NLDS: A WPA Review

As Game 3 of the NLCS rages on, I wanted to take a look back on that roller-coaster division series and analyze it using the best stats I know that can numerically illustrate the up-and-down nature of each game. Therefore, I looked at both WPA (win percentage added) and LI (Leverage Index) to rate the players and the moments of each game.

Quickly, WPA uses data from years and years of game to calculate the winning probability of a team before and after a certain event, and then finds the difference. For example, when Rick Ankiel hit that titanic homerun off of Ramon Ramirez in the 11^th inning of Game 2, the Braves went from a 44% chance of winning to an 82% chance, so Ankiel is credited with .380 win percentage added, and vice versa for Ramirez. In this way, batters or pitchers who come through in a big spot are rewarded for their “clutchiness” (or not, if they flop).

Leverage Index is a measurement of the “pressure” of a given situation. 1.0 is average, so a score over one is above-average pressure, and below one is below average. For my purposes, this stat is particularly useful for illustrating exactly how important a specific play is in the course of the game.

Alright, let’s get to the games:

Game One: Giants win 1-0, Giants lead series 1-0

Player of the Game: Tim Lincecum, +0.754 WPA

Well, that should have been obvious. Lincecum went for a complete game, two-hitter with 14 strikeouts. Plenty has already been written about this game already, so I’ll just add this: I have an uncle who is about as hard-core of a Phillies fan as there is. He texted my dad after Lincecum’s game and told him that Timmy’s performance impressed him much more than Roy Halladay’s no-hitter the night before.

Goat of the Game: Freddy Sanchez, -0.166 WPA

Sanchez went 0-4, grounded into a double play in the first, grounded to the pitcher in the 3^rd that got Cody Ross tagged out in a rundown between third base and home, ground out to second in the 5^th, and ground out to first to lead off the 8^th. Not a whole lot to say about this game for Freddy, except that he saw a grand total of 8 pitches in four at-bats. Not a good day for him.

Play of the Game: Cody Ross singles to left, Posey scores (4^th inning)

This play would amount to the only run of the game, and was worth .13 WPA all by itself. This play, of course, followed the controversial call at second base on a Posey steal (he hadn’t stolen a base all season). It was obvious upon replay that Posey was out by approximately a foot at second, and of course that mars the game. In conjunction with the other missed calls in the Division Series, I expect Major League Baseball to take another look at replay this winter.

That being said, I think Cody Ross played himself into a contract this series, and, especially, this postseason. He was, by WPA, the Giants’ second most important player (after Aubrey Huff) and he, of course, hit those three (so far) home runs in the NLCS. In fact, as I type this, he just hit a single to left with two outs to score Renteria, giving the Giants a 1-0 lead in Game 3. I fully expect Bochy to offer arbitration to Ross this winter, and to have him on as a 4^th outfielder for around $3-4 million. He’s certainly much better than Jose Guillen, can run the basepaths and play outfield (+9 DRS, +3.4 UZR/150).

Game 2: Braves win 5-4 in 11 innings, series tied 1-1

Player of the Game: Pat Burrell, +0.242 WPA

Speaking of players that deserve another contract, how about Pat Burrell? The Bay Area native put up 2.5 fWAR this season after signing a minor-league deal May 29^th. Want to hear the most shocking stat of the day? Guess how much value Burrell accumulated with his glove. Has to be a negative number right? I mean, he’s a self-described “water buffalo” in left field.

Well, you’d be wrong. According to FanGraphs, Burrell has a DRS score of +6 and an UZR score of +4.9. The things you learn.

Nearly all of his value this game derives from his first inning home run against Tommy Hanson, plating three runs and setting the Giants off on the right foot. Yeah, how’d that work out?

