May 5, 2008

WPA Through April

Win Probability Added (WPA) measures how much impact specific plays had on the outcome of each game and assigns that value to the individual players responsible. For example, hitting a grand slam in the seventh inning when the score is already 10-2 has less WPA value than drawing a walk to lead off the ninth inning of a 2-2 game. The grand slam didn't have much impact on the likely outcome of the game, whereas the walk had a major impact on each team's chances of winning.

There are much better and longer explanations of WPA than that one, of course. If you're interested in learning more about it, Dave Studeman's WPA primer at The Hardball Times is a good place to start, and both Fan Graphs and Baseball-Reference.com offer tons of information on the subject. It's far from a perfect stat and is not meant to definitively prove how valuable each player has been, but WPA is an interesting tool to use in looking back at what has already taken place.

It's important to note than WPA doesn't measure any defensive contributions, which means that strong defenders don't receive full credit for their value. Beyond that, WPA doesn't place offensive contributions in the context of position, so an .850 OPS from a catcher or shortstop is treated the same as an .850 OPS from a designated hitter or left fielder. There's nothing that can be done about measuring defense via WPA, but it's relatively easy to put the numbers in better context by using positional adjustments.

With the help of Fan Graphs creator David Appelman, I've taken the Twins' raw WPA totals for April and adjusted them based on the MLB average at each position. Most adjustments are minimal, but starters are given a boost relative to relievers and hitters who play up-the-middle positions are given a boost relative to hitters who man corner spots. The end result is a sort of adjusted WPA (adjWPA), but before getting to that let's first take a look at the raw totals through April 30:

HITTERS                PA      AVG      OBP      SLG      WPA
Justin Morneau 110 .268 .345 .495 0.63
Joe Mauer 99 .295 .357 .386 0.12
Craig Monroe 50 .255 .300 .426 0.10
Brian Buscher 9 .250 .333 .375 -0.01
Matt Tolbert 54 .300 .340 .360 -0.04
Delmon Young 108 .265 .306 .314 -0.10
Denard Span 34 .258 .324 .258 -0.11
Jason Kubel 101 .237 .257 .381 -0.24
Brendan Harris 99 .287 .337 .379 -0.25
Adam Everett 29 .185 .214 .222 -0.26
Mike Lamb 84 .205 .226 .282 -0.28
Carlos Gomez 104 .265 .279 .373 -0.34
Nick Punto 44 .250 .318 .250 -0.36
Mike Redmond 14 .154 .214 .231 -0.41
Michael Cuddyer 40 .297 .350 .405 -0.47

As you can see, the offensive totals for April weren't pretty. Only Justin Morneau, Joe Mauer, and Craig Monroe provided a positive WPA--with only Morneau significantly above average--and the team as a whole batted just .260/.305/.362 while racking up -2.00 WPA. Seeing Morneau atop the list isn't news, but it's surprising to see Michael Cuddyer at the bottom given that he missed over half the month with a finger injury and hit a decent enough .297/.350/.405 when he did play.

Cuddyer's low WPA basically comes from two games (April 3 and April 25) in which he combined to go 1-for-10 while leaving 12 runners on base. He had -0.58 WPA between those two games, but 0.11 for the rest of the month. Monroe is the opposite, because on April 22 against the A's he went 3-for-4 with three RBIs, including the game-tying homer, producing enough WPA (0.32) to leave him as a positive contributor for April despite going 9-for-43 (.209) while accumulating -0.22 WPA the rest of the month.

I've heard it said that Carlos Gomez's overall struggles in April were lessened by his "single-handedly" winning games. While hyperbolic, that's actually true to some extent given that Gomez had four games with at least 0.10 WPA. To put that in some context, Morneau totaled six such games in April despite a vastly superior overall WPA. However, along with four huge games Gomez also had six games with WPA worse than -0.10. He was either very good or very bad, and the end result was -0.34 WPA.

