Cabrera's Value

On the local SABR mail list there's been a discussion on what Orlando Cabrera is worth. We've seen a similar debate several places the last week. One side looks at his stats and points out that he topped out at "servicable" offensively and "suspect" defensively. The other side looks at the run the Twins made with him in the lineup and looks for impacts beyond the stats.

This debate gained a little momentum this weekend when Ron Gardenhire suggested that the Twins might still like Cabrera at second base as if this is a realistic possibility. It isn't. The Twins can find plenty of better free agents to play second base at a salary that is less expensive than Cabrera. And Cabrera can find a lot better offers that allow him to play shorstop. Sometime managers just want to say something nice about their ex-players. That's all this means.

But Cabrera's value takes a little twist when one looks at a neat statistic that's growing in popularity called Win Probability Added or WPA. To borrow from Denny Green, it is what is says it is: it adds (or subtracts) the affect a player had on the game by measuring how that player affected the probability of his team to win.

For instance, if player hits a single in four at-bats his WPA could vary significant depending on when he made those outs or got that hit. If he struck out three times while the bases were loaded and there was one out, his WPA for that game would likely be horrendous (and he would replace Punto as the voodoo doll of choice amongst Twins fans). If, on the other hand, he made three innocuous outs but got a key single in the bottom of the eighth that drove in the tying and winning runs, his WPA would be quite high (and he might be the player of the game.)

In other words, WPA goes beyond a player's overall statistics and beyond the myth that those hits are spread out more or less evenly. Some of those hits (or outs) are exceedingly more valuable than others, both positively and negatively. It often explains why fans who watch games thing a player had a better (or worse) year than his stats indicate.

It certainly could with Cabrera. His overall stats were mediocre, but in two months he posted a WPA of .527, which ranked fourth on the Twins hitters over all of 2009. Here's the list, by the way:

Name WPA
Joe Mauer 3.64
Jason Kubel 2.20
Justin Morneau 0.78
Orlando Cabrera 0.53
Michael Cuddyer 0.48
Matt Tolbert 0.30
Joe Crede 0.12
Jose Morales -0.12
Delmon Young -0.14
Denard Span -0.15
Brian Buscher -0.29
Mike Redmond -0.43
Alexi Casilla -0.61
Nick Punto -0.84
Brendan Harris -1.35
Carlos Gomez -1.97

Cabrera might not have had a great stint with the Twins statistically, but he absolutely made a difference in whether or not they won games in his two months here. Now that doesn't mean a thing for the future. It doesn't justify resigning him, and I'm far more excited about getting JJ Hardy here than I would have been if he returned.

But the love for Cabrera isn't just blind homerism. Indeed, it suggests a level of attention beyond overall stats, and an awareness of the situations in which those stats were compiled.

Coupla Notes
1. I haven't mentioned it much lately, but you can still download 1/3 of the TwinsCentric Offseason GM Handbook absolutely free at TwinsCentric.com. I have no idea why you wouldn't. What else are you doing today?

(You can also now by the hard copy, too!)

By the way - wanna guess who our very first trade target was? Yep - J. J. Hardy.

2. A couple of days ago I mentioned that Twins fans shouldn't discount the possibility that some top-notch Type A free agent second basemen might be available without needing to give up a draft pick. Recent news supports that.

First, we found out that the Tigers are desparate enough to cut payroll that they're willing to trade Curtis Granderson or Edwin Jackson. These guys are key cogs on their team, and they're both going to be WAY below market value next year. If they Tigers need to part with them, they certainly aren't going to offer Placido Polanco arbitration that could pay him at (or above) market rate next year.

And now we've found out that the relationship between Orlando Hudson and the Dodgers is icy, but the Dodgers have been talking to the Mets about Luis Castillo. It sure sounds like both sides are ready to move on, but the Dodgers have to realize that if they offer arbitration, Hudson will have to take it. I'll be surprised if they offer it. I'll be shocked if he doesn't accept it if they offer it.

Gettin 'Er Duen

What do the Twins need this second half? They need a bunch more “Brian Duensings.”

This week I got to participate in the Sports on Demand web show on Fox 9 (which you can see here). On of the midseason awards we presented was the Twins Most Valuable Player, which belonged to Justin Morneau, who was chosen unanimously. This is not a surprise.

But second place, at least on my ballot, should surprise.

When I look at MVP, I like to look at a statistic called Win Probability Added. WPA isn’t like most of the advanced offensive metrics: you’re not taking a bunch of other stats and adding them or dividing them together. WPA just measures how much each player impacts the probability of their team winning a game. So, if a player who comes to bat with his team having a 47% chance of winning a game, and leaves the at-bat with his team having a 51% chance of winning the game, he gets .04 points. If they have a 44% chance of winning the game when he’s done with his at-bat, he loses .03 points. That’s it.

