Measuring Defense w/ Statistics - No Thanks

I really admire the work that people like John Dewan and Bill James do. They are pioneers in the world of baseball analysis and they, along with countless others, have helped change the way "baseball people" go about their business.

That said, I haven't found a single defensive statistic that I find terribly useful.

I love the concept behind Dewan's Plus/Minus system (http://www.fieldingbible.com/) but there are still too many things that are not quantifiable enough for me to buy into it completely. I also think some of Dewan's new "Defensive Runs Saved" concepts are interesting, yet unconvincing.

I should point out, right off the top, that to my knowledge Dewan doesn't promote either of these stats as "Be All, End All" of defensive evaluation.

The problem is that these things are all subjective, relying completely on the evaluations/decisions made by people watching video of Major League games. Even if those people are "baseball people" (former coaches, scouts, players, etc.) we're still talking about subjective analysis. "Baseball people" don't always see things the same way, afterall.

Now, defensive statistics are better than they used to be and maybe someday Dewan, James or some other forward thinker will have a break through and get it down pat. The problem, however, is there are too many defensive factors that simply cannot be measured.

Positioning, for example.

Let's say Player A has better physical ability than Player B. Quicker feet, better balance, stronger arm, more accurate arm, etc. If they're both in the same pre-pitch spot on the field Player A can make plays that Player B cannot.

On the other hand, Player B is a more "heady" player. He's more up to date on the tendencies of opposing hitters - and perhaps his team has more accurate scouting reports - and therefore he's in the correct position more often than Player A.

Player B could be making "routine" plays all the time because he's in the right position while Player A makes "excellent" plays, in part, because he's not in the best pre-pitch position.

Simply put, there is more to being a strong defensive player than physical skill. Preparation, anticipation and instinct are all significant factors in a player's ability to field his position and none of those factors are quantifiable.

Is the player in the proper pre-pitch position based on scouting information? Is he in tune with how his team is planning on pitching to the opposing hitters? Is he anticipating the play pre-pitch based on these factors?

Does the player throw to the right base consistently? Is the player a "heads-up" or "head in the clouds" kind of guy? Does he get into the proper relay position? Does he hit the cutoff man? How does he play the ball off the wall?

Is the player "saved" by his fellow defenders (like Mark Grace saving Shawon Dunston's butt all the time) or is he the one doing the "saving?"

I just don't see how anyone can place a numerical value on these things. They are bits of invaluable information when it comes to a player's ability to field his position and they are also immeasurable. You can see them with your own two eyes if you know what to look for, which most hardcore baseball observers do, but I just don't see how you can quantify it.

There are other more tangible factors that are not factored in, at least not to my knowledge or satisfaction...

1) Actual velocity of the ball hit (not just a generic term like "soft" or "hard" hit ball, actual MPH - any hitter will tell you, there is a distinct difference between 90 MPH and 95 MPH and when watching the game on video that difference won't be noticeable on a ball put into play)

2) The condition of the field (wet or dry, shorter grass or longer grass, etc. - it's not like every big league field is exactly the same)

3) Variation in spin/bounces (every ball hit is a little different - a "routine" ground ball right at a fielder isn't necessarily "routine" as some balls "kick" or "bite" at the last moment based on how they were hit)

4) Trajectory (was the ball hit at a 45-degree angle, 60-degree angle or 75-degree angle? did it have backspin or topspin? was the ball cutting or slicing? I need more detail than a simple "fly ball/line drive" label)

5) Conditions (sunny w/ a big sky? cloudy? windy? raining? twilight? shawdows? those are all factors in how plays get made, especially difficult plays)

All of those things are "known" factors but they cannot be easily converted into a number. Without those pieces of information nobody can come to an accurate conclusion about the quality of a player's defensive game. Maybe some of that evens out over the course of a 162-game season but not all of it does.

The aforementioned defensive statistics can be helpful when it comes to filling in the blanks on some players when people just don't have the time/ability to judge for themselves, but they're just not complete enough for me to value them over scouting information.

There is just no logical way to turn pages and pages of scouting information into a number. Even though "real" scouts use the 1-8 scale to grade players' physical ability (which I don't much like either) it is included in a detailed report that provides some of the nuance/detail I wrote about above. Plus, most of the "tools" scouts are grading can be measured - velocity on throws, bat speed, 60-yard dash time, home-to-first time, etc.

What all of these defensive statistics come down to, essentially, is a human being assigning a number to a play based on his own subjective analysis. In that sense, defensive statistics are no different than the subjective analysis in a scouting report.

Defensive statistics cannot provide the type of detail a written or spoken-word report can.

Then again, the defensive statistics themselves aren't the real problem.

It's the people who use them improperly - replacing detailed professional analysis with a number - that are probably at the root of my frustration....

KW