Golden Opportunity: Using SDI to Evaluate the Twins Defense

I am glad we are developing more metrics for defense, as they are hard to quantify normally.  We all the know the old ways of errors is not good, as players with greater range would get more errors because of more chances at high risk plays.  Players with less range would not get errors because they could not get to balls to be charged with one.

It is good to see Polonco improving overall, and the eye test kind of backed that up.  Sano I think much of his issues were not knowing when to go after a ball.  He will never be a plus defender, but he did better than I was expecting, which I expected dumpster fire so that is not saying much.  

For me, with all the shifts trying to look at range based on normal alignment is flawed.  I would judge OF by reaction, route efficiency, speed, arm strength and accuracy.  If you get good jumps, run good routes to ball even if you are slow you can get to balls someone that is just fast but cannot judge the fly well.  Then arm is huge.  Buxton seems to be high on all levels.  

For IF, similar I would judge reaction, quickness, along with arm strength and accuracy.  Of course how many balls they boot are important.  With all the shifting it would be hard to judge someone for not getting to a ball because out of position. That is where batted ball measures is important. 

This was interesting, but the thing that caught my eye was what I call Rosario bias - "The 2020 season exemplifies how a small sample size can make a player look better or worse than their career numbers."  This could have been the lead to the entire essay, but you chose it for Eddie because he obviously looks better than you expected or wanted.  We will miss Eddie when he is gone.

Overall the defense is below a championship level and that needs to change.  You did not show the 3B or 2B rankings.  I am guessing that our players did not get enough games at 2B to qualify for their list.  I do find it fascinating to see Goodrum and Schoop ranking so high at SS and 2B,  

Gonzales was number 8 among 3B, the last one to have a positive score (0.2).  I was surprised to see Devers (Boston) really bottom out as the worst 3B and Ramirez (Cleveland) also with a strong negative score. 

Then there is the big shocker - the worst CF on the list - Mike Trout!

Thanks for the article Cody. 

I do feel you glossed over Kepler's decline a bit. True Gallo had a great year defensively but Kepler was down defensively and if his 2020 play continues he will have no chance at winning a gold glove.

Also, I'm curious to know how the catchers fared defensively. The home town broadcasters constantly fawn over their defensive prowess, but I'd be interested in what the stats say.

The Strategic Defense Initiative? I suppose Maeda could knock down an incoming missile but I'm not sure I'd trust even Buxton to catch one.

Not all errors or misplays have equal weight--just like hitting. This is the trouble with defensive metrics. Polanco's error in Game 1 of the playoffs lost the game for the Twins. The weight of the error is the equivalent of a closer giving up a two-run homer with two outs to lose the game. If this had been the first inning, sure it would have hurt, but the intensity of the error would seem somehow much lighter in retrospect.

Not all errors or misplays have equal weight--just like hitting. This is the trouble with defensive metrics. Polanco's error in Game 1 of the playoffs lost the game for the Twins. The weight of the error is the equivalent of a closer giving up a two-run homer with two outs to lose the game. If this had been the first inning, sure it would have hurt, but the intensity of the error would seem somehow much lighter in retrospect.

This is a good point. I don't know the details of the parameters used to calculate SDI but I would think it could be configured to account for changes in win probability, unearned runs, or even pitches thrown. There are probably other relevant parameters as well. And for all I know it may already do that.

Thanks for the article Cody. 

 

I do feel you glossed over Kepler's decline a bit. True Gallo had a great year defensively but Kepler was down defensively and if his 2020 play continues he will have no chance at winning a gold glove.

 

Also, I'm curious to know how the catchers fared defensively. The home town broadcasters constantly fawn over their defensive prowess, but I'd be interested in what the stats say.

In Baseball reference both Garver and Avila had a 0.3 and Jeffers 0.0 dWAR.  

What has always interested me is a players range - we know Buxton can cover half the outfield now, but what about Kepler and Rosario?  In the infield how far does Polanco cover?  A SS has to have range.  I know his arm is not the best, but how much does he cover?  That is really the only position in the IF that has as much range factor - 2B would be second.   At 3B I want reaction time - those bullets that come down the line make the 3B almost a hockey goalie. And at 1B I want to know the ability to pickup balls that have to be short hopped from the other IF throws.