The War On Fielding Percentage

Despite today’s announcement of the National League Gold Glove winners, including Scott Rolen winning the eighth of his career, people are still talking about Derek Jeter’s inclusion in yesterday’s list of American League winners.

After the outrage and mockery following the announcement of Jeter’s win for a defensive prowess at shortstop that simply doesn’t exist, a funny thing began happening: justification.

Even though several defensive metrics and baseball observers saying otherwise, several people defended Jeter’s selection with a blind fervency usually reserved for religious exercises.  The most often used rally cry for defense of his defense was a little statistic call fielding percentage.

Yesterday, Drew Fairservice wrote an excellent piece on why batting average matters less than on base percentage (OBP), slugging percentage (SLG) and on base plus slugging percentage (OPS).  Today, given the recent Gold Glove news, I thought we’d take a look at fielding percentage.

I realize that a divide exists among baseball fans on the use of statistics.

I’m not going to call anyone an idiot for using one statistic over another.  It just so happens that newer metrics seem more reasonable to me than other, older alternatives, when it comes to measuring a player’s worth.

I’m hoping that it will be somewhat useful to those quoting Jeter’s fielding percentage to, at the very least, understand why so many baseball fans were dismissive of it.

From Bill James in the 1977 Baseball Abstract:

What is an error?  It is, without exception, the only major statistic in sports which is a record of what an observer thinks should have been accomplished.  It’s a moral judgment, really, in the peculiar quasi-morality of the locker room.  The fact of a baseball error is that no play has been made but that the scorer thinks it should have.  It is, uniquely, a record of opinions.

Michael Lewis in Moneyball further explains:

Errors had been invented in the late 1850s, when fielders didn’t wear gloves, the outfield went unmowed and the infield ungroomed, and the ball was bashed around until it was lopsided.  In 1860, a simple pop fly was an adventure.  Any ball hit more than few feet from a fielder on leave from the Civil War was unplayable.  Under those circumstances, it might have  made some kind of sense to judge a fielder by his ability to cope with balls hit right at him.

Range didn’t matter back in the day because the elements limited everyone’s ability to get to balls.  That’s just not the case today.  Continuing to judge errors with the same criteria in the modern game seems unreasonable because the most obvious way to avoid getting an error is to not reach the ball at all.

In other words, many balls that a bad fielder doesn’t get to end up counting as hits instead of errors.   That’s why you can’t use a player’s lack of errors as proof positive that he’s a good fielder, which is essentially what fielding percentage does.

There are other alternatives, but no fielding metric is perfect because there are so many variables to consider.  A few of the more popular are Ultimate Zone Rating (UZR), Plus/Minus and Defensive Runs Saved (DRS).  Put simply, these statistics count up the times a fielder is able to get to a ball that his average contemporary wouldn’t, while subtracting value for when he can’t make a play on a ball that the average defender at his position would.

It’s a bit more complicated than this in practice, as measuring these numbers involves breaking the field up into several different zones.  For UZR, statisticians then keep track of the out rate in each zone and the percentage of balls in that zone that end up as outs.  From there, they can calculate the average out rate and compare it to the out rate of each individual player.

That difference is then multiplied by the number of balls that were hit in that player’s zone to get a zone rating. In order to put it in terms of runs, the rating is multiplied by the Zone Ratings that are calculated for each zone the fielder covers, and then totaled.

From there, it’s further adjusted for park factors, batted ball speed, which side of the plate the batter was hitting from, the pitcher’s groundball/flyball ratio and the number of baserunners and outs at the time.  Using run expectancy charts, all of these rates can be converted to runs.

Confused?  You’re not alone.  There’s no question that it’s far easier to simply compare the number of errors a player makes to the number of chances he gets, but it’s also far easier to stay in bed all day, never leave the house and only watch television.  It’s easier, but you don’t get an accurate portrayal of what’s actually happening in the world.

The rest of the 2010 Gold Glove Award winners;  Albert Pujols 1B; Brandon Phillips 2B; Troy Tulowitzki SS; Carlos Gonzalez OF;  Shane Victorino OF; Michael Bourn OF; Bronson Arroyo P; and Yadier Molina C.