Wrapping Up Win Values

So, we’ve walked through the pitcher win values formula, and they’re now available on the leaderboard and the team pages going back to 2002. As a wrapup, let’s take a look at some final win value housekeeping notes.

If you sum up the 2008 win values for the position players and the pitchers that we have here on the site, you’ll notice that we’re handing out 583 wins to position players and 445 wins to pitchers. That’s 1,028 marginal wins – 57% of them are earned by position players and 43% of them are earned by pitchers. You know that whole “good teams are built around good pitching” thing? It’s bunk. Position players are more valuable than pitchers.

With a league wide total of 2,430 wins available every season (81*30) and 1,028 of those being distinguished as wins above replacement level, that means that the average team is 34.26 wins above replacement. 81 – 34.26 = 46.74. That’s what our numbers say a true talent replacement level team would have won in 2008. Just shy of 47 wins, or a .289 winning percentage, is the replacement level we’re using. It will vary slightly from year to year, but the .290 to .300 win% range is what is generally accepted as replacement level.

Over the last four years, Oliver Perez has been worth a total of 2.5 wins in nearly 600 innings. And he wants $12 million a year for five years. Maybe he’s the left-handed Gil Meche, and he’s going to turn the corner immediately after signing a big contract, but there’s about a 5% chance of that being true and about a 95% chance that he’s the most overrated pitcher in baseball.




Print This Post



Dave is a co-founder of USSMariner.com and contributes to the Wall Street Journal.


14 Responses to “Wrapping Up Win Values”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Rob says:

    I do find it funny Oliver Perez wants $12m while Odalis Perez who projects better might have to settle for an NRI.

    Vote -1 Vote +1

  2. CajoleJuice says:

    To be fair, I don’t think it counts as overrated when it’s only you and your agent. Any sane Mets fan will tell you that they couldn’t stand another season of him.

    Vote -1 Vote +1

  3. CajoleJuice says:

    P.S. Hopefully, Omar Minaya is a sane Mets fan. Recently, it looks like he is.

    Vote -1 Vote +1

  4. SleepyCA says:

    “Position players are more valuable than pitchers.”

    Is that realluy true? The typical team has 56% position players and 44% pitchers (assuming 11/14) or 52% position players (assuming 12 pitchers).

    Vote -1 Vote +1

  5. According to THT and Baseball Prospectus, it is not bunk, pitching and defense is the key to success in the playoffs.

    Taking your methodology, it would be interesting to see what the proportion of position to pitcher Win Value is for winning teams vs. losing teams, to see if that yields any further information on their difference.

    Also, looking at the values, roughly there is 8 main position contributors, or an average of about 7% per position player, while for pitchers there are 6 main contributors, the 5 starters plus closer, which also works out to about 7% per pitcher.

    Vote -1 Vote +1

  6. Librarian says:

    I’m no sabermetrician (or mathematician), but this makes sense:
    “The typical team has 56% position players and 44% pitchers (assuming 11/14) or 52% position players (assuming 12 pitchers).”
    And if the starting pitching staff composes 20% of the roster, does that make SPs more valuable?

    Vote -1 Vote +1

  7. Mike A. says:

    “…57% of them are earned by position players and 43% of them are earned by pitchers. You know that whole “good teams are built around good pitching” thing? It’s bunk. Position players are more valuable than pitchers.”

    Isn’t that backwards? If good hitting is more abundant, wouldn’t it make sense that good pitching is more valuable since it’s not as plentiful?

    Vote -1 Vote +1

  8. David says:

    I think if you do an analysis like this:

    1) contribution of your “average” outfield player in terms of wins
    2) contribution of your “average” starting pitcher in wins
    3) contribution of your “average” relief pitcher in wins

    2) would be somewhat higher than 1), which would be significantly higher than 3). Even at first glance, this would seem to be true. Just eyeballing the list of starters with >100 IP, your median starter ranks at something like +2.3 wins. Your median reliever (>40 IP) is like +0.5 wins, while your median outfield player (>300 PA) is like +1.9 wins. (the arbitrary cutoffs are just to make the median a meaningful statistic)

    Now mathematically that isn’t quite accurate because, obviously, because the median isn’t going to be equal to the mean. But I can’t imagine it’s far off.

    Vote -1 Vote +1

  9. David Pinto says:

    How does the position players break down offense/defense? What % is offense, and what percent is defense. Win shares puts the pitching/defense percentage over 50%.

    Vote -1 Vote +1

  10. philosofool says:

    Well, as an argument that I just totally pulled out of my ass and I have no idea whether it stands up on analysis, defense and offense are equally important. (That premise is supported by the pythagorean wins formula.) Defense is 75% pitching and 25% fielding defense (this is the part where I’m really making shit up), while offense is 100% position player in the AL or 89% position player in the NL, and all pitchers hit about the (terrible) same. If that’s right, then you’d expect position players to account for about 12.5% more of the wins than pitchers. Oh look, my little possibly bullshit argument produces the right results. I leave it to someone knowing a shit ton more about this stuff than me to figure out whether the fancy argument that agrees with Dave’s conclusions has any merit.

    Vote -1 Vote +1

  11. tangotiger says:

    Over the last decade, pitchers got around 42-43% of a team’s payroll. So, there’s a certain logic that they are getting 42-43% of the team’s wins, not to mention around that many number of roster spots.

    The off/def breakdown has to come in at exactly 50/50 by design, and any deviation from that is simply some rounding or slight base points that need to be altered a smidge.

    Great job by Dave in explaining the whole system. Hopefully people can appreciate how simply this whole thing is (and should be). The challenge for any other system is to make it better, WITHOUT making it more complicated.

    Vote -1 Vote +1

  12. Darren says:

    Where does Pitcher hitting come into play with total Wins above replacement. I would assume that each team’s pitchers hitting stats are so brutal that it would bring down the total teams wins above replacement level. Or should be assume that there is a positional adjustment for hitting pitchers that is so high that it cancels this out. Not sure I am making sense, but this seems to be the only thing missing from the total picture

    Vote -1 Vote +1

  13. Erich says:

    Thank God! Dave you have delivered me from confusion! All I’ve ever wanted from these win value explanations was a decent context to view them in.

    A team made up of replacement level players finishes a season in the 290 to .300 win% range.
    I GET IT!
    HALLELUJAH!

    Vote -1 Vote +1

  14. Brian says:

    I would expect position players to accumulate more than half of the win values, because the game is exactly 50% run accumulation and 50% run prevention. If you wrap your head around that basic framework, it becomes clear that the game is 50% offense and 50% defense, with defense comprising both pitching and position player defense (this is the shortcut method to ‘prove’ that position players are worth more, on average, than pitchers).

    Your figures seem to imply that the game, at least above replacement value, is 50% offense, 43% pitching and 7% defense. Intuitively, these numbers seem ‘right’ to me, or at least scaled correctly.

    Any thoughts?

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>