Goat of the Game: Buster Posey, -0.353 WPA

Seems wrong, right? I mean, the goat of the game has to be Sergio Romo, or Brian Wilson, or Ramon Ramirez, right? According to WPA, the second biggest play of the game was Posey’s double-play groundout to Troy Glaus (!!), which cost the Giants -0.33 WPA. Add in that he didn’t do much else for the rest of the game, while Ramirez racked up 0.14 WPA in the 10^th before giving up Rick Ankiel’s monster shot in the 11^th (0.38 WPA) to finish with only -0.154 WPA for the game.

Play of the Game: Rick Ankiel homers off of Ramon Ramirez, 11^th inning.

One of the few times I’ve ever shouted at my TV before the hitter actually had hit the ball. The pitch was only about halfway to the plate, and I was already shouting “NO!!” I mean, it was a fastball right down the middle of the plate, belt high. I could have hit that halfway to Berkeley.

Game Three: Giants win 3-2, Giants lead series 2-1

Player of the Game: Buster Posey, +0.413 WPA, and Jonathan Sanchez, +0.486

I did an offensive and pitching player of the game because starting pitchers get so many more opportunities to affect their WPA. I chose not to do it for Game 1 because no Giant really did anything of note on offense. Fair? No. Sue me.

Posey gets most of his WPA from one play that will be remembered for a long time: his groundball to second that Brooks Conrad promptly Bucknered into right field. That play garnered 0.36 WPA for Posey, and sent me into hysterics. I cannot remember a sporting event that sent from an extreme high, to an extreme low, and back to an extreme high in such a short time period. Looking at the Win Probability Chart is really quite amusing (http://www.baseball-reference.com/boxes/ATL/ATL201010100.shtml#wpa).

About 2 years ago, Joe Posnanski wrote “I never argue with people who say that baseball is boring because, well, baseball is boring. But then, suddenly, it’s not. And that what makes it so great.” It has been a long time since I saw a final two innings that gripped me as much as the 8^th and 9^th of this game did.

Quickly, on Jonathan Sanchez: kid looked great. He struck out 11 in 7 and a third, 2 hits, a run, and (importantly) only one walk. I know Lincecum and Halladay took all the playoff-debut thunder, but Sanchez didn’t do too shabby himself in his first-ever playoff start.

Goat of the Game: Sergio Romo, -0.535 WPA

This one was easy. Romo gave up a 2-run home run to Eric Hinske after Bruce Bochy got out-managed by the old dog, Bobby Cox. Another scream-at-my-TV moment. That home run alone was worth -0.55 WPA, but Romo stayed in to get the next two batters out at +0.01 WPA each.

I feel bad for Romo, who had one of the better middle-relief seasons by a Giant in recent memory. He had a 2.18 ERA, a WHIP under 1, and a K/9 over 10 with a BB/9 of less than 2 (hard to do, folks.) But he had such a terrible NLDS that I doubt Bochy will give him any work of note in the NLCS.

Play of the Game: Eric Hinske homers off of Sergio Romo, 8^th inning.

In the words of Forrest Gump: “Well, that’s all there is to say about that.”

Game 4: Giants win 3-2, Giants win series 3-1.

Player of the Game: Cody Ross, +0.262 WPA

Jesus Giants, you want to stop with the one-run games? You’re killing me over here.

Cody Ross broke up Derek Lowe’s no-hitter with a one-out solo shot in the fifth inning. That tied the game at 1 and was worth 0.17 WPA all by itself. He then managed another 0.11 WPA when he singled Buster Posey home with two down in the seventh to give the Giants a 3-2 lead, a score that would hold. Man, is there anything this guy doesn’t do*?

*So I was reading my Twitter feed right after Game 3 of the NLCS ended (hell yes Giants!) and I saw that three consecutive MLB posts called Cody Ross, ‘Babe.’ I found this to be clever and funny, and immediately texted some of my Giants-loving friends that joke. As it turns out, apparently Tim McCarver called him that during the game and I had somehow missed it. Tim McCarver? Not the funniest or most clever person in the world. Damn.

I’ve talked about Ross enough, so just one last thing: by WPA (as I said earlier), he was the Giant’s second most important hitter this series (after Huff). Some people just get hot at the right time, huh?

