That’s what we’re spinning today here in the studios of WXRL. You’ve never heard of it? Must be some newfangled Internet radio podcast thing, right?
Sort of. WXRL is one of those statistics that make Bruce Jenkins’ head explode as he chases the nerdhats with calculating machines off his bucolic greensward of yesteryear. It’s definitely an Internet thing, because you can go to Baseball Prospectus, the secret-password stathead basement clubhouse, and look it up, sort it, tab it, LOVE IT.
I speak today of WXRL because Randy Messenger gave up some bad hits to the bad Dodgers last night, and the Giants lost the game. Still, Randy Messenger has a 2.75 ERA, which is very very nice. He must be very very good, no?
The problem is, ERA doesn’t really tell the story of a reliever’s season. Here’s an extreme example: Joe Loogy comes in 20 times with the bases loaded and two outs. All 20 times, he lets all three runners score, but no runs are ever charged to his record. What’s more, in five of those games, his team takes the lead the next inning, and Loogy is credited with the win. That’s a 5–0 record with an untarnished 0.00 ERA in 20 games. Wow!
If we had watched all 20 games, we’d see past the Potemkin village of his numbers, but what about a less extreme example? Say, a guy with a 1–3 record and a 2.75 ERA in 46 games? How do you measure a reliever’s effectiveness at preventing baserunners from scoring in high-pressure situations?
Ladies and gentlemen, I give you WXRL. It stands for “Win Expectation above Replacement, Lineup-adjusted,” which means it gives extra props to guys who enter in a crucial situation and retire, say, the 3–4–5 hitters with the bases loaded. It gives more credit for one-run saves than for three-run saves. It measures all relievers, be they closers, set-up guys, or middle relievers.
(For a discussion of WXRL versus other advanced relief-pitcher stats, click here.)
The top ten in WXRL this year: Putz, Saito, Wagner, Pena, Okajima, Betancourt, Papelbon, Hoffman, Neshek, and Nathan. Note four of the ten (in boldface) are not closers.
The stat is measured in how many wins a player has contributed to his team. Putz has given the Mariners 5.2 wins, Nathan has given the Twins 3.3 wins.
The highest-ranked Giant is Brad Hennessey with 1.7 wins, good for #48 on the list. The worst Giant reliever is Randy Messenger. In his 28 2/3 IP with the Giants, Messenger is in negative territory; he has cost the team a sliver of a win (-0.03). It’s hard to reach negative numbers in this stat; most on the list are at least at the neutral zero or above. (In his 23 2/3 IP with Florida, Messenger added just 0.15 wins.)
This doesn’t mean he won’t improve, but it suggests right now he’s not the guy to bring in when you need a strikeout to preserve the game.
I love stats like these, although I don't really think of them as predictive in the way that K's, walks and homers are. (Unfortunately, Messenger's strikeout rate refuses to catch up to his stuff.) They're great for award voting and otherwise reviewing a season or a career, though. For instance, it seems pretty clear to me that no other reliever should receive more Cy Young votes than Putz, who is well above every other reliever in WXRL.