Editorial Note: The latest information graphic from the talented Craig Robinson of Flip Flop Fly Ball shows us the team wins above replacement of position players (and pitchers when they bat) for every Major League Baseball team, as well as the biggest every day contributor on each team. Adam Jones has permission to reference this piece for any MVP campaign purposes. Jeff Francoeur does not. [D.P.].

Comments (24)

  1. It is ironic that last night on MLB Tonight, the guys were drooling about how great of a defensive player Francoeur is

  2. Mike Fuckin’ Trout.

  3. You’ve done it again, Craig! Amazing job.

  4. Fucking Baltimore. How are they still winning ballgames?

  5. This isn’t a smartarse question, it’s genuine confusion. WAR is meant to indicate how many wins a player contributes above an ‘replacement’ player. In that case, shouldn’t WAR average out at zero across the leagues? The graph indicates that even the worst teams are much better than ‘replacement level’, and a 1.7 wins above replacement is nothing outstanding, while 1.7 below is appalling. I don’t understand.

    Also, if WAR is meant to be a great metric to use, how come it only seems to have a moderate correlation with the teams that are doing well? Sure, the good teams have the best WARs, by and large, but there are some good teams with bad WARs.

    • A “replacement player” is the average guy who replaces your guy if he goes on the DL. So a AAA or bench player. Thus, most teams ought to be well above that.

      • So the replacement guy is the mythical average AAA player, not the actual player. For example if Encarnacion goes down the replacement is the average AAA SS and not Hetch right?

    • A team made up of replacement level players would win about 35-45 games.

    • The comparison is not the “Average” player in MLB, but the average player you’d be able to find if a replacement were necessary. If it were based on the “average” player, then, yes, it should equal out to zero.

      The 1.7 question: 1.7 above replacement is in the range of an “average” MLB player. -1.7 is significantly worse than a replacement level player and far, far below the average MLB player — in other words, a AAA player would be preferable as a replacement to such a player.

      Finally, the total WAR should correlate with wins-losses (though it is an individual level stat, and will not mirror what the team actually accomplishes — in much the same way that a team does not win every game in which it scores 6 runs or lose every game in which scores 2 runs, etc.) However, another reason why this graph is further from actual wins-loss totals is because it does not include pitching.


  7. ben – a replacement player is basically a AAA player and the theory is a team full of replacement players should win 50-55 games in a MLB season (I forget if its 50 or a number in the 50′s).
    The one factor that usually skews team WAR to actual wins is closers / back end of the bullpen. Toronto and Milwaukee who have had closer issues are high on team WAR but both under .500 whereas Baltimore, Cleveland and Tampa Bay have been very good in the back end of the pen (lower team WARs compared to actual wins).
    Its not perfect but a pretty good guide on overall talent and performance of teams

  8. Really close correlation between team WAR and team performance there ain’t it?

    • Yeah I think WAR is a bit over valued as the new go-to stat. It provides a good snap shot of a player’s overall worth, because it combines defensive and offensive metrics, but it’s far from perfectly turned.

      Also, for all the completely just complaining about pitcher wins, why are we using another win-stat? Yes, very different, but trying to approximate how many team wins an individual player contributes to a season seems ridiculous to me.

      Also: a completely replacement level team is 35-45 wins? 10 wins is a hell of discrepancy.

      I’m still not a fan of WAR. I think it oversimplifies. There’s no easy one-stat way to value one player over another. You always have to look at the whole package, which is what WAR is attempting to do, but I think it fails.

      • It would be better without the defensive aspect.

        • Yeah, I suppose. I have a decent idea of how advanced defensive metrics work, and I think they’re pretty cool, but I think trying to wrap a player’s overall value into one tidy number is misguided.

          Also, what do we make of the cardinals having the highest team WAR up yonder?

      • Don’t miss out on the positional adjustments and playing time components. WAR is good at showing how even a seemingly bad player who stays on the field adds value.

        No, it isn’t perfect. But I don’t understand the impulse to throw the baby out with the bathwater because it doesn’t tie every last little thing up with a tiny bow.

        • i wonder if Team WAR is the problem. It seems to me it’s useful when looking at a single player, but Team WAR doesnt’ really teach me much… I dunno.

        • My problem with it is that it even attempts to tie every last little thing up with a tiny bow.

          I still look at it, but even as a fan who probably digs a little deeper than your average homer, I still have to look up what exactly the goddamn thing is and how it’s calculated every so often.

          It’s just so far removed from anything you actually see happen on the field (because of it’s complexity–not that it doesn’t reflect what actually happens in a game), I fear people are valuing it to highly and reference it to often without fully understanding it.

          BTW, anybody listen to Hayhurst and Cosentino? I’m pretty sure somebody is feeding them these new fangled stats to spit on air without explaining to them what they actually are. They get hayhurst to explain advanced stats and he fucks it up all the time.

Leave a Reply

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