Goat of the Game: Aaron Rowand, -0.124 WPA

There was no real goat to this game, so I’ll pass over this quickly. Rowand’s strike-out against Johnny Venters in the 7th could have really hurt the Giants, but Cody Ross followed it up with his single to score Posey. No harm, no foul.

Play of the Game: This one’s an interesting one.

There are actually four plays that are tied with a 17% swing in Win Expectancy. First, Brian McCann’s home run off of Madison Bumgarner in the sixth broke a 1-1 tie. Cody Ross’s home run in the top of the sixth made it tied in the first place. Melky Cabrera’s ground out to 3^rd ended the game, and Juan Uribe reached on Troy Glaus’ error in the 7^th, tying the game at 2. To add on to that, the 5^th most important play, Brian Wilson striking out Omar Infante with one out in the ninth, had a Leverage Index score of 7.71! That means that at-bat was seven times as pressure-packed as an average at-bat. As this confusion indicates, this was a hard-fought game all the way through, and to name one play of the game wouldn’t be right.

Series Review

Well, that wasn’t good for my heart. Four consecutive games decided by one run. Home runs and blown saves and Cody Ross, oh my! But, in the end, the Giants advanced to the NLCS (and hold a 2-1 series lead!!!!) and all is right with the world. Some notable WPA figures:

Aubrey Huff: +0.278 WPA. He had a negative WPA in only one game, Game 1, and put up a huge 0.269 WPA effort in Game 3.

Buster Posey: +0.176 WPA. Good enough to tie him for third on the team (with Pat Burrell). Huge Games 1 and 3, but that double play ball in Game 2 killed him.

Pablo Sandoval: -0.167 WPA. And that was in only 2 games. Probably a good idea to keep him on the bench.

Andres Torres: -0.245 WPA. The worst score for a hitter. He didn’t have any downright awful games, but he only put up a positive score in one game (Game 1). He was benched for Game 3 of the NLCS, and that’s probably a good decision, because he’s definitely slumping.

Brian Wilson: +0.358 WPA. Blew a save in Game 2, came back in huge spots in Games 3 and 4. Andrew Baggerly tweeted earlier today, “Brian Wilson should change his entrance music to Rolling Stone’s ‘Paint It Black.’” Needless to say, the blog heartily agrees.

Sergio Romo: -0.679 WPA. Oof. It was a rough series for Romo, who started the Braves’ rally in Game Two and finished it in Game Three. He’s a fantastic setup man though, and I’m sure he’ll be back as good as ever.

Well, all in all, that was quite a series. After many trials and tribulations, the Giants are back in the NLCS for the first time since 2002, and I’ve can’t help but think of Scott Spiezio. Dammit. Giant’s baseball: torture, indeed.

Pitcher Abuse Points: Revisited

A few days ago, I was wasting time while at work reading the excellent Rany on the Royals blog, written by Baseball Prospectus correspondent Rany Jazayerli, and Rany linked to an article he wrote from BPro entitled “Pitcher Abuse Points.” The article was published to the site on June 19th, 1998, and offered up a system that ‘awarded’ points to pitchers as their pitch count went up. There’s a lot of fascinating stuff in the article, but what I really like about it is how long ago it was published. These days, we understand that, hey, Dusty Baker, it might not be a great idea to have Mark Prior throw 330 innings before he turns 23, but, as it will turn out, that was the norm not so long ago.

At the beginning of the article, Rany explains why one can’t just use innings pitched as a measurement of how taxed a young pitcher’s arm is, and this is obvious. David Cone and Greg Maddux both threw a ton of innings as young pitchers, but only one of them retired early with arm problems. As Rany puts it, “it’s the pitches, stupid.”

After that bit of explanation, Rany introduces what he calls “Pitcher Abuse Points.” The scale went like this: pitches 1 through 100 received zero points, each pitch between 101 and 110 received one point, each pitch between 111 and 120 received two points, each pitch between 121 and 130 received three points, and so on. Intuitively, this makes sense: as the pitch count gets higher, each pitch is more tiring on the arm. Now, as we just said, this system isn’t going to work for everybody but it does tell us something.

