Rumbling, Grumblings, and Bumblings: Mauer, Stark, WARP, and Platoon Splits

by devil_fingers on Apr 10, 2009 1:30 PM PDT in Sabermetrics Comment 13 comments

Some Friday thoughts while pondering how the Royals will get by without Jose 'Domincan Joe Carter' Guillen for the next few weeks.

There's nothing to add about Nick Adenhart's death,other than to say how sad it is that three people were killed and a fourth seriously injured, and to send thoughts and prayers out for the families.
I thought of doing a "small sample size fun" type of first week column, replete with "the Yankees are doomed" jokes and so on, but why do something that other people do better... yes, I realize that taken to the full extent, that means I would quite blogging...
Here's something interesting: Well, sort of. If you're a nerd. Jayson Stark discovers WARP! It's nice that BP is getting more mainstream play. At the risk of killing any potential chance I might have of winning BP Idoll (if I can get it together to enter), I find it really appropriate that Prospectus is more visible than ever, and that even ESPN writers are using their stuff.... at least a year after every baseball nerd on the internet stopped looking to them for cutting edge stuff. So it's perfect that Stark uses WARP, a stat that is so broken (it's replacment level is ridiculously low, and the defensive stat, FRAA, is, um, not great) that BO actually bothered to respond to the critics and (began) to fix it. 9.6? That seems high. I looked on Mauer's Davenport Translation Card... For 2008, his WARP1 is listed as 8.2, his WARP2 as 9.3, and his WARP 3 as 9.2. I wonder where Stark got 9.6? I won't know, seriously. Did they fix the WARP on the DT cards yet? I thought 9.6 sounded high, and FanGraphs has Mauer at 5.9 in 2008... but they don't use defense for catchers. Rally has Mauer at 9 WAR, though, so maybe BP's on to something. Rally includes catcher defense (+12 for Mauer) baserunning (Mauer as +2), GiDP (Mauer is 0 -- average). One big difference is the batting runs above average. FanGraphs has Mauer at 27.5, while Rally has him at 45. Rally adjusts a player's contribution according to team offensive levels, so the Twins ridiculous number with runners on last season may be a big part of the difference. Thoughts?
A request: I'm working on a project on trying to project hitter platoon splits for 2009. I was hoping that readers would give me some ideas for some hitters to use from around the league -- Im' looking particularly for lefty and righty hitters (switch hitters are interesting, but in a different way). I have somem ideas of my own, but getting contributions in terms of what players to watch would make it more interesting for everyone. I'm looking for a variety of guys -- lefties and righties, guys with exceptionally large spltis, no splits, reverse splits... Maybe just in recent years. I want to see how much they regress to the mean.

Have a nice holiday weekend, if you so choose!

0 recs | Comment 13 comments | Share

Story-email Email Printer Print

Comments

I like WARP? What are the flaws with it? What is better?

by backtocali on Apr 10, 2009 8:48 PM PDT reply actions 0 recs

DF's link to Tango's blog post explains a lot.

Their replacement level was off by a few wins. This is…kind of large given the standard data range.

by jwiscarson on Apr 10, 2009 10:40 PM PDT up reply actions 0 recs

yes, follow the link

WAR (Wins Above Replacement) as found on the FanGraphs player pages or Rally’s historical WAR site are far superior on a number of levels: replacement level, positional adjustments, offensive statistics, etc.

Bringing you more-or-less replacement level analysis and commentary to Driveline Mechanics and elsewhere since sometime in 2008.

by devil_fingers on Apr 11, 2009 9:56 AM PDT up reply actions 0 recs

The defensive stat that they use

FRAA, is basically advanced range factor, making it light years behind UZR/TZ. Also, their replacement level is way too low.

St. Louis Cardinals... defying win expectancy since 2008

by vivaelpujols on Apr 12, 2009 1:11 AM PDT up reply actions 0 recs

Some Rangers suggestions for your platoon splits project:

Hank Blalock (L), Ian Kinsler®, and Marlon Byrd®.

Hank has a pretty serious split over his career, but Kins and Marlon are much closer. I’d suggest Saltalamacchia if you were analyzing switch hitters. It might be worth following Chris Davis (who has reverse splits, albeit over a small sample size), but word on the Ranger grapevine says that he may be sent down to the minors if he can’t cut down on the Ks (five in 13 PAs thus far this year).

by jwiscarson on Apr 10, 2009 10:48 PM PDT reply actions 0 recs

Wow, those registered symbols should really be an R in parentheses.

by jwiscarson on Apr 10, 2009 10:50 PM PDT up reply actions 0 recs

thanks

hopefully others will jump in, too

Bringing you more-or-less replacement level analysis and commentary to Driveline Mechanics and elsewhere since sometime in 2008.

by devil_fingers on Apr 11, 2009 10:05 AM PDT up reply actions 0 recs

Ludwick

Ryan Ludwick has got a reverse platoon split going on.

by Laura on Apr 11, 2009 11:52 AM PDT reply actions 0 recs

Ian Stewart has been hitting LHP's better than RHP's going back to 2007

if you look at his minor league splits. He may be an interesting one to look at.

by Rox Girl on Apr 11, 2009 12:07 PM PDT reply actions 0 recs

I'll check it out -- that's a great example,

and he’s interesting in general, not just because of his prospect status, but because he’s making the 3B to 2B switch.

If I don’t include him, it’s only because I’m not sure he’ll have enough PAs in the majors for my crude methodology (I’m not comfortable incorporating MLEs at this point, especially on this topic) for it to be very helpful.

Bringing you more-or-less replacement level analysis and commentary to Driveline Mechanics and elsewhere since sometime in 2008.

by devil_fingers on Apr 11, 2009 3:14 PM PDT up reply actions 0 recs