PITCHERS               PA      AVG      OBP      SLG      WPA
Joe Nathan 43 .220 .256 .317 1.21
Nick Blackburn 159 .315 .350 .416 0.87
Dennys Reyes 30 .143 .200 .179 0.65
Scott Baker 122 .256 .287 .436 0.41
Boof Bonser 151 .250 .293 .379 0.29
Pat Neshek 48 .209 .255 .395 0.04
Bobby Korecky 11 .250 .455 .250 0.00
Matt Guerrier 65 .288 .354 .441 -0.07
Brian Bass 80 .288 .350 .575 -0.16
Kevin Slowey 14 .286 .286 .643 -0.16
Juan Rincon 41 .243 .317 .432 -0.19
Livan Hernandez 154 .310 .338 .490 -0.28
Jesse Crain 34 .226 .294 .484 -0.43
Francisco Liriano 56 .366 .509 .415 -0.67

Twins hitters combined for -2.00 WPA in April, but the pitching staff nearly balanced that with 1.50 WPA. Taken together that equals -0.50 WPA or a half-win below average, which is what the Twins were by going 13-14 in April. Joe Nathan led the way by converting 9-of-9 save chances with an 0.82 ERA in primarily big-pressure, high-leverage situations. Dennys Reyes was almost flawless, throwing nine scoreless innings while allowing just one of a dozen inherited runners to score.

However, Reyes totaled "only" 0.65 WPA because he faced 30 percent fewer hitters than Nathan and worked in spots that weren't quite as crucial. Pat Neshek also worked in high-leverage situations and held batters to .209/.255/.395, but had two disastrous appearances, totaling -0.45 WPA on April 7 and -0.54 WPA on April 14. Aside from those two games his WPA for April was a Nathan-like 1.03, but the value of WPA is that what you do in crucial situations has a huge impact.

Relievers were the Twins' strength in April, as the bullpen combined for a 3.54 ERA and 1.06 WPA over 84 innings. The rotation was more of a mixed bag, although as a whole the starters posted 0.45 WPA. Nick Blackburn was fantastic at 0.87, and both Scott Baker (0.41) and Boof Bonser (0.29) checked in solidly above average, but Livan Hernandez was well below average at -0.28 and Francisco Liriano was a mess at -0.67.

After looking at the Twins' raw WPA totals through April 30, let's switch to the adjusted numbers once each player is compared to the MLB average at their respective position. In other words, Morneau is compared to first basemen, Mauer is compared to catchers, Blackburn is compared to starters, and Nathan is compared to relievers. Positional adjustments don't cause any huge shifts yet because of the limited number of games and plate appearances involved, but there are some changes:

                      adjWPA                              adjWPA
Joe Nathan + 1.17 Matt Tolbert - 0.03
Nick Blackburn + 0.92 Matt Guerrier - 0.13
Dennys Reyes + 0.62 Kevin Slowey - 0.16
Justin Morneau + 0.45 Denard Span - 0.16
Scott Baker + 0.45 Delmon Young - 0.21
Joe Mauer + 0.34 Livan Hernandez - 0.23
Boof Bonser + 0.34 Juan Rincon - 0.23
Craig Monroe + 0.04 Brendan Harris - 0.24
Pat Neshek 0.00 Brian Bass - 0.24
Brian Buscher 0.00 Adam Everett - 0.25
Bobby Korecky 0.00 Carlos Gomez - 0.31
Nick Punto - 0.34
Mike Lamb - 0.35
Jason Kubel - 0.37
Mike Redmond - 0.38
Jesse Crain - 0.46
Michael Cuddyer - 0.53
Francisco Liriano - 0.65

Once positional adjustments are made only eight Twins ended April with positive WPA, which is a low total even considering the team's 13-14 record. Nathan, Blackburn, and Reyes led the top-heavy WPA distribution, while Morneau and Mauer were the lone hitters to contribute significantly above average for their position. Beginning May with a three-game winning streak has quickly changed the WPA picture, but this was the first of my planned month-by-month looks at WPA throughout the season.


Once you're done here, check out my latest "Daily Dose" column over at Rotoworld.

No Comments

No comments yet.

Sorry, the comment form is closed at this time.