October 12, 2008

My Twins MVP Ballot

Now that everyone has had a couple weeks to get over how the Twins' season ended, it seems like a good time to examine who the most valuable (and least valuable) players on the team were this year. There are certainly any number of different ways to approach such a task, but my preferred method is to utilize several different player-evaluation metrics and attempt to synthesize the information from them to form what should be one well-rounded opinion.

It'd be easy to go by gut feel, conventional wisdom, or a single statistic, but taking a wisdom-of-crowds approach by involving as much objective information as possible is more worthwhile. Rather than just post a bunch of numbers and reveal my team MVP ballot, let's walk through the different metrics used and examine what each one had to say about assigning value to Twins players this season. If you're unfamiliar with something being referenced, click on the links to learn more.

First up is Value Over Replacement Player (VORP), which takes a player's production and compares it to the "replacement level" for his position. Replacement level is considered readily available and easily obtained, so VORP examines how many runs someone was worth compared to a low-level player who a team could acquire at minimal cost to replace him. Joe Mauer gets compared to catchers, Justin Morneau gets compared to first basemen, and so on, with playing time being a big factor.

                    VORP                              VORP
Joe Mauer 55.5 Carlos Gomez -0.9
Scott Baker 46.0 Michael Cuddyer -0.9
Justin Morneau 45.5 Livan Hernandez -1.5
Joe Nathan 30.2 Juan Rincon -2.9
Kevin Slowey 30.1 Craig Monroe -3.3
Nick Blackburn 23.2 Adam Everett -3.7
Denard Span 20.6 Mike Lamb -10.5
Jason Kubel 20.1 Boof Bonser -11.4
Dennys Reyes 17.6
Craig Breslow 15.6
Glen Perkins 15.0
Delmon Young 13.1
Nick Punto 12.6
Jesse Crain 10.9
Brendan Harris 9.8
Francisco Liriano 8.8
Alexi Casilla 8.4
Brian Buscher 4.3
Matt Tolbert 3.8
Brian Bass 1.7
Matt Guerrier 1.6
Mike Redmond 0.6

As you can see, because replacement level is such a low baseline for comparison, few players racked up negative VORP totals. By hitting .328/.413/.451 in 633 trips to the plate, Mauer was 55.5 runs better than a replacement-level catcher offensively. In other words, if the Twins lost Mauer and were forced to replace him for an entire season with someone like September call-up and minor-league veteran Ryan Jorgensen, the lineup would score 55.5 fewer runs.

By tossing 172.1 innings with a 3.45 ERA, Scott Baker led the staff by being worth 46.0 runs more than a replacement-level pitcher. In other words, if the Twins lost Baker for the entire season and replaced him in the rotation with someone like rookie September call-up Philip Humber, the pitching staff would have allowed an additional 46.0 runs. VORP is an important concept, but it's also easy to understand. How many runs would a team gain or lose by replacing a player with some freely available scrub.

For instance, during the offseason the Twins signed four veteran free agents in Livan Hernandez, Mike Lamb, Adam Everett, and Craig Monroe. All four proved to be busts at a combined cost of $17 million, but they also cost the team 19 runs compared to a replacement-level player (without defense factored in). On a team-wide level 10 runs typically equals one win, so choosing the veteran foursome instead of four minimum-salaried replacement-level players cost the Twins about two wins and $16 million.

Next up is Runs Above Average (RAA), which is similar to VORP except that all comparisons are made to "average" rather than "replacement level." Average is a much higher baseline for comparison, so the result is far fewer positive contributions and far more negative contributions. Having a negative VORP means that you were horrible, but the same isn't necessarily true of having a negative RAA because an entire team full of perfectly average players would go 81-81. There's plenty of value in being average.

                     RAA                               RAA
Joe Mauer 43.0 Matt Tolbert -1.1
Scott Baker 25.2 Brian Buscher -1.4
Joe Nathan 22.4 Mike Redmond -2.2
Justin Morneau 18.9 Glen Perkins -3.2
Dennys Reyes 12.3 Brendan Harris -3.5
Craig Breslow 11.2 Alexi Casilla -5.0
Kevin Slowey 10.9 Juan Rincon -6.1
Denard Span 7.2 Brian Bass -6.2
Jason Kubel 6.0 Delmon Young -6.5
Jesse Crain 3.7 Matt Guerrier -7.2
Nick Punto 1.1 Craig Monroe -7.5
Nick Blackburn 0.1 Adam Everett -7.8
Francisco Liriano 0.0 Michael Cuddyer -10.1
Mike Lamb -17.4
Livan Hernandez -18.4
Carlos Gomez -21.8
Boof Bonser -25.3

Mauer stands out even further from the pack in RAA, largely because the average MLB catcher hit just .257/.325/.390 this season. Except for pitcher, that's the worst offensive production from any position on the field. As noted in this space last week when looking at Mauer's historically great start to his career, most people fail to realize how difficult it is for catchers to consistently post great numbers offensively. As a whole catchers simply don't hit, which is why the Twins have such a huge advantage with Mauer.

