what kind of hitter you are.

The best sliders in baseball (Carlos Marmol, Jonathan Papelbon, Francisco Rodriguez, Chad Billingsley) usually break only 5 to 8 inches or so, but there are plenty of pitchers whose sliders sit in the 4-inch range. But two inches (and often one or none) simply isn't enough to fool major league hitters, who are so well trained that they make mid-swing adjustments in hundredths of a second, and so strong that they can hit a ball out of the park while breaking the bat.

Whether this is a physical problem for Wang or not, I don't know, but I doubt it. Perhaps his foot still hurts, and he's somehow favoring it, throwing the slider less often because it bothers him physically. This is unlikely, as any difference in his delivery due to throwing the slider would be tantamount to "tipping" his pitches, and batters would have picked up on it long ago.

Perhaps he's still rusty, having missed some time due to the foot injury, and doesn't yet have the "feel" for the slider. This seems very likely to me, as pitchers often talk about how difficult it is to get a feel for their sinker, slider or cutter, and how much practice this takes. Having missed more than half of last year, and having thrown only about 21 innings so far this year in the majors (plus 13 scoreless innings in AAA) Wang's a little behind on his usual regimen.

Maybe this means that with a bit more practice, he'll get that feel for the slider - and with it, his confidence in the pitch - very soon, enabling him to keep hitters a bit more honest and not so frequently serve them the heater they already expect. I just wish the "practice" didn't have to come in Yankee Stadium, and against the damn Mets.

Admittedly, I have not looked over all the available data. Wang has thrown hundreds of pitches this year and thousands in his career, and I simply don't have the time to examine every one, but this hopefully gives us an idea of where to look for answers the next time Wang takes the mound.

UPDATE: My apologies for the false information, but it turns out that Wnag's next start is scheduled for Wednesday, against the Nationals, not Sunday. That was Burnett's regular spot, and he pitched well for once. Wang should have been scheduled for Today, Monday, which is an off day and (it would seem) a perfect opportunity to skip Wang in the rotation.

Instead, for some reason Joe Girardi has chosen to start him on six days' rest and push the rest of the rotation (other than Sabathia) back a day against the Washington Nationals. Maybe Joe agrees with my assessment that Wng just needs more work, and figures that if there's any team he can beat, it's the Nats (16-45).

Stumble Upon Toolbar

10 June 2009

Yankees Need to Step It Up Against Red Sox

Well, it's been a whole month, so I guess we were due to resume the Yankees-Red Sox rivalry. As you'll no doubt recall, when last we met our heroes - or, at least the last time the Red Sox met them - they were a sad sack of a team, hovering about the .500 mark. Xavier Nady, Brian Bruney and Alex Rodriguez were all on the Disabled List, and the ineffective Chien Ming Wang would soon join them, as would Jorge Posada.

To make things worse, none of their big name free agent acquisitions seemed to be panning out. Mark Teixiera was hovering around the Mendoza Line, while CC Sabathia and AJ Burnett both had ERAs around 5.00. Additionally, the patchwork bullpen, put together mostly on the cheap from the Yankee farm system, had failed them miserably, allowing a composite ERA of almost 8.00 in their first five games against Boston, though they'd done mostly respectable work against the rest of the league.

That team lost all five of its early season contests against the Red Sox, with the starting pitching largely to blame, as they got only one Quality Start in those five games. That performance - Andy Pettitte's 6-inning, 4-run (3 earned) outing on April 26th - met those requirements in name only, and anyway the Yankees could do nothing with Justin masterson that day.

The third inning of that game, when the Yankees had a 1-0 lead, marked the last time the Yankees have led the Red Sox in their season series, including last night's 7-0 loss. The Sawx scored a run in the bottom of that inning and then three more in the 5th (including Jacoby Ellsbury's infamous steal of home plate) and have had no reason to look back since. Indeed, the Yankees have given them no reason to glance over their shoulder.

So, with the season series resuming Tuesday night, the Yankees had good reason to be in high spirits. They had Alex Rodriguez and Jorge Posada back, and Chien Ming Wang slated to start the second game of the series. Mark Teixiera had found his stroke since A-Rod came off the DL, and several Yankees (Damon, Jeter, Cano, Cabrera) were hitting around .300, many of them with power. They sat atop the AL East, with the best record in the league coming into the game....

...and then they lost miserably.

A.J. Burnett, the big name pitcher who came up so very small against the the Yankees' biggest rival in April, managed to lower the bar for himself even more last night, allowing five runs (three earned) without escaping the third inning. Granted, he's faced the toughest slate of hitters in MLB this year to date, but still, more is expected of a man who's earning more money than the gross domestic products of some small island nations.

Burnett's fastball was plenty fast, usually in the 95-96 mph range, but he seemed to have little idea (or interest) in where it would end up. Working quickly, as if to get it over with rather than to get batters out, Burnett threw 84 pitches - less than half of them for strikes - with his curve proving to be especially erratic. He threw only five of 16 curveballs for strikes, and one of those was a single by Kevin Youkilis anyway.

With the curve clearly not working, the Red Sox could just sit on the fastball and wait for him to throw a rare strike. And when he did throw strikes, they were belt high, out over the plate, which is why the struggling David Ortiz was able to hit one of them 420 feet into the stands in center field, only his third homer of the season.

What's more, it seems from looking at the pitches on MLB Gameday that Burnett all but refused to pitch anyone inside, perhaps out of fear of another suspension for not actually hitting someone. JD Drew's 2-run double in the second inning, on an 0-2 pitch, was hit off a 96 mph fastball that was supposed to be inside (based on where Posada had set up behind the plate) but ended up on the outside corner. Nobody's fastball is good enough to leave it out there and expect to get batters out.

The relief, such as it was, did better but was hardly impressive. Neither Brett Tomko nor Jose Veras threw even half of their pitches for strikes, and though David Robertson was OK, and the group as a whole allowed only two more runs in over five innings of work, the Yankees' hitters couldn't touch Boston's pitching, getting only two hits off them all night.

Tonight Wang takes the hill against Tim Wakefield, who's 7-3 this year but is only 10-17 with a 5.03 ERA against the Yankees in his career, so those who do baseball handicapping would presumably say that the game is up for grabs. Wang, for his part, has been dreadful as a starter this year, 0-3 with a 23.62 ERA, but perhaps he's on the mend and can give the Yankees six solid innings for a change. If not Phil Hughes will be available out of the bullpen, but by then it may be too late.

Stumble Upon Toolbar