A Call to Arms

Prompted by Richard Lederer’s article at TBA, in which he discusses how the Angels have consistently outperformed their PECOTA-predicted record, and the discussion that has ensued over at The Book Blog, I have a simple request for the baseball community:

Somebody figure out what the heck is up with the Angels.

If this has been done before, then point me in the direction. My hunch is that is is more “marginal secret something + statistical noise” than an extreme of either. But anyway, I’m wondering what everyone thinks.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Bill James Things
Next: Joe Girardi, sabermetrician »

Comments

  1. Alex Poterack said...

    I feel like this can open up some new worlds for sabermetric analysis; I think one of the biggest stumbling blocks between stats people and non-stats people is understanding the pythagorean conversion between runs and wins; it’s not always intuitive that scoring more runs or preventing more runs leads to more wins, because the timing of it matters.  I think obviously the pythagorean relationship models reality extremely well; however, I think the Angels’ performance is evidence that maybe we don’t understand it as well as we think we do.

  2. DrBGiantsfan said...

    Just a guess off the top of my head, but maybe PECOTA equates walks with plate discipline.  The Angels’ philosophy has always been to wait for the right pitch, but when it comes, jump on it and hit it hard.  They are almost unique in being a team with good plate discipline(well, except for Vlad the last several years) but low walk rates.  So maybe in the Angels’ case, PECOTA overvalues walks?  Again, just a guess.

  3. Detroit Michael said...

    I suggest reading the Mike Scioscia section from Chris Jaffe’s book Evaluating Baseball Managers.  The book as a whole makes a forceful argument that a lot of the things that we call “luck” when we examine a single season are really managerial (often including coaching and front office help too) outcomes when one looks at 10+ seasons.

  4. Jimbo said...

    My vote is “statistical noise” + manager influence over time. I like on the blog where the term “efficient” is used. If the Angels are an efficient team, making the most of their runs produced, how would that be quantified? Couple things that I was curious about…

    Over the last 5 seasons, the best winning percentages in baseball are:
    Yankees 59.0%
    Angels 58.6%
    Red Sox 57.7%
    Phillies 55.2%
    Cardinals 54.1%

    If you calculate % of wins achieved via save, those five rank as follows:
    Angels 56%
    Phillies 48%
    Cardinals 47%
    Red Sox 46%
    Yankees 45%

    Pretty large gap. Is it possible the Angels get a lead and “play for the save” far more efficiently than others? Tactics such as defensive switches, or sacrificing for an ‘insurance run’ will suppress runs scored in a win. Of course other teams do it, but perhaps the Angels do it best/most? Perhaps their team is actually tailored to excel this way intentionally?

    Other thing I noticed in that same span, the Angels made ~300-400 fewer pitching changes than the other four teams above (based on Games / Games Started). Not sure if that is significant to the discussion here, but one could probably hypothesize a few ways it throws the PECOTA formula off. Perhaps the Angels have success (win loss success that is) sticking with a pitcher when others would go to the bullpen, or they are more willing to let a pitcher shoulder the occasional blowout. If memory serves, it does seem like Scoscia is less inclined than most to replace relievers based on handedness.

    I suspect a lot is manager/organization influence. I also suspect that’ll be hard to prove empirically.

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>

Current day month ye@r *