“The [Florida] Marlins have spent about $396 million on player salaries from 2000 through 2010, with 873 victories and a World Series title to show. The [New York] Mets have spent about $1.212 billion on salaries in those seasons, with 878 victories and no championships.

In other words, the Mets have spent about $816 million more than the Marlins to win five more games and one fewer title.”

(Source: NY Times, 25.08.2011)

12 Responses to “Factoid of the day: Moneyball Redux”

  1. Sara Davis says:

    Those numbers are meaningless without a profit statement. They are in business to make money. Did the expense provide larger profitability for them?

    I have no idea and don’t care; but only the fans care about the wins and probably would allot any expense to achieve it. The investors want the profit/expense ratio.

    Thus the frustration of professional sports. Where is the priority Profit or Sports?

  2. Dan Pink Dan Pink says:

    @sara —

    Good questions — both about these teams’ profitability and about whether winning or profits (or resale value of the club) is the goal. Meantime, those same documents in that Deadspin link show that the Marlins are pretty profitability: http://www.miamiherald.com/2010/08/24/1789068/leaked-data-show-strong-profits.html

    Yet by another measure, the Mets are the third most valuable team in baseball: http://newyork.mets.mlb.com/news/article.jsp?ymd=20100407&content_id=9124692&vkey=news_nyy&fext=.jsp&c_id=nyy

  3. Michael Johnson says:

    This is an example of what my PhD adviser used to call “death by anecdote.” These two teams represent rather extreme deviations from the norm. I just ran the correlation between team salaries and current team record for the 2010 season: it is .30. Not a huge correlation, but all else being equal, higher paid teams win more games.

    In a highly complex system like Major League Baseball, many factors affect team performance. I guess that’s why they say the game is played on the field and not on paper.

  4. Dan Pink Dan Pink says:

    @michael —

    excellent. point. and thanks for running that correlation. the figure out came up with (0.3) surprised me a bit. i expected it to be a tad higher.

    that said, there are players like our own stephen strassburg (with very high salaries and lots of promise for later years) who might confuse things a bit.

    interesting issue.

  5. Juliosus says:

    I’m not much of a baseball fan, but this conversation brings to mind why more and more teams are subscribing to the Bill James sabermetrics method of using observable statistics to make personnel decisions. I imagine it’s no coincidence that the Red Sox hired James in 2003 and two World Series followed not long after.

  6. Michael Johnson says:

    Dan’s speculation that the correlation should be higher made me curious as to whether this year is different than others. And he’s right: the correlation between total salary and total wins from the 2008-2010 seasons is currently .52.

    The Marlins got the best value from 2008-2010, winning 233 games on a total salary of $114,287,000 (in other words, each win cost them $490,502.15 in player pay). The Yankees actually got the worst value for the same period, winning 270 games on a total salary of $616,864,155 (each win cost them $2,284,682).

  7. Dan Pink Dan Pink says:

    @michael — very interesting. and “costs per win” is a fascinating metric. i wonder if front offices use it.

  8. Payroll and wins do have a positive correlation of about 0.2, which is certainly not strong. If you draw a scatter plot and a regression line you will find some teams above, like the Marlins, some teams below, like the Mets, and some teams on the line like the Yankees. Let me know if you want to see some data.

  9. Bill Petti says:

    I think the focus on money distracts from the real insight from Moneyball. The lack of resources was a precipitating condition that forced Beane and the A’s to find a novel way to compete against better funded teams. What they did was find a better, more efficient way to win. Their analysis, building on James, showed that the variables that had the greatest impact on wins (e.g. OBP) were undervalued by most teams and therefore players with those attributes could be a acquired for below-market prices. This worked out great for a team that wanted to compete for wins, but had to do so with a limited budget.

    It is the analysis–the looking for a better, more efficient way to “win” and not being afraid to buck conventional wisdom –that, for me, is the real takeaway of Moneyball.

    As better funded teams realized that attributes like OBP were highly valuable and under priced they began to spend higher amounts of their budget on them, which likely accounts for why the correlation between wins and salaries remains high. Think about the Yankees and Red Sox. They are still winning a lot, but for different reasons, and that maintains the high correlation even though they are essentially playing “Moneyball”.

  10. And the Marlins keep it exciting each year. Will it be a year of random players after a fire sale or a championship contender? Who knows! 🙂

  11. Evan says:

    I agree with Sara… what about profitability?? What if a club spends on the type of players that bring out more fans and sell more Jerseys? Being a met fan, I doubt that was the case here – but to illustrate – remember that owner who signed a midget as a publicity stunt? Unlikely those dollars bought that club more W’s ( Might have helped the mets though) A follow up post comparing the profitability of the said clubs over that horizon in requested.

  12. Richard says:

    Marlins tend to be feast or famine. They hibernate stock up on young talent allow it to blossom as a group and if it does they succeed, sell off assets and go back into hibernation. For small market teams this becomes the best strategy to occasionally create a winner. It also creates a fickle fan base (customers). Money ball ballparks have plenty of empty seats. As I write this Tampa is in a race for first place loaded with talented young inexpensive players playing in front of 18,0000 in a 40,000 seat ballpark, and they recently went to the World Series….

Leave a Reply