September 26, 2003
Take it to the Bridge
Into the mailbag we shall go...
Hi, Gary. I heard you on the radio, and you said that in your opinion, one of the first general managers that should be let go is Pat Gillick in Seattle. How can you possibly say that? The Mariners are a perennial powerhouse in the AL, won 116 games just two years ago, have the best pipeline of talent from the Pacific Rim, and a stocked farm system. He's done everything that a general manager could possibly do. What possibly justification could there be for firing him, and who's available that could do the job better?You make a good case. However, I still think that the Mariners would be better served by a different GM, with Gillick taking a more focused role within the organization. (I didn't advocate Gillick's termination...) Gillick has a tremendous resource base to work with, and the results he's delivered simply haven't been adequate. Seattle's looked haphazard, both in terms of player acquisition, and, more importantly, in terms of talent utilization.
It's not that it's a bad thing to have Bret Boone or the overrated Ichiro Suzuki around; the real damage is done when you pick up a bunch of relatively crappy role players for too much money, then ask them to perform duties that are beyond their capabilities. Check out the role players on the Seattle bench...
Player VORP 2003 Salary ------------------------------------------ McLemore, Mark -0.9 $ 3,000,000 Wilson, Dan -5.9 $ 3,500,000 Cirillo, Jeff -9.5 $ 6,425,000 Davis, Ben 4.2 $ 1,000,000 Mabry, John 0.9 $ 600,000 Bloomquist, Willie -4.5 $ 300,000 Sanchez, Rey 4.8 $ 1,300,000 Colbrunn, Greg 2.8 $ 1,750,000 Sum -8.1 $17,875,000And really, this is pretty generous. Some of these guys ended up in part time roles because they were ineffective, and in certain obvious cases, such performance should have been expected by any diligent observer. In fairness, the Colbrunn deal's been unlucky because of an injury. But even if you observe these decisions at the time they were made, these were bad decisions.
Think about the implications of this. The Mariners basically gave up a game to replacement level, and paid nearly half of the salary gap between them and Oakland to do it. That's brutal. It's pretty much the opposite of a high-OBP front office; expending resources to dig yourself a hole. At the end of the day, the GM has to balance a number of limitations and restraints, and give the team the best chance to succeed. In order to do this, a GM has to understand the value of a player on the field, and understand how much that's worth, in terms of financial and logistic resources. That's the core of the GM's job, and Gillick hasn't done it well enough to retain his position, in my opinion.
Dear Mr. Huckabay:
Got a lot of mail about that, split approximately 50/50. Half the people thought I was bringing politics to BP, the other half were grateful for the shining of some light on the Ueberroth administration. I certainly don't mean to introduce politics to Baseball Prospectus. I thought I was doing a pretty reasonable job of focusing strictly on what can only be described as Ueberroth's awful job as commissioner. I'll grant that it was the media's kid glove treatment of Ueberroth that prompted the article, and I made a big mistake in including the final line of the column, which really did shift the apparent focus of the piece.
Or, to sum up, my bad. Some days, I can't write. (Simpsons fans: "We call those 'Fridays.'")
Hello Gary! In your chat session, you stated that you thought Keith Woolner's research into replacement level was the most important work in sabermetrics. Why is it so important? I think that your work on PAP is much more important. It can change the way teams handle their pitchers. The pitchers will be healthier, and the teams will be better because of it. Why is replacement level more important than that?
First off, let me clarify something. Pitcher Abuse Points was a system developed by Keith Woolner and Rany Jazayerli, not me. And if you review it, you'll find a curve fit that you'll be lucky to find again in your whole life.
The answer's pretty straightforward: replacement level is essential to know because it's the only way you can accurately assess marginal value. Let's say you can sign Joe Slugger, who's likely to play a pretty good corner outfield spot and post .280/.360/.550 each year, for $6,000,000 annually. Is that a good deal? There is absolutely no way to tell unless you know what your options are--you need to know the marginal value of that player's production, and for that, you have to know what your replacement options are. Or, put another way, you have to know replacement level.
In the example provided, if you can't swing a dead cat without hitting corner OFs that hit .270/.350/.500, and are available for the league minimum of $300,000 annually, then signing Joe Slugger isn't a particularly compelling idea. On the other hand, if teams are scarfing up the Kerry Robinsons of the world to start in their outfields, the calculus might well go the other way.
It is simply not possible to make an informed decision about the value of a player unless you know the value of the available options.
I can't believe this is the same guy that went to Ralston School and went to the 8th grade dance with Cheryl Thomas and hung out with Steve Pieraldi and Joe Donahue. If you're the Detroit Tiger GM, what steps do you take, and in what order?
I really hope you don't look like Kathy Bates. And you misspelled Joe's name--it's Donaghue.
On a general level, the steps for the Tigers aren't really that different from the steps for any club. There would be worse places to start a career as a GM than Detroit. Think of the freedom you have to experiment or execute a long term plan. Detroit fans are solid; they'll support a winner, and if you can put together a team that looks pretty good, with a young nucleus, and win 75 games in 2005, you're going to get some well-deserved good press. But onto the question...
Then, start the hard work. Plan for player acquisition. That means the draft, it means Rule 5, it means free agents and castoffs. Figure out the best risks and aggressively take them--it's not as if you have a hell of a lot to lose. Buy low, develop, and trade/let go as necessary to build your resource base. Go sign a Graham Koonce and give him the ABs necessary to see if he can play. If he can, figure out if he's going to contribute to the next contending team. If he is, great. If not, he's inexpensive and productive, which should allow you to go get some young players with actual talent. If he doesn't work out, you're not really any worse off except for some small lost opportunity.
It's more interesting, in my mind, to come up with a list of things you shouldn't do:
Blasphemous as it sounds, teams need to sometimes slow down and say "We're not as good as we think." It's difficult to do, but failure to do so can result in a nasty cycle of mediocrity, like the Tigers before 2003, or the Mets and Orioles of recent years.