On Wednesday, I gave a few examples of Second-Order Hitting Ultimate Value Index (UVI2) and explained a few things about it. Now, I'm doing the same for First-Order Pitching Ultimate Value Index (UVI1). Sorry I'm not using pitcher UVI2 here; I've been having some technical issues with that part of my spreadsheet. All UVI2 does is adjust HRA anyway, so it's not too big a deal. Let's begin.
How high can you go?
Remember, pitcher UVI is inverse to hitters; so 0 is the best. In order to do that, a pitcher would have to strike out every batter he faced. Obviously, that's pretty much impossible, so let's look at some good ones.
Johan Santana's 2007 Adjusted Line: 216 IP, 192 H, 1.13 WHIP, 4.08 ERA, .317 BABIP, .229 AVG, .277 OBP, .400 SLG, .677 OPS, .438 UVI
Notice that everything in here looks pretty good except for the ERA, the HR (remember, HR, BB, K, GB/FB% and HBP stay the same), and the BABIP. Santana's career BABIPs are usually lower than this. On one hand, this serves as an example of why I should include LD% in UVI (it would bring the BABIP closer to his normal level; on the other hand, it shows that Santana's strong flyball/homer tendencies may be a warning sign. His groundball% did drop some last year.
Erik Bedard's 2007: 181.7 IP, 142 H, 1.10 WHIP, 3.63 ERA, .305 BABIP, .207 AVG, .272 OBP, .336 SLG, .608 OPS, .391 UVI
Bedard was much better than Santana last year. His groundball percentage was over 10% higher, and so was his strikeout rate. Don't get too bent out of shape over the ERAs; the Expected Runs Allowed here is simply just taking Total Bases Allowed and dividing by four. It's crude, but it's only here as a sort of frame of reference until we all get better acquainted with UVI's ins and outs.
C.C. Sabathia's 2007: 243 IP, 232 H, 1.11 WHIP, 3.64 ERA, .309 BABIP, .241/.275/.362/.637/.391 (AVG/OBP/SLG/OPS/UVI)
We have a tie. Obviously, Sabathia deserves the Cy more than Bedard due to the higher IP total. Note that Sabathia's Expected AVG is 34 points above Bedard's, but his good control makes up for it. Also, note that Sabathia's Adjusted Line is actually better than his actual numbers (except for ERA).
Josh Beckett's 2007: 202.3 IP, 184 H, 1.11 WHIP, 3.60 ERA, .308 BABIP, .233/.274/.353/.627/.388
UVI-wise (and consequently ERA-wise; as you can see, they're in direct correlation), Beckett is just a bit better than Sabathia or Bedard. It appears Sabathia was about 1% worse in 15% more innings. You make the call on who deserves the Cy Young. Beckett, like Sabathia, gets an extra six outs in the Adjusted Line.
J.J. Putz's 2007: 66.3 IP, 53 H, 1.00 WHIP, 3.36 ERA, .312 BABIP, .210/.255/.333/.588/.371
Before I ran this, I saw the number ".201" under BABIP for Putz's actual statline, and I thought that my translation would make his '07 look bad. Turns out, he still winds up a cool .017 ahead of Beckett in the UVI column, even with 16 outs turning into hits. Unlike Santana, Putz has no track record of low BABIPs, so this seems like a pretty accurate look at how good his 2007 really was.
Joba Chamberlain's 2007: 22.7 IP, 16 H, 0.97 WHIP, 2.97 ERA, .320 BABIP, .190/.253/.273/.526/.330
Like Putz, Chamberlain clearly has plenty of value, low 2007 BABIP or not. His BABIP in this line actually goes up 91 points from his true '07, but hitters still bat just .190 off him. That's exceptional--how true it is given the small sample size has yet to be determined.
Jake Peavy's 2007: 217.7 IP, 186 H, 1.17 WHIP, 3.60 ERA, .311 BABIP, .222/.286/.325/.611/.381
Peavy wound up better than any of the elite AL starters. This should be no surprise given that he faces a lot of pitchers and pitches half his games at Petco Park. In UVI2, I would suspect he comes down. Not to say that Peavy wasn't a great pitcher last year. He loses 14 outs in this translation; his .279 BABIP was well below his career numbers.
Brandon Webb's 2007: 236 IP, 210 H, 1.19 WHIP, 3.51 ERA, .290 BABIP, .229/.288/.317/.605/.370
Is it heresy to say that the unanimous NL Cy Young winner didn't deserve it? Webb gives up one more hit in this line than he actually did, so his 2006 season is basically for real. As I mentioned before, Peavy has some luck attached, and Webb threw more innings for a better team. Which one sounds better now?
Aaron Harang's 2007: 230.3 IP, 217 H, 1.17 WHIP, 4.03 ERA, .314 BABIP, .239/.286/.389/.675/.427
Why Second-Order UVI Is Better Than First-Order UVI: Exhibit A. Harang's 28 homers are partly due to the Great American Bandbox in which he pitches half his games, so compared so someone like Peavy, he looks pretty bad.
Carlos Zambrano's 2007: 211 IP, 203 H, 1.44 WHIP, 4.62 ERA, .308 BABIP, .243/.334/.380/.714/.455
Zambrano, like Johan Santana, has a consistent history of low BABIPs, but of the ten pitchers I've examined so far, eight lines look pretty good, and this is just first-order stuff, so I'd say it's not bad. Zambrano's triple-digit walk totals certainly don't help him, and his GB% is in a downward trend. This line may be a bit of an exaggeration (I expect that in second-order UVI it looks a bit better), but it's not all rosy for No. 38 on the North Siders.
Takashi Saito's 2007: 60.3 IP, 45 H, 0.96 WHIP, 3.25 ERA, .304 BABIP, .199/.252/.314/.566/.360
The most underrated reliever in baseball. Period. Like Putz, Saito still has a better UVI than all of these top starters even after you increase his BABIP--97 points in this case. That he's doing this at 37 is remarkable.
In conclusion, it looks like with a decent sample size, .350 would be really close to the best you can do.
How Low Can You Go?
As with hitting, the max UVI is 4.000--which would occur if the pitcher allowed a HR every plate appearance.
It's tough to really pick out a line to show as an example of copious badness, because the translations change everything around. Just as an example, though, let's take Josh Towers' 2006, where he actual line features an 8.42 ERA. His adjusted line is as follows:
67.3 IP, 77 H, 1.40 WHIP, 5.48 ERA, .316 BABIP, .276/.324/.516/.840/.548
This tells us three things:
1.) Josh Towers was unlucky in 2006.
2.) Josh Towers still sucks.
3.) It looks pretty impossible to put up a UVI upwards of .600 without getting sent down.
So, as with hitters, average seems to be in the .460-.470 range, with the overall range of UVIs looking like .350-.600.
Note that since all this is adjusted, there is likely to be less of a standard deviation. You will find in the piece on second-order hitting UVI that several batters top the .600 mark, although .350 seems to be about the floor there as well.
More UVI studies will be up shortly.