Let’s start with the list of most abused arms in 1998. As Rany computes it, Randy Johnson and Roger Clemens were the two most abused arms of the first half of 1998, with PAP per start scores of 38.2 and 36.2, respectively. There’s not a lot of surprise there, as they were their clubs ace, the stopper, the bullpen saver, the [insert workhorse cliché here]. The third name gives us something to work with: Bartolo Colon. He measures a 34.0 PAP per start score in 1998 at the age of 23.

Colon’s an interesting case. From 1998 to 2005, he pitched over 1,700 innings without major shoulder or elbow injury. It wasn’t until then that he started experiencing arm problems, as many 33-year-olds with almost 1,800 career IP do. In the 2005 playoffs, he suffered a torn rotator cuff that caused him to miss most of the 2006 season, and he’s been plagued with shoulder and elbow injuries since. However, the fact remains that he pitched effectively for eight seasons before experiencing arm problems, posting an ERA+ of 110 or higher each of those seasons except 2004. He even won a Cy Young in 2005.*

*It should be noted that that Cy Young Award was a complete sham, and perhaps the dying breath of the power of the win. Colon had a 3.48 ERA to go along with his 21 wins, which led the league. That was good for a 122 ERA+ - certainly excellent, but barely Cy Young-worthy. The award that year almost certainly should have gone to Johan Santana, who posted a 155 ERA+ in 231 innings, but only recorded 16 wins with a punchless Twins lineup supporting him.

Continuing our look at young pitchers, numbers 5 and 6 on Rany’s list don’t really fit the prediction either. They would be Livan Hernandez and Pedro Martinez. Pedro, of course, would go on to be perhaps the greatest pitcher there ever was from 1999 to 2001*, so we don’t need to discuss that any further.** Hernandez was before, and has been, a just-below-average pitcher. He posted a 95 ERA+ from ’96-’98, and a 97 ERA+ since then. It should be noted that he’s pitched over 2500 innings from 1999 to the present, and has never suffered a major arm injury.

*Martinez would post a 2.01 ERA in those 3 years, with 760 K’s in 547 innings against only 94 BB’s, and astounding 8.09 K/BB ratio. That all works out to a 246 ERA+, which, for perspective, has only been achieved by 3 other pitchers in one SEASON since integration. (Those three pitchers? That’s our AFLAC trivia question.)

**Though it is worth mentioning that in 2001, Pedro only made 18 starts due to a rotator cuff injury.

The next young pitcher on the list is more interesting. Jesus Sanchez scored a 21.9 on the PAP index, the year he threw 171 innings as a rookie. He was below league average that year, and would be his entire career. He would only one other season throw for more than 171 innings, in 2000 he would throw 182. But one wonders what might have been. Rany says, “Sanchez…is having his right arm throw to the wolves.”

The next name on the list is another interesting one, and one of my sneaky-favorite players on my youth: Brad Radke. I’ve always felt Radke was one of the more underrated players from my 9-15 years, posting a 116 ERA+ despite a unseemly 4.35 ERA. Indeed, after his 1998 season (his third with 200+ IP), he posted 200+ IP’s 6 seasons over the next 7, though he did miss about 12 starts in 2002.

He rather memorably carried the Twins to a postseason berth in the 2006 season (at the expense of his right shoulder), especially replacing Francisco Liriano (more on him later) after he went down with elbow soreness. Of course, he suffered from a torn labrum since the beginning of the season, and had pitched through it the entire season. He also suffered a stress fracture in his right shoulder in late August, but only missed one start. The extent of his injuries forced him to retire after that season, at the age of 33.

