High Heat Stats Podcast: Episode 2 WAR–What is it Good For?

A new HHS podcast is available!

We have added a page for all the podcasts. You can click on “Podcast” in the navigation bar at the top of the page, or click right here.

It would be very helpful if you could leave a rating or review on our iTunes page. Just search for High Heat Stats, then leave your review.

8 thoughts on “High Heat Stats Podcast: Episode 2 WAR–What is it Good For?

  1. 1
    brp says:

    The segment about Puckett was really interesting in total.

    Regarding the fielding I was trying to think of a similar outfielder and I came up with Tony Gwynn- I believe generally regarded as a good (or at least average) fielder when young, and then gained just a ton of weight.

    But again the Gwynn data are strange: he went from a -23 in 1989 to +28 in 1991. There is some explanation in that he split RF/CF in 1989 and was presumably a poor CF and by 1991 was full-time in RF.

    However, after a +19 in 1992, it drops to -8, -11, -18 the next three years. Then, magically back up to 3, and then down to -15, -17.

    So in 1991, Gwynn’s WAR is 5.6 despite a 118 OPS+ and only 2.5 in 1989 despite a 132 OPS+.

    It’s still tough for me to wrap my arms around these metrics, especially when they can vary so much for a player… but I guess we accept that variance more easily from the plate than in the field. Still trying to buy in on fielding statistics. Not there yet.

    • 2

      brp—you do raise some very interesting points.

      It’s a good catch about Gwynn playing center field in 1989. That means his Total Zone is compared to a whole different class of outfielder. And he probably just didn’t stack up. The other outlier I see is 1996. That year he had an excellent fielding percentage (as aways), but a pretty lousy range factor. I guess for some reason Total Zone doesn’t feel it was that bad. Could be some batted ball data influencing that (perhaps his range factor is low because of more ground ball pitchers or something).

      That said, there are still two distinct halves of Gwynn’s career.

      Age 22-32: +76 runs (even with that -28 season in there)
      Age 33-41: -70 runs

      This, like Puckett, is not a man who got lighter with age. I think the basic career arc of his Total Zone tells a reasonable story. I think Total Zone just isn’t as steady from year to year as we’d hope. It is much more prone to streakiness as there are far fewer samples to work from.

      Plus, I think we should always remember that hitters can go into slumps and fielders can go into slumps. I’m not using that to explain the ups and downs for Gwynn, but consider this:

      Batting runs for Gwynn:
      1982-1983: 6.4 batting runs/700 PA
      1984-1989: 30.2
      1990-1992: 10.7
      1993-1998: 38.2
      1999-2001: 21.0

      • 3
        brp says:

        The larger year samples definitely do make more sense, and that 22-32/33-41 table matches up both with what you’d expect given Gwynn’s age and waistline.

        Surprised Gwynn ever had such low batting runs as he did in 90-92; and there again is my difficulty accepting fielding variance as readily as batting.

        Anyway, looking forward to finishing out the podcast today.

      • 4
        RJ says:

        “I think Total Zone just isn’t as steady from year to year as we’d hope. It is much more prone to streakiness as there are far fewer samples to work from.”

        I’ve noticed this too when looking at a player’s Rfield totals. I suppose the question is, is this variability a demonstration of a flaw in the statistic (like a low sample size as you say) or a true representation of fluctuation in a player’s performance?

        Somebody around here brought to my attention the fact that one of the reasons Mike Trout’s defensive numbers are down this year is because he’s not made any home run saving plays, compared to several last season.

        He’s probably not had those opportunities though. I wonder (and somebody please stop me if I’m talking rubbish) if that’s not like marking a player down for not having any bases-loaded hits, even though he never batted with the bases loaded. Whilst making a HR-saving catch shows defensive skill, the circumstances that allow you to make that catch, ie the location of the ball being just over the fence, are largely beyond your control.

        Again, I’m pretty green in these matters, so please set me straight if that was all nonsense.

        • 5
          RJ says:

          Re-reading that penultimate paragraph, I seem to be implying that Trout is getting negative Rfield by not making stupendous plays, which I’m sure isn’t the case. I guess my point is, how valuable are those plays? If the answer is “really valuable” then the year-to-year variation makes sense. It does make it all *appear* unsatisfyingly random though.

          • 6

            Mark Simon had an excellent post about this topic: http://espn.go.com/blog/sweetspot/post/_/id/36800/why-are-trouts-defensive-numbers-so-bad

            This is a spot on observation:

            He’s probably not had those opportunities though. I wonder (and somebody please stop me if I’m talking rubbish) if that’s not like marking a player down for not having any bases-loaded hits, even though he never batted with the bases loaded. Whilst making a HR-saving catch shows defensive skill, the circumstances that allow you to make that catch, ie the location of the ball being just over the fence, are largely beyond your control.

            Now, this doesn’t explain why he’s negative. But it does go a LONG way in explaining why he isn’t astronomically positive.

            I hope the wonderful Mr. Simon doesn’t mind if I quote that whole section of the article:

            Trout had four plays that BIS scored home-run robberies last season. In their world, those are very important plays. As BIS founder John Dewan has explained, this is the one play in which a player literally saves a run, no matter what the situation.

            Those four plays accounted for 7.4 of Trout’s runs saved last season — basically one-third of his total.

            Trout has not made any homer-robbing catches this season. That’s not necessarily a knock against him. Homer-robs are really hard. Four in one season isn’t something that is easily repeatable.

            In 2009, Adam Jones had four. He’s had two since then — one in 2011 and one a couple of weeks ago. Gary Matthews Jr. had four in 2006 and then had one the next season. Nook Logan had four in 2005 and didn’t have any in the remaining 145 games of his career.

            So, that right there is a lot of it. But there’s also a lot more.

          • 7
            bstar says:

            RJ, ’twas me who originally put that nonsense in your head.

            Here’s the comment:

            http://www.highheatstats.com/2013/06/sunday-game-notes-5/#comment-59749

          • 8
            RJ says:

            @bstar Haha, the bit that Adam quoted above as a “spot on observation” by me is basically a rewording of the key points in your post. As always, you guys are making me smarter about these things.

Leave a Reply

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