I like wOBA for hitting. I think it does the best job of balancing getting on base and hitting for power. As far as pitching goes, I like xFIP the most because it’s FIP but it normalizes the homerun rate. As the saying goes: “pitchers give up flyballs, but hitters turn them into homeruns.” Defensively I look at UZR but I strongly believe that there’s no stat yet that can perfectly explain defense. Which is part of the reason why I don’t completely trust FIP either.

Offense: VORP. Pretty all inclusive, adjusts for position, era, park, simple, easy to give a rough conversion into wins. Very accurately researched.

Pitching: I’m not a fan of any one stat yet. Pitching is still a relative menagerie of things in my mind. You have to look at various rates, GB’s, BIP data. I have a major problem with 3-true-outcomes numbers because pitchers DO have an affect over the number of grounders and pop ups and things. If we can get a FIP and add the 4th outcome – ball in play data – to it (ie, adjust for their number of line drives given up compared to the average) then I’d be very happy with the metric.

Defense: Since none of the pitching ones really stands on its own to me, I’m taking two (and sorta 3) fielding ones, oddly enough. I like UZR (paired with the out of zone plays number) for fielding, it seems very well measured and matches good observation the most consistently (and doesn’t seem overly variant form year to year like some fielding stats). I also very much like defensive efficiency to judge team defense (which is linked into judging pitching then). It isn’t perfect since it doesn’t take into account the skill of the pitching staff, but it’s a pretty good look.

I like OPS+ because it adjusts for league averages, so you can compare how Crede is faring against, say, Gaetti.

I like UZR/150 because it tells us how a guy is doing right now in terms of how he would end up if he doesn’t improve or regress.

Pretty important stat.

I’ve never actually seen any data regarding exactly how accurate FIP, xFIP, and the like are at projecting pitchers’ performance. Supposedly they are better than just using ERA, but I’ve never seen any figures of exactly how much better. Anyone know any more about this, or where to find it?

Offense: I like BIP—LD%, GB%, FB%, HR:FB, etc. My second choice is OBP, although wOBA is pretty great as well.

Defense: UZR, or UZR/150. Fielding Bible’s +/- system is also a phenominal, and phenominally simple, system.

Pitching: It’s all about TTT for me. HR, K, BB. Sure, you can be a fly-ball pitcher, which works against you, but that doesn’t make the TTT system any less accurate. Ultimately, if you’re on the right side of average in all three of these categories, you’re going to be a good pitcher.

I don’t know if there are any places that actually calculate this, but I think a valuable stat would be the the ERA of catchers when they are calling games.

For stats that actually exist.

Offense: I like OPS, but lean more towards the OBS side vs the SLG side it. ie. if two players have the same OPS but one has a higher OBS than SLG I would rather have the OBS on my team.

Defense: I am not really sure, UZR seems ok, but its new to me.

Pitching: I like WHIP a pitcher can get lucky and get out of innings, but if a WHIP and ERA are drastically different something is up.

wOBA, UZR and FIP (would be xFIP, but it’s not easily available on fangraphs).

I’d prefer to break down everything based on runs (baserunning, moving the runner over, etc.), but right now I’m stuck on defense, as well as having three under 5 year old kids in the house…

WPA for relievers is a great one. WPA is a “neat” stat in general but not necessarily one Mike on Tilt is right on one several counts. I realy like WPA for relievers. In general, WPA is “neat” but I don’t see it as particularly predictive, but for relievers who often come in with two outs and men on, it’s more worthwhile valuable. I always like to look at their strand rate, too, which at extremes tends toward luck. Both those things are in some ways captured by xFIP, my hands down favorite for pitchers in general, since it indirectly incorporates GB rate and eliminates a(n at least mostly) luck-fueled variable — HR/FB.

Regarding hitting metrics, here’s a great series if you missed it:

Mike also nails the fielding question: run too far with any one given zone type number at your own peril. When they agree, though, it’s difficult to argue.

There was a great article recently at the () putting forth a new metric the guy calls Simple Zone Rating. The kicker was that Derek Jeter was found to be the all time second worst infielder in terms of total number of plays made below average, which is probably why the article stuck with me and why I’m pointing it out here.

my links didn’t work. they are (cut and paste, I guess):
http://www.hardballtimes.com/main/article/the-great-run-estimator-shootout-part-1
http://www.hardballtimes.com/main/article/the-great-run-estimator-shootout-part-2
http://www.hardballtimes.com/main/article/fleeter-than-birds-part-1/