Morneau's relatively low RAA total will no doubt shock the people who view him as the Twins' clear-cut MVP, but those people likely weren't paying attention to positional comparisons anyway. Mauer fares so well in RAA because catchers as a group are poor offensively, but the opposite is true for Morneau and first basemen. Throughout baseball history first base has been home to sluggers and this year was no different, as the position featured the second-most offense behind only right field.

                      AVG      OBP      SLG      OPS
Joe Mauer .328 .413 .451 .864
MLB Catchers .257 .325 .390 .715

Justin Morneau .300 .374 .499 .873
MLB First Basemen .266 .349 .447 .796

Morneau was an excellent hitter this year, but so were a lot of first basemen. Compared to his position Morneau's advantage was 13 percent in AVG, 7 percent in OBP, 12 percent in SLG, and 10 percent in OPS. Meanwhile, compared to his position Mauer's edge was 28 percent in AVG, 27 percent in OBP, 16 percent in SLG, and 21 percent in OPS. Based strictly on their hitting, Morneau was 8-10 percent better than the average first baseman while Mauer was 20-22 percent better than the average catcher.

Most people evaluate hitters by looking at raw numbers, whether it's batting average, homers, RBIs, or OPS. However, there's a huge difference between offense coming from a hitter-friendly spot like first base, left field, right field, or designated hitter and offense coming from a defensive-driven position like catcher, shortstop, second base, or center field. All of which is why having Mauer at catcher is a bigger advantage for the Twins than having Morneau at first base.

Not only is Mauer significantly better than the catchers other teams are collectively putting behind the plate, he lets the Twins get a big bat in the lineup while still leaving first base, left field, right field, and designated hitter open for other potential big bats. Mauer and an average first baseman would create more offense than Morneau and an average catcher, and while often ignored the logic supporting that fact is one of the basic building blocks for performance-based baseball analysis.

VORP and RAA both view all plate appearances and innings as equal, so tossing a scoreless inning with a 10-run lead counts just the same as throwing a scoreless frame in extra innings. Win Probability Added (WPA) is much different, because it's designed specifically to measure the impact that each play has on a team's chances of winning and assigns that value to the individual players responsible. For example, WPA values a walk-off homer much differently than a solo shot in a 10-2 game.

Fan Graphs creator David Appelman has helped me take the Twins' raw WPA totals and adjust them based on the MLB average at each position. Most adjustments are minimal, but starting pitchers get a boost relative to relievers and hitters who play up-the-middle spots are given a boost relative to hitters who man corner positions. All of which results in a sort of adjusted WPA (adjWPA), which like VORP and RAA does not factor in defense:

                  adjWPA                            adjWPA
Joe Mauer 6.22 Dennys Reyes -0.14
Scott Baker 3.06 Craig Monroe -0.22
Joe Nathan 3.01 Brendan Harris -0.28
Justin Morneau 2.69 Nick Punto -0.34
Denard Span 1.45 Brian Bass -0.37
Francisco Liriano 0.64 Michael Cuddyer -0.38
Nick Blackburn 0.49 Matt Guerrier -0.41
Brian Buscher 0.45 Mike Lamb -0.50
Kevin Slowey 0.42 Matt Tolbert -0.54
Craig Breslow 0.33 Alexi Casilla -0.71
Jason Kubel 0.27 Juan Rincon -0.89
Glen Perkins 0.27 Jesse Crain -1.03
Adam Everett 0.19 Mike Redmond -1.25
Delmon Young -1.91
Livan Hernandez -2.18
Boof Bonser -2.49
Carlos Gomez -2.62

Seeing Mauer atop the adjusted WPA list is no surprise, because he led the Twins in raw WPA just like he did in VORP in RAA. In fact, Mauer led the entire league in WPA among position players at 4.88. He posted that total in 633 plate appearances whereas the average MLB catcher produced -1.34 WPA per 633 trips to the plate, which is how he gets an adjusted WPA of 6.22. Morneau ranked second on the team in raw WPA with 3.87, but given his 712 PA the average first baseman would've had 1.18 WPA.