A quick rundown of the other young pitchers on Rany’s list: Jaret Wright was 22 when he scored a 15.1 PAP, and he missed most of three seasons with a chronic shoulder injury, starting in 1999. Shawn Estes pitched regularly until 2004, and then suffered several shoulder problems, including undergoing Tommy John surgery that caused him to miss the entire 2006 and 2007 seasons. Jason Schmidt pitched effectively for several years, including a great 2003 season when he posted a 2.34 ERA and struck out a batter an innings. He signed with the Dodgers in 2007, but was shut down after 25 innings due to shoulder soreness.

Ah, but of course I forgot to mention perhaps the most interesting name on Rany’s list: Kerry Wood. Wood was one of the most impressive rookie pitchers of all time, going 13-6 with a 3.40 ERA and a 12.6 K/9 ratio in his rookie season in 1998. He famously pitched a one-hit, zero-walk, twenty strikeout game, good for a 105 Game Score, the highest since 1974 and 18th best since 1952, when Baseball-Reference begins tracking Game Scores. He pitched 166 innings that rookie season, and scored a 14.2 PAP. Here’s what Rany has to say about Wood:

“And, of course, I can't write this article without mentioning Kerry Wood. At 21, he's the youngest name on this list, and he's in the middle of the pack as far as abuse goes. He hasn't thrown more than 128 pitches in a game this year, but he has a number of outings in the 120+ range. I don't think he's in grave danger of injury - he's a big guy with good mechanics, relies on his fastball, and doesn't throw a splitter. But I do think that Jim Riggleman should take a little more care of the most prized arm of the decade.”

We all know what happened. Wood would miss the entire 1999 season after tearing his ulnar collateral ligament, and underwent Tommy John surgery. He’s struggled with arm problems ever since. I sure hope Riggleman doesn’t find himself managing “the most prized arm of the decade” again; I would hate to see what might happen to that poor young man.

Pining for Fred Lewis

Welcome to my silly summer project, the Designated for Assignment blog. As some of you know, I’ve long held a small interest in writing and I figured, while I work three days a week over the summer, I might as well indulge my interest and work on my writing skills at the same time.

If you’re reading this, you’re probably a friend or family member whom I have pointed in this direction, or perhaps a follower on Twitter, all 4 of you. I’m still trying to figure out this whole Twitter thing.

What will I be writing about, you ask? Well, I don’t really have a clue, but as you all surely know, I am borderline obsessed with baseball, so inevitably the majority of the posts will be about that. I’ll try to mix in some team analysis, mostly Giants, with some broader-picture posts and some fun research projects. I’ll also be throwing in some random posts about nothing in particular; we’ll just kind of see how this goes.

You may note that the picture to the right is of former Giants and current Blue Jays left fielder Fred Lewis, who the Giants essentially gave away on April 15th after spending four years as an effective fourth outfielder. Lewis was drafted in the second round of the 2002 amateur draft, and when he made a six game cameo appearance at Triple-A Fresno in 2004, he appeared on Baseball America’s Top 100 Prospects list at #78. After 11 PA’s with the big boys in 2006, Fred posted a .292/.366/.550 slash line in 192 PA’s at Fresno in 2007 before coming up for good.

It was his year and a half at Fresno when I first became a fan of Fred’s. He was, to me, the kind of ballplayer I wanted to be: lithe, effortless, fast but with a touch of power, a good outfielder who played with a certain grace. I became convinced that Lewis was the Giants’ left fielder of the future.

Ah, but to be a Giants fan is to know disappointment. The Giants, instead of starting an exciting young outfield including the 26-year-old Lewis and the 24-year-old Nate Schierholtz (who had slugged .560 and played an astonishingly good right field in Fresno the year before), decided to resign a 33-year-old Randy Winn and to give Phillies reject Aaron Rowand a five year, $60 million contract. Rowand was 30 years old and, although he was coming off an excellent year in 2007, he had been below average in ’06 and ’05, and he promptly put up an OPS+ of 94 in 2008, 90 in ’09, and an abysmal 68 so far this season.

