Examining what went wrong for Rangers, right for Giants

With the final game of the 2010 season wrapped up, it’s easy for fans to be caught up in remembering a few highlight moments. World Series games tend to live on by showing one brief moment. Usually, the more dramatic, the longer the series stays with us.

So much about the game of baseball is built on repetition; everything seems cyclical as it builds from a high to a ho-hum point only to be reset again. It’s only natural that we have a need to condense everything into one exciting swing or unexpected error. (Although, one can find it curious that Johnny Damon outrunning Pedro Feliz in Game Four of last year’s World Series stands as a high point, but maybe that was due to a lack of surprises.)

But in keeping with tradition, it must be said that Edgar Renteria’s seventh-inning home run in Game Five was pretty dramatic and should go down as the defining moment.

It also, again, reminded us of Cliff Lee’s mortality during this World Series.

Cliff Lee: What went wrong?

Besides the obvious answer—what beat the Rangers was a rush of excellent Giants pitching—one would be foolish to ignore San Francisco’s timely offense as the Giants took advantage of mistakes made by Cliff Lee.

To see Lee lose both of his starts came as a complete surprise. Looking over his line during the series: 11.2 IP, 14 H, 1 BB, 13 K, everything seems to be Cliff Lee-esque except for those 14 hits allowed. During Game One it became obvious Lee had trouble getting his curveball over for strikes and his change-up seemed to be worthless, making him a two-pitch starter (fastball/cutter). In that first game, Michael Young’s error in the third inning and Lee’s shaky command that resulted in a Andres Torres hit-by-pitch on an 0-2 fastball began a bad chain of events.

Another bad spot for Lee was his pitch selection against Pat Burrell in the fifth inning (especially that unnecessary curve he threw on a 2-2 count; the delivery was off and the pitch was so far out of the zone that no one would have swung). Lee had no business walking Burrell and with two outs that was an important at-bat.

Lee also missed his spots as many fastballs drifted over the fat part of the plate. Joe Pawlikowski at Fangraphs recently ran a few charts showing the location of Lee’s pitches and compared his Game One start to his last outing against the Yankees in the ALCS.

In Game 5, Lee seemed to be in much better command of all of his pitches… except for one.

How did the Rangers’ pitching staff attack the Giants’ one through five hitters?

With apologies to Giants heroes like Renteria and Juan Uribe, it seemed best to look at the overall strategy the Rangers may have had against the top of the Giants’ lineup.

Torres saw a total of 90 pitches in 23 plate appearances, which equals about 3.9 pitches per plate appearance. As a switch-hitter, Torres has seen more right-handed pitching in his caree,r but is essentially the same hitter on both sides of the plate (.251/.320/.432 vs LHP in 354 PA; .250/.320/.431 vs. RHP in 671 PA).

Overall, Torres saw 38.9 percent four-seam fastballs, while curveballs and two-seamers came in at exactly 16.7 percent each. With the four-seamers, Torres whiffed only 2.9 percent of the time while he was able to put that pitch into play at a 25.7 percent rate. The pitches that gave Torres the most trouble were curveballs and cut fastballs. In Game One, Torres was thrown a total of four cut fastballs and swung at three of them (most tailed down and away). He was able to get a hit from one. Torres wasn’t thrown a single curveball in Game One.

Freddy Sanchez is a career .324/.358/.453 vs. left-handed pitching in 997 plate appearances. In 22 plate appearances during the series he saw 82 pitches: 41 percent were four-seam fastballs, 18.3 percent sliders, 13.4 percent two-seamers. The pitches that gave him the most trouble: curveballs (8.5 percent thrown) and change-ups (7.3 percent).

Buster Posey had 21 plate appearances this series and hit .300/.333/.450. He saw 96 pitches. Four-seamers and sliders were thrown the most, with sliders giving him the most trouble by missing his bat 21.7 percent of the time.

A Hardball Times Update
Goodbye for now.

Cody Ross had problems putting curveballs into play all season, and judging by this chart, he seems very likely to swing at one below the strike zone.

image

After failing to convert any of the 10 thrown this series into play, Ross saw just 10.5 percent of that pitch overall. This series Ross saw a total of 95 pitches in 21 plate appearances and hit .235/.381/.471.

Aubrey Huff finished with a .294/.368/.588 in 19 PA. He saw only 54 pitches, but he made contact with most of them. Huff hit ground balls 35 percent of the time but his two home runs do elevate his numbers. Four-seam fastballs and curveballs were the pitches of choice; however, two seamers, cut fastballs and sliders accounted for 39 percent and Huff was able to put them into play at a high frequency.

Why did the three-four-five hitters for the Rangers struggle?