Adam Everett fares far better in WPA then in VORP or RAA, because while he batted just .213/.278/.323 and managed only 27 hits, one of them was the game-changing, butcher-boy double against the Rays. WPA views that play as having a tremendous impact, whereas VORP and RAA view it the same as all other doubles. At the other end of the spectrum, Delmon Young and Carlos Gomez both fare horribly in WPA because they repeatedly failed to come up big in crucial situations.

Similarly, Jesse Crain does well in both VORP and RAA because without context his 3.59 ERA in 62.2 innings out of the bullpen is very strong. However, WPA accounts for the fact that Crain struggled in key spots, allowing a .277/.360/.523 mark in high-leverage situations compared to a .250/.311/.381 line in low- or medium-leverage situations. Crain was much worse when the game was on the line, and WPA factors that in while VORP and RAA treat all of his innings as equal.

One thing that VORP, RAA, and WPA all have in common is that they each fail to account for defensive contributions, which is obviously a major factor when assessing a player's overall value. In an attempt to rectify that problem, I've gathered defensive stats from The Hardball Times, Baseball Think Factory, ESPN.com, and Baseball Prospectus to come up with a defensive estimate for each player that can be tacked on to the hitting and pitching value found in VORP, RAA, and WPA.

Throwing defense into the mix further widens Mauer's lead in every metric, because he was one of just four MLB catchers to log 1,200 innings behind the plate, ranked second in the AL by gunning down 36 percent of steal attempts, and helped lead a young pitching staff to a 4.18 ERA. Accounting for defense also moves other position players up the lists. For instance, Gomez ranks 23rd, 29th, and 30th based on poor hitting, but his great glove wipes away a major chunk of that negative value. Here's my ballot:

 1. Joe Mauer                   16. Carlos Gomez
2. Justin Morneau 17. Brendan Harris
3. Scott Baker 18. Jesse Crain
4. Joe Nathan 19. Delmon Young
5. Denard Span 20. Matt Tolbert
6. Kevin Slowey 21. Michael Cuddyer
7. Jason Kubel 22. Mike Redmond
8. Nick Blackburn 23. Matt Guerrier
9. Alexi Casilla 24. Adam Everett
10. Nick Punto 25. Craig Monroe
11. Brian Buscher 26. Brian Bass
12. Francisco Liriano 27. Juan Rincon
13. Glen Perkins 28. Livan Hernandez
14. Dennys Reyes 29. Boof Bonser
15. Craig Breslow 30. Mike Lamb

Any MVP case for Morneau must revolve around his RBI total, because Mauer dominates in metrics that evaluate overall offensive and defensive value in the context of position. Plus, even an RBI-fueled case for Morneau misses the mark if it looks only at raw totals. Morneau easily led the Twins and narrowly missed leading the league with 129 RBIs, but that's largely due to his coming to the plate with more runners on base than any hitter in baseball (which, in turn, is largely due to Mauer getting on base).

Morneau came to the plate with an astounding 558 runners on base this season. Not only did that lead MLB by 37 and lead the AL by 74, it was the highest number of runners on base that any hitter has had since 1996. People making the case for Morneau as MVP like to focus on his being a "clutch hitter" and "run producer," but in reality that's simply a product of being a very good hitter with a massive number of RBI chances.

Morneau drove in 18.9 percent of the runners that were on base for him, which is a great rate. However, Mauer drove in 18.7 percent of the runners that were on base for him. Both players cashed in runners at the same rate, but the difference is that Morneau came to the plate with 151 more guys on base than Mauer (thanks largely to having Mauer's excellent .413 on-base percentage in front of him). No wonder Morneau drove in 44 more runs than Mauer (while scoring one fewer and making 73 more outs).

If you want to simply hand the MVP title to the guy with the most RBIs, then Morneau is the easy pick. He's a very good player who drove in a ton of runs, and for many people that's where the debate ends. However, if you want to hand the MVP title to the guy who was actually the most valuable player based on all-around offensive and defensive contributions within the context of his position, then Mauer is an even easier pick and has a legitimate case for being the best player in the league.

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.