Anyway, back to Lewis. He received only 180 PA’s in 2007 and hit .287/.374/.408, and in 2008 he delivered a .282/.351/.440 line in 521 PA’s, good for a 105 OPS+. For some reason, he only got 336 PA’s in 2009 as the Giants gave a combined 615 PA’s to Schierholtz and Eugenio Velez (more on him in a minute), who delivered with OPS+’s of 81 and 83, respectively. Lewis didn’t even make the big league club this spring, and, after putting up a ridiculous .409/.536/.773 line in 29 PA’s in Fresno, was sent to the Blue Jays for $75,000.

After being told by the team that drafted him that he was unwanted, and three seasons as an average-at-worst outfielder, and after being sent to a more difficult league in the most competitive division in the sport to boot, Lewis has put up a slash line of .280/.324/.460 in 217 PA’s for Toronto, good for a 112 OPS+. Meanwhile, the Giants have given Rowand, Schierholtz, Velez, Mark DeRosa, Andres Torres, and John Bowker a combined 794 PA’s in the outfield this season, and they have OPS+’s of 68, 93, 67, 42 (!!), 134, and 58 respectively, with Torres representing the only batter who measures above average.

For some of that massive heap of epic fail, explanations can be gleaned. DeRosa, who’s 42 OPS+ almost defies belief*, signed for 2 years and $12 million before this season, and seems to have some sort of wrist injury that he’s been dealing with. Rowand is costing the G-men $12 million this season, and, incredibly, for the next two seasons as well. If there’s one thing a general manager has trouble accepting, it’s realizing that your $60 million investment isn’t exactly performing the way you had hoped. Schierholtz is only 26 and is just waiting for his power to come around, but in all honesty is receiving the PA’s he should have gotten two years ago and probably isn’t any better than a fourth outfielder. Torres is raking, and Bowker is another young-un who has hit in the minors and just needs exposure to big league pitching.

*Since expansion in 1993, there have been 16 seasons with 300 PA’s and an OPS+ of 42 or lower. The lowest? Luis Sojo, who put up a sterling 32 OPS+ for the Mariners and Yankees in 1996, with a slash line of .220/.250/.272. DeRosa probably won’t get to 300 PA’s this season, but it’s worth mentioning.

But it’s the curious case of Eugenio Velez that’s got me so furious. About a year and a half younger than Fred, he broke in with the big league club in 2008, a year after Lewis’s arrival. In the three seasons and 666* PA’s since then, he’s got a career OPS+ of 80, and an OBP of just .303, right at the Mendoza line of that statistic. He played well-below-average defense at second base in 2008 and 2009 before management decided they’d rather he play well-below-average defense in the outfield. But Velez offered an eye-popping tool, his considerable speed, which entranced management into keeping him around to put up a .275 OBP in 54 PA’s this season, while Lewis gets sold for chump change and promptly on-bases 50 points higher and slugs 90 points higher than his de-facto replacement.

*666? Ominous.

I just don’t understand why Brian Sabean decided it was time for Lewis to go and Velez to stay. The numbers clearly show that Lewis is a valuable player, and significantly more so than Velez. At worst, Fred is a major-league average outfielder, and for $455,000 he deserved better than to be kicked to the curb. These are the kind of decisions that can add or subtract two or three wins a season, and in the NL West that could be the difference between making the playoffs and watching them from home. But the Giants management never really has paid much attention to numbers, not when there’s a 34-year-old, injury-prone outfielder with a career 96 OPS+ who’s been traded twice in the last 13 months on the market for $6 million a year. Gotta catch ‘em all!

You know, it was the triples that made me fall for Fred Lewis at first. He hit 47 of them in his 2500 or so PA’s in the minors, and he proved those numbers were no fluke by hitting 11 in 2008, 7 in triple-happy AT&T Park. The triple is, of course, the most exciting play in the game, all nine fielders struck into motion as the runner rounds second at full speed, flops face-first into third and wraps his arms around the bag. The whole play is poetry in motion; it’s often described as “the most exciting twelve seconds in sports.” It’s hard not to love a player who regularly finds himself standing on third after a shot in the gap.

So long, Fred. You deserved better.