Sunday, January 23, 2005

Marginal payroll per Marginal win

I was inspired by the 25+" of snow to break out the Baseball Prospectus. I guess when you're really far into winter it's only natural to think of summer.

Anyway, I was interested in learning more about the Mets. Clearly, with their recent moves they are making a big push for the WS this year. That being said, BP04 did a nice review of the margin payroll per margin wins from '86 to '03. Basically, if you haven't read the piece, they compare the teams marginal payroll, Team X's payroll at start of season minus league minimum time 28. Basically, take a 28 player team of John Does and whatever payroll is over that is your marginal payroll. Then they take your winning percentage minus .300 and multiple by 162. Basically, they assume a team of John Does will have a winning percentage of .300 and multiplying by 162 evens out shortened seasons and whatever. The idea being if you have a lot of wins for little money you're efficient. If you have a lot of wins and pay a lot of money you are the Yankees. If you have few wins for a little money you didn't spend enough and if you have a few wins for a lot of money you "paid a lot for that muffler" - not the Midas touch.

Anyway, here are a couple observations:

1. Mets 2003 MP/MW $6.1 MM! The highest in the league and the second highest absolute value. Detroit's MP/MV of -$7.3 MM is more insult to injury

2. Red Sox 2003 MP/MV $2.0 MM puts them at 16th best in the league, right dab in the middle. The best being Oaklands MP/MV $883,140. In addition, many of the teams ahead of the Red Sox in MP/MV obviously didn't make the post season so there is something to be said about compromising on $ for wins. (More evidence that a combined money-ball/yankee-ball strategy is effective).

3. The Mets are $2.0 MM higher than any other team. That being said, they did have a terrible year spoiled by a lot of injuries, if I remember correctly. Regardless, they are adding a ton of payroll this year so they have put even more pressure on the club to win. If Pedro is really capable of winning at the rate he has for the past couple of years and Beltran adding a little offense (but not much) are the Mets really putting themselves in a better position?

This post really doesn't have a point, other than the fact that I marvel at the willingness of the Mets to spend and spend and spend for no return. If their recent moves don't work out do you think is spells a change of the gaurd in the Mets' front-office? Do you think they will become a "standard" of what not to do for teams to come?


  1. Yeah, the cost-per-win thing is pretty cool -- apparently it's what got Michael Lewis interested enough in the A's to write Moneyball.

    Do you have the yearly stats for all teams? I'd be interested to see how the Rangers did during and after the Park-and-ARod the Yankees playoff teams typically do...etc.

    XL spreadsheet!!!

  2. Damn! They only have stats going back to 95. But I like the Texas question. First Arod joined Texas in 2000. So leading up to 2000 here is Texas' MP/MV:

    1997: $1.6 MM (77 Wins)
    1998: $1.2 MM (88 Wins)
    1999: $1.5 MM (95 Wins)
    2000: $2.9 MM (71 Wins)
    2001: $3.4 MM (73 Wins)
    2002: $4.3 MM (72 Wins)
    2003: $4.2 MM (72 Wins)

    Wow! I too would have thought adding the "best player in baseball" to my 1999 squad would have improved my chances of making the playoffs. Makes me wonder if A-rod has a problem relating to his teammates. Maybe he doesn't mean to, but his reputation could alienate players and cause them to underperform. I doubt this will happen in NY because they have a lot of big confident egos.

    Chan Ho Park Joined the Rangers in 2002, which might explain the $1 MM leap in MP/MV since the Rangers had the same number of wins in both 2002 and 2003.