Freakonomics Suggests This Was the "Year of the Fielder"

[StlSluggers]

Interesting take on this low-scoring season.

First, he points out that the scoring drop has almost entirely occurred on the AL side; something I didn't know.

Interestingly, not all of baseball is suffering from fewer runs: the decline has been limited almost completely to baseball’s American League. AL teams are scoring .37 fewer runs per game, from 4.82 to 4.45; National League teams, meanwhile, have only reduced their scoring by .04 runs. Of the drop of 1,105 runs in the major leagues from last year, the American League has been responsible for a whopping 823 of those.

Then, he does on to draw correlations that he feels suggest that improvements on defense actually contributed to the offensive swing (not pitching).

If you find this to be tl;dr, here's the conclusion:

It’s not crazy to blame the scoring drought on fielding improvements: great fielders often contribute doubly to reduced scoring rates in their second role as below-average hitters. The 2010 Mariners are a good example: they added defensive tools in the off-season and continued to be one of the better fielding teams in the league this year, but their offense fell from the worst in the league in 2009 to one of the worst of all time in 2010. So when you’re putting a great fielder in your lineup, you’re also usually hurting your own offensive production, effectively reducing overall scoring on both sides of the ball.

Hayes Davenport, who provided research for the “Year of the Glove” segment on “Marketplace,” is a staff writer for Comedy Central’s “Big Lake” and “The Nick Swardson Sketch Show.” He is a certified stathead: his senior thesis at Harvard (from which he graduated in 2009) was called “Anti-Rationalist Rhetoric in American Baseball Journalism.

[StlSluggers]

Personally, I'm more inclined to buy into the "good fielders were bad hitters" argument based on the numbers he presented.

Those r-squareds weren't exactly inspiring (though maybe they're above average for the industry... i don't know what other correlations tend to look like).

[rookies and cream]

Haven't had time to digest the article but r squared of .33 = r of .57, which according to Cohen (1988) represents a large correlation.

Basically the r squared of .33 means that 33% of the variation in runs can be explained by defense. That is pretty significant I think.

[StlSluggers]

Haven't had time to digest the article but r squared of .33 = r of .57, which according to Cohen (1988) represents a large correlation.

Basically the r squared of .33 means that 33% of the variation in runs can be explained by defense. That is pretty significant I think.

In my line of work, if I'm not looking at a +/-0.6, it's not worth looking into to. Like I said, maybe it's different in baseball.

[Skin Blues]

Basically the r squared of .33 means that 33% of the variation in runs can be explained by defense. That is pretty significant I think.

And 33% by pitching, and 33% by hitting. haha, I know it's not that simple, and I'm not as adept at stats as I was when I was actually taking the courses. But I do think it's likely that defense is the new "moneyball" advantage, which could account for the lower run totals. Mostly because it's so hard to valuate. It'd be interesting to see if it's the entire American League or a select few teams. A lot of managers still play the intuitive traditional way for defensive alignment, who to play where, what constitutes "good" (diving catches and errors vs getting a good jump and being in good position) and whatnot which leaves a lot of room for improvement over areas like pitching and hitting which are much easier to quantify who is good and who is bad, and how to get the most out of them.

[mweir145]

Bautista realized teams can't defend home runs, so he decided to hit 54 of them.

[Neato Torpedo]

Basically the r squared of .33 means that 33% of the variation in runs can be explained by defense. That is pretty significant I think.

And 33% by pitching, and 33% by hitting.

rofl

[bleach168]

When he uses a bigger sample size, only 25% of the drop in runs scored is accounted for by defense.