I like WPA because it places a player’s performance inside the context of their situation. Pitching a scoreless inning in an 8-1 blowout doesn’t help your team much. Doing the same thing in a tie game does.

You find out that players with higher WPA scores are often fan favorites. Why? Because fans watch the game, and understand that this guy’s impact on their season was far higher than their numbers might indicate. For instance, if you liked Orlando Cabrera last fall, you had good reason. His stats were good, but his WPA was excellent. His impact far exceeded his time here.

If you look at WPA this season for Twins hitters (you can find them here), Morneau is where you would think he would be – atop the list and by a far amount. If you look at the list of pitchers, you find Carl Pavano, Jon Rauch and Francisco Liriano, but you find them in 2^nd, 3^rd and 4^th place. Guess who is at the top of that list…

A: Brian Duensing

How does one become one of the team’s most valuable players as a middle-innings reliever? The SOLE answer is consistency. Duensing has appeared in 38 games. Only in five of them has he decreased the team’s chances of winning. Those five are:

6/19 at Philadelphia – This was Saturday’s “comeback” game in Philadelphia. He gave up a solo home run in the 7^th when the Twins were down 8-4. Because the lead was already large, it didn’t have much of an impact. Total percentage impact: -1.2%

7/4 vs Tampa Bay – With the Twins already down 5-1, Duensing took over for Nick Blackburn with one out in the seventh, inheriting runners on first and second base. He got out of the inning but the two runners scored on a double by Gabe Kapler. Again, team is already losing, so a small impact. Total percentage impact: -3.4%

6/10 vs. Kansas City Royals – The Twins had been down 8-1 but had closed the gap to 8-6 as Duensing pitched the top of the ninth. After getting two quick outs, he gave up a home run to Wilson Betemit and a single before being placed to Matt Guerrier. The Twins continued their comeback in the bottom of the frame, but lost 8-9. Total percentage impact: -3.9%

Now we get to the one’s that hurt, because the game was close.

6/23 at Milwaukee – With the Brewers leading 3-2 in the seventh inning, Duensing walked a batter, got an out, intentionally walked a batter and hit Prince Fielder with a pitch. One of those runners eventually came around to score. Total percentage impact: -7.5%

5/22 vs Milwaukee – This was the Saturday game at Target Field, where the Twins grabbed an early lead, the Brewers tied it in the ninth, and the Twins finally won it in the 12^th after missing all kinds of previous chances to win it. I expect to find out that Duensing was the main culprit for that comeback. Nope. He just started the seventh with a single and a walk when the scores was just 4-2. Then he was pulled. Guerrier got him out of the inning without any runs scoring. Total percentage impact: -13.6%

Bu that’s it for negative WPA. By contrast, he has impacted 31 games positively (two were neutral). But what is amazing here is that these are not Herculean endeavors. These are a guy doing his job in a close game. Here are the five games he impacted most positively:

5/12 vs Chicago White Sox – The Twins were clinging to a 3-2 lead and Duensing started the 8^th. He struck out Juan Pierre, and got AJ Pierzynski to ground out to second. That’s it. Guerrier got the third out of the inning and the Twins eventually won the game 3-2. Total percentage impact: +9.9%

5/1 at Cleveland – The Twins blew a 4-2 lead to lose this game 5-4 in the 11^th inning when Jesse Crain was unable to clean up a mess that Alex Burnett created. Duensing pitched a scoreless 7^th with a 3-2 lead. Total percentage impact: +10.9%

4/24 at Kansas City – Duensing entered the eighth inning of a tie game with one out and a runner on first. He walked David DeJesus but got a line drive double-play to get out of the inning. This one was mostly luck. Total percentage impact: +11.3%

4/17 vs. Kansas City – Nick Blackburn had given up 5 runs in five innings, but the Twins had matched it. Duensing pitched the 6^th and 7^th innings of the tie game. He actually got a “W” for this one too, because Orlando Hudson hit a home run to lead off the bottom of the seventh. The Twins won the game 6-5.

4/9 at Chicago White Sox – This was the first game versus the White Sox, and the big question was what Liriano would look like away from Fort Myers. But the pitchers that carried the team that night were Guerrier and Duensing, who each pitched two scoreless innings in a 3-3 game. Duensing shut down the 9^th and 10^th frames, and also got the “W” when JJ Hardy singled in Joe Mauer in the 11^th inning.

There are another 25 examples like that.

Brian Duensing is never going to be a Cy Young winner, and is probably a longshot to make it to an All-Star game. But he’s doing his job on a consistent basis, and the impact it is having on the team is overwhelmingly positive. The Twins have more than enough talent on this team already. If they can coax more players to perform like Duensing – day-in and day-out positive impacts – this pennant race can still be over early in September.

----------------------

Have you downloded your 2010 Trade Deadline Primer yet? How about the free 1/4 book sample? Why not? It's FREE. And it's going to help you sort through all the trade B.S. that is rushing your way. Get it! Hurry!