Graph of the Day: David Wells vs. Mariano Rivera

by Jack Moore on Jun 30, 2009 4:50 PM EDT Comment 12 comments

I read this story on MLB Trade Rumors earlier today, and was very intrigued. Here's a snippet -

Former Yankees GM Gene Michael told John Harper of the New York Daily News that he considered dealing Mariano Rivera to the Tigers for David Wells back in 1995. Rivera was still considered a starter then, and he was one of the players the Tigers had interest in. Michael, now a scout, says he held onto Rivera because of his control and mid-nineties fastball.

Of course, this trade initially seems completely ridiculous. Rivera is the best closer of all time, and while David Wells had some good seasons, he's certainly not on the level of Rivera, right? Well, here's the career path for these two pitchers.

As you can seee, the best 10-12 seasons for these two are very, very similar. Although Wells never put up the huge numbers of most hall-of-fame starters, he was consistently well above average. Rivera simply could not put up anything above 5.5 wins due to not throwing enough innings. To truly see how even this trade could have been, let's look at their career year-by-year.

The similarity between the two from 1995 to 2005 is striking. Hopefully this gives some idea of the worth of starting pitchers compared to relievers.

----------

WAR data from Rally's WAR database at www.baseballprojection.com

0 recs | Comment 12 comments | Share

Story-email Email Printer Print

Comments

Noobish question here possibly.

wouldn’t Rivera pitching in higher leverage situations lead to him being even a little more valuable than WAR would suggest? Or does WAR take that into account (with an adjustment or something, i’m actually not completely sure how WAR is calculated)?

by Lum on Jun 30, 2009 5:45 PM EDT reply actions 0 recs

WAR considers leverage in calculating relievers

It might actually go a little too far in incorporating leverage. I don’t recall whether it handles the chaining of relievers correctly (when you lose your closer, his high leverage innings don’t go to a replacement level pitcher, they go to the next guy in the chain).

by Dan Turkenkopf on Jun 30, 2009 5:48 PM EDT up reply actions 0 recs

The way Rally calculates it

Pitchers are credited with extra leverage halfway between their actual leverage and 1.0. This is due to the effect of chaining, a relief ace who normally has a leverage of 2.0 and is injured. He will be replaced by a setup man, who is normally an above average pitcher himself, and all pitchers will move up into higher impact situations. The net effect is less than if we assume that the relief ace is directly replaced by a replacement level pitcher.

straight from his stat definitions page

http://www.baseballprojection.com/war/statdef.htm

---
http://www.beyondtheboxscore.com
http://www.rightfieldbleachers.com

by Jack Moore on Jun 30, 2009 6:47 PM EDT up reply actions 0 recs

Thanks I couldn't remember the details and didn't realize he had a description of the method up.

by Dan Turkenkopf on Jun 30, 2009 6:50 PM EDT up reply actions 0 recs

Well, it can certainly still be debated.

But it’s also certainly better than just straight up LI

---
http://www.beyondtheboxscore.com
http://www.rightfieldbleachers.com

by Jack Moore on Jun 30, 2009 7:31 PM EDT up reply actions 0 recs

I would prefer a combination of their actual LI and their deLI

Regressing to average doesn’t seem to plausible in this situation.

Derosa.

by vivaelpujols on Jun 30, 2009 11:49 PM EDT up reply actions 0 recs

It's not really regressing to average.

I mean, it is mathematically, but that’s not the reasoning behind it.

The formula is a shortcut for taking into account chaining. If you take a typical bullpen and remove the closer, bump everyone else up a position in LI, and throw a rep level reliever in the least important spot, the the WAR of the closer turn out to be pretty close to his non-leveraged WAR times (1+LI)/2.

Definitely room for improvement, but it’s pretty good.

Beyond the Boxscore Not a member? Sign up.

by Sky Kalkman on Jul 1, 2009 8:46 AM EDT up reply actions 0 recs

Seems like there would be an argument to be made for chaining starters...

especially on a team such as the Yankees. They won’t start a replacement level guy for very long, so innings eating seems to be overrated if we look at it in this way. There are a lot of ways to get David Wells’ production, but a dominant guy like Rivera is far more rare.

by Canseco's Roid Party on Jul 2, 2009 2:32 PM EDT up reply actions 0 recs

What about if a bunch of good to great starters became relievers?

I’m inclined to think it’d still be tough to post a 2.00 ERA, but there are a ton of starters who could be really damn good in the bullpen, but are starting instead of relieving. Do hypothetical performances make Mo’s performance any less rare? Are +20 run fielding seasons at 1B made less impressive because we know many shortstops and some 2B/3B could do it if they were moved to 1B?

Beyond the Boxscore Not a member? Sign up.

by Sky Kalkman on Jul 2, 2009 5:17 PM EDT up reply actions 0 recs

Chaining starters is less important I think

Because they all tend to have roughly the same leverage. So when replacing the starter with a theoretical replacement level pitcher the replacement innings are of the same importance as the replaced ones.

That’s not the case for relievers.

If you’re suggesting that because of the resources at a team’s disposal the actual replacement level is higher than “true” replacement level, that’s a possibility, although it’s probably less true than most might think. Above replacement level is a lot harder to come by during the season than during the offseason, and has associated costs. Remember the Yankees ran Small, Chacon, Redding and Ponson out there in the past few seasons.

by Dan Turkenkopf on Jul 2, 2009 6:41 PM EDT up reply actions 0 recs

If David Wells and Mariano Rivera were a package ticket for HoF voting, would they get in or not?

Beyond the Boxscore Not a member? Sign up.

by Sky Kalkman on Jun 30, 2009 10:36 PM EDT reply actions 0 recs

I'd say yes-

Wells alone doesn’t have enough good seasons or any short stretch of seasons of total dominance, but he was always a plus pitcher and his control was consistently excellent. Rivera will get in on the first ballot alone, due largely to the overvaluing of saves. Even so, he is the greatest relief pitcher of the current era by a wide margin, the greatest among the “only for saves” usage pattern. The two were an imposing combination with very low walk rates and alot of K’s. If they were one person it would be a no brainer.

by Slugger O'Toole on Jul 1, 2009 9:05 AM EDT reply actions 0 recs