As a followup to my earlier piece introducing my new stat, I thought I'd put it in some context. I'll split my analyses up some. In this piece, we will examine second-order UVI for hitters to get a better feel for what it offers.
As I explained in my earlier article, second-order hitting UVIs account for a player's home park along with his performance. Since all the examples I'm about to use are from the major leagues (for familiarity's sake) there's no need to go into third-order UVI here, since all that does is account for level.
How high can it go?
Well, UVI is on the same scale as slugging percentage, so if a player hits a homer every plate appearance, their UVI will be 4.000.
For a slightly more realistic idea of UVI's ceiling, a better idea is Barry Bonds' incredible 2001. In that year, Bonds had a second-order UVI of .897, which goes along with the idea that UVI tends to run slightly higher than slugging percentage. Since it is almost inconceivable to imagine a better season than Bonds' 2001, it is safe to assume that a .900 UVI is about as good as it gets.
For those of you who argue that Bonds' season should be thrown out, fine. For another idea of a ceiling, let's look at A-Rod's 2007 for an example of a great season. Rodriguez had a .687 second-order UVI last year, so his '07 was only 7/9 as good as Bonds' '01. Wow.
Some others to chew on:
Jimmy Rollins' '07: .572
Matt Holliday's '07: .573
Ooooh...wow...I expected Holliday to win this by a mile, but Coors drives his first-order advantage from .17 to .01. Yes, Holliday is the better hitter, but Rollins' baserunning helps not only with steals, but with grounding into fewer double plays. Given that Rollins plays a tougher defensive position than Holliday, maybe he deserved that MVP award after all...
Ryan Braun's '07: .626
For someone to put up a UVI like that with Braun's kind of walk rate means he's one hell of a hitter. I know, I know, you didn't need me to tell you that, but look at how he dwarfs Holliday and Rollins. Did he deserve the MVP? Given his horrific defense, it's tough to really say yes, but Braun sure hit the heck out of the ball.
Ryan Howard's '06: .665
I put this up simply because it's got the highest HR total of any season since the Steroid Era. It also serves as an interesting study, though, because Howard walks a lot, strikes out even more, and is a big handicap on the bases. Apparently it all works out for him pretty well.
Hanley Ramirez's '07: .631
Ramirez's appearance out of relative nowhere had me skeptical, but you can see that his UVI blows Rollins' and Holliday's out of the water. Yes, he's a well-below-average SS, but he's better there than Derek Jeter, and he hits better as well. That will be great for whoever Florida eventually trades him to. How in hell have they won two World Series over there?
How low can you go?
The worst a hitter can possibly do would be to ground into a double play every time, which would yield a UVI of -1.000. When doing 3rd-order UVI translations from short-season leagues, the translations can often go into the red due to ridiculous K rates.
As far as 2nd-order UVI goes, it usually won't get anywhere near zero with a decent sample size. For example, Bobby Crosby put up a .374 UVI last year with a batting line of .226/.278/.341. Therefore, most second-order UVI's usually sit above .350. Generally, if a hitter can't top that, they get sent down or released.
UVI As A Problem-Solver: The Dodgers Outfield
Perhaps UVI's greatest asset on the hitting side of things is its combination of hitting and baserunning. Proponents of speed players often scoff at stats like OPS because they don't take into account all the extra bases a speed player can make through stealing and bunting. UVI does account for that.
Now, the Dodgers have four outfielders supposedly worthy of starting jobs: Andruw Jones, Matt Kemp, Juan Pierre, and Andre Ethier. Statheads like me hate Pierre, and think he is best relegated to fifth outfielder duty, while traditionalists think he's a prototypical leadoff man. Anyone's mind can be erroneous when thinking about such things, including mine, so I decided I would let the numbers do the talking and see what they had to say.
Second order UVI's of the Dodgers' four outfielders, 2006-07:
Matt Kemp: .473 in 2006, .515 in 2007
Andre Ethier: .478 in 2006, .460 in 2007
Juan Pierre: .445 in 2006, .430 in 2007
Andruw Jones: .567 in 2006, .451 in 2007
Ahh...so the statheads win. Joe Torre made a good call. What's interesting here, however, and something to keep an eye on, is how close Ethier's, Jones', and Pierre's 2007 UVIs were. So far this season, it has been Ethier who has gotten the most starts, and Kemp is third. Kemp seems miles ahead of the rest, and the optimal alignment would seem to be Ethier in left, Jones in center, and Kemp in right.
Notice also that for all their youth and virtues, neither Kemp nor Ethier has come close to Jones' 06. This is proof that a low ISW (in Kemp's case) or low ISO (in Ethier's case) makes it tough to put up a big UVI. Jones hit .262 in 2006, while Kemp hit .342 in 2007. Jones' slugging percentage was .531 to Kemp's .521, but it's the walk rate difference that really helps Jones out. This also serves as test point No. 5258 that batting average doesn't mean a whole lot in the end.
Test Point No. 5259: Juan Pierre.