After a successful run in the ALCS, the Rangers’ Josh Hamilton, Vladimir Guerrero and Nelson Cruz hit a wall this World Series, combining for a batting line of .124/.156/.257.

In 56 combined plate appearances, they struck out 13 times, giving them a 23 percent overall strikeout rate. Over the season, Hamilton came in at 18.3 percent, Guerrero 10.1 percent and Cruz at 20.3 percent. Guerrero struck out in five of his 15 plate appearances and seemed uncomfortable throughout the series, but how do we account for Hamilton and Cruz?

Individually, Hamilton hit .100/.143/.250 with 20 AB, two hits, one home run, one walk and three strikeouts. Cruz came in at .200/.200/.450 with 20 AB, four hits, two doubles, one home run, no walks and five strikeouts.

Among the 76 pitches thrown to Hamilton, 12 were change-ups. According to the invaluable website texasleaguers.com, Hamilton had the most trouble with this pitch; he whiffed at a rate of 16.7 percent. On the other pitches, Hamilton was able to make contact. However the Giants were successful (for the most part) in inducing weak contact.

image

Looking at other pitches that Hamilton swung at during the series, we see that he chased quite a few sliders and change-ups down and away.

image

Among the 70 pitches Cruz saw, 31.4 percent were sliders while 20 percent were change-ups. In Game Five, Cruz was able to take one of Lincecum’s sliders over the left field fence. In terms of whiff rate, a lot of these sliders and change-ups did work to miss his bat at about 30 percent of the time.

Another factor that did Cruz in was his inability to handle many of the four-seam fastballs thrown down the middle of the plate. Looking at those 70 pitches Cruz saw, 42.9 percent qualified as four-seam fastballs. Below is a chart showing the four-seamers that Cruz swung at:

image

That’s quite a bit of white cowhide thrown down the middle; however, the above graph translates into this:

image

Looking at the managers:

What about the aspect many fans and commentators love, second-guessing lineups and bullpen moves. For the most part, Giants manager Bruce Bochy managed a good series. He wasn’t afraid to fiddle around with the lineup and use his bullpen match-ups to gain an advantage.

The problem with Ron Washington (besides the decision to start Guerrero in the outfield, which has been discussed ad nauseum) was his inability to get one of his better relievers in high-leverage situations. During this series, Alexi Ogando was used mostly in garbage situations, probably not the best use of a resource that was able to strike out six opposing batters in 3.2 innings of work (without walking anyone). I’m sure Ogando will have plenty more opportunities to prove himself.

In conclusion:

It’s staggering to think how patched-together both of these teams looked at the beginning of the season. By December 2009, Kevin Millwood and Scott Feldman looked like the only two pitchers guaranteed a rotation spot in Texas. Millwood would eventually be traded and Rich Harden and Colby Lewis would be brought in as bargain rotation-fillers. Both Harden and Lewis were seen as minor gambles with many expecting the former to pay off. The rise of Lewis along with the noble experiment to return C.J. Wilson to the starting rotation did pay off, and then they traded for one of the best pitchers in baseball, Lee.

Of course, San Francisco had pitchers like Tim Lincecum and Matt Cain to depend upon as well as a bullpen that was quietly effective in 2009. Yet, no one knew what to make of the Giants’ dreadful offense.

Pablo Sandoval was seen as a certified offensive cornerstone while last-minute free-agent acquisitions like Huff were brought in to help score runs. What happened? Sandoval slowly faded as his production was cut in half and Bengie Molina was traded to make room for a promising rookie backstop. Toward the end of the summer, the Giants pick up a few discards in Burrell and Ross and a 21-year-old rookie named Madison Bumgarner seemingly overcame his velocity woes and—we have a championship team.

Can it really be that easy?


7 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Rich Snyder
13 years ago

The last time the Giants won a World Series was 1954. That was when Willie Mays made the most improbable catch ever in a World Series to rob Vic Wertz of what would probably have been the game winning hit. That catch changed the whole course of that series and led to the Giants 4-0 sweep. The most memorable play of this series was the ball that bounced off the top of the wall off the bat of Ian Kinsler. If anything could be as improbable as May’s catch, this was it. How unbelievable that was!!!—and you didn’t even mention it!!!

DrBGiantsfan
13 years ago

I don’t think it’s a slap at Cliff Lee(or to he Phillies’ big 3 starters) to say that they just might have been overrated by a lot of observers in relation to the Giants pitching staff.  It wasn’t just that the Giants had been on one of the greatest pitching tears of all time since around September 1.  They were a pretty darn good pitching staff all season and in 2009 too. 

Tim LIncecum:  2 time Cy Young award winner and 3 time NL strikeout leader.  Why you wouldn’t think he could more than hold his own against Cliff Lee or Roy Halladay is a real head scratcher.

Matt Cain:  Perennially underrated, especially by sabermetric types who just can’t believe that it’s actually possible for ERA to outperform xFIP or whatever other calculated stat du jour you want to use.  Memo:  Matt Cain is a horse!

Jonathan Sanchez- Sanchez was one of the only Giants pitchers who underperformed in the postseason.  Down the stretch he was their best performer and won the NL West clincher against San Diego.  Had he been up to par, the pundits would have looked even worse!

Bumgarner- I’ll grant that this guy’s performance in Game 4 astonished even me.  Great looking young pitcher, but I’m not sure anyone could have seen THAT coming.

Brian Wilson- Best closer in baseball all year.  Turned it up a notch in the playoffs, but no real surprise here.

Bullpen- Once the Giants acquired Lopez and Ramirez, their bullpen, from top to bottom, was the best in baseball.  No surprise at all that they dominated in the postseason.  They even left a couple of guys off, Ray and Runzler, who would not have been the last man in any of the bullpens they faced!

What went wrong for the Rangers is they ran into the best pitching staff in baseball!

Graham
13 years ago

Thanks for the piece, but I must say—I’m very uncomfortable with leaning on PitchFX data for substantive analysis at this point.  It’s extremely exciting to think about all of the analytical possibilities the system presents, but I think the accuracy of PitchFX is greatly overestimated at this point. 

Consider the case of Tim Lincecum.  Timmy does not throw a four-seam fastball—or, if he does, he throws maybe a couple per game.  Everything else is a two-seamer – he’s confirmed this in interviews, and closeups of his grip would reveal the same thing.  Yet PitchFX claims that only 14.5% of his pitches this year were two-seamers (and it further claims that he didn’t throw any prior to this year).  And then there’s his slider—an awesome new pitch he developed in September, used to dominate the Braves in game one of the NLDS, and then more or less abandoned until game five of the World Series.  Watching his two starts against Philadelphia and his first start against Lee in the WS, it was pretty clear that Timmy didn’t really have his slider at all (which may have been due to the blister problem that recurred against the Braves).  But PitchFX had him throwing 26 sliders in one of the games against Philly (can’t remember which one). 

I’m willing to accept that Lincecum is one of those guys that just defies easy classification pretty much across the board.  I’m sure PitchFX is far more accurate with other pitchers than it is with The Freak.  But….I just don’t think it’s accurate enough to be used as a be all / end all of performance analysis.

Vince Caramela
13 years ago

Graham,

I agree with the inaccuracies of PitchFX and I must admit to being a slight newbie in assessing the data.

My goal was to look at both video (it’s funny how antiquated the word “video” sounds today) and PitchFX data in terms of what pitches were thrown and see where each certain batter succeeded and struggled as a whole.  It’s not bullet proof by any means but I was interested to see just how truly dominated a few Ranger hitters were.

I’ll admit the data esp. the fastballs Cruz saw around the middle of the plate was surprising (small sample size alert, be damned).

obsessivegiantscompulsive
13 years ago

First, Vince, nice job dissecting the PitchF/X data.  As you noted, it’s not the best data.  But it’s the best we got.

As others noted, the Giants pitching staff seems to defy analysis.  And they all made excellent points. 

The thing is, to me, while PitchF/X handles the plane of the strike zone well enough, it is not capable of detecting what experienced observers call “stuff”.  And apparently the Giants staff has a lot of stuff.

What sabermetrics has not done well with yet in terms of pitching is detecting the rare pitchers who have stuff, who can do things with their pitches that other pitchers can’t.  What happens is that when these pitchers then produce outlier data points, like Cain does with his BABIP (Zito too), sabers say that they are going to fall back to the mean, but then after 6-7-8 years of saying that, that don’t seem so true anymore. 

It is not just that the hitters struggled against the Giants and their pitchers’ selection of pitches, as you were trying to show in this nice article, but that the pitchers had other things going that helped them get batters out regularly, not just the Rangers 3-4-5 hitters.  That was a huge reason why the Giants led the majors in keeping runs allowed low.

obsessivegiantscompulsive
13 years ago

Rich, that was a pretty good sign that things would go the Giants way, but I was thinking that the 7 run explosion, while not as tidy as a catch or a near-HR ball, in game 2 was what did it for me.  It just seemed like the Giants could not be stopped, it lasted so long.

Rich Snyder
13 years ago

—The thing is, to me, while PitchF/X handles the plane of the strike zone well enough, it is not capable of detecting what experienced observers call “stuff”.  And apparently the Giants staff has a lot of stuff.—
I agree with you 100% obsessive—what I was trying to point out was that there were factors in this Giant win that can’t be measured by PitchF/X. I think the truth is NOBODY was going to beat the Giants this year. The Yanks???—forget it—they wouldn’t have done as well as the Rangers.