How do we measure fielding in baseball?
craig44
Posts: 11,255 ✭✭✭✭✭
in Sports Talk
I mean this more for the old timers. I keep reading lots of posts on other websites, and posters will mention how great such and such was on defense and how much that adds to their overall value. For example, honus Wagner and pie traynor are commonly called two of the best defenders at their respective positions. My question is, how do we know? Is it all anecdotal? I can't imagine there is much video footage or anyone left who saw them play. Were there defensive statistics recorded then other than errors committed? All this, and I am not totally sure how accurate/valuable the current metrics are for modern players. Can anyone enlighten me?
George Brett, Roger Clemens and Tommy Brady.
0
Comments
By far the best single metric is Win Shares, but it is less definitive than it is the best educated guess. It's also not available anywhere for free (as far as I know). If anyone doesn't own Bill James' book "Win Shares" and you're serious about wanting to properly evaluate players, you really need to get it. On baseball-reference, the best stat is total zone runs (labeled "Rtot" in the Standard Fielding section); this stat is only 100% reliable (based on complete data) for 1974 on, and gets less reliable as you go back to 1930, at which point it isn't calculated at all. Defensive WAR is directionally reliable more often than not, but it can be completely wrong. Sort of like batting average, it's better than nothing but you should always try to confirm it with other stats, at which point why bother having looked at it at all?
Beyond that, baseball-reference has fielding percentage and range factor both for the player and for the league each year. In general, the better the player the higher both of these stats will be above league average. But, range factor is a lot more important. A shortstop who gets to an extra 100 ground balls and makes 10 more errors and 90 more outs will have a lower fielding percentage than the shortstop who didn't get to those balls, but he will have helped his team MUCH more.
Least reliable is anecdotal evidence, including Gold Glove awards, and for modern (post-1974) players it should probably be ignored entirely. But the further back you go the more likely it is that anecdotal evidence has to play a part in the evaluation because there just isn't much else to look at. An excellent rule of thumb - players who have won a Gold Glove were probably very good fielders when they won their first one. After that, fielding excellence is less important to winning additional Gold Gloves than the player's batting average. By pretty much every measure, Clete Boyer was a better third baseman than Brooks Robinson, at least at his prime, but by the time Clete established himself at third Brooks had already won a Gold Glove and he was a better hitter so he kept winning them year after year. Boyer finally won a Gold Glove when he got traded to the National League. At that point, Aurelio Rodriguez was probably the best third baseman in the AL, but he had to wait until Brooks was a part-timer before he finally won a Gold Glove. Robinson was a great fielder, and you can identify that by the fact the he won a bunch of Gold Gloves. The problem is that he won too many, and fielders as great or even better than he was get lost because of it.
So, if you look at AL third basemen in the 1960's / 1970's you'll see Brooks Robinson with 16 Gold gloves, Clete Boyer with none (one in the NL) and Aurelio Rodriguez with one. If there is any other stat or anything at all that distorts the comparative skills of those three third baseman more than that I'd have to see it to believe it.
George Brett, Roger Clemens and Tommy Brady.
Ok, that's kind of what I thought. So there really isn't any way to measure the real old timers like Wagner or speaker other than anecdotally. About range factor, it seems that the pitching staff could be a big influence on this stat. It seems that if you are fielding behind a high strikeout staff, that each fielder would get many less chances even if they have great lateral movement and natural range.
That's exactly right; the nature of the pitching staff can have a big effect on range factors. Of course, it averages out in the league factor, and on average it's a good stat for every fielder, but any given fielder could be way off.
The only stat that does a halfway decent job of trying to factor in the pitching staff is Win Shares. James would be the first to admit that it still involves some averages, and some assumptions, but he developed tons of formulas for each position to account for what he was able to account for, and the result is pretty good.
The best stat you can find on baseball-reference is the total zone runs and it does a very good job of assigning value to all the balls that are hit at a fielder. But, to your point, a player that gets a lot more balls hit at him will have a higher Rtot value than an equally good fielder who doesn't get as many chances.
Bottom line, don't try to evaluate any fielder by one stat, even Win Shares. Find as much information as you can and in most cases all of the stats paint a picture that's clear enough, if not entirely in focus. For the old-timers like Wagner, you can probably tell who the very best were and who the very best were and who was in between, but not much more than that. For them, contemporary accounts are still not worth much, but since you don't have much to begin with they do probably add to the picture. For the old-timers, to the degree I ever have any reason to judge how good they were in the field, I just rely on Win Shares.
Agree defense is so hard to rate.
Take one of my favorite players Pudge R.
1992 makes most errors but wins GG. 13 most for catcher.
It is in my opinion he was so good at throwing runners out they simply didn't try stealing.
That is a big factor....now a runner isn't in SP.
Is there any stats on how much a runner did/did not run on a particular catcher?
% caught?
So many factors can go into an error for a catcher it can make your head spin.
Is there stats like how many digs a 1st basement saves an infielder?
1948-76 Topps FB Sets
FB & BB HOF Player sets
1948-1993 NY Yankee Team Sets
He does appear though to have a bit of extra time on his hands with all his in-depth responses.
To answer the OP. Not very well.
I look at fielding % and (oh no) observation, Gold Gloves are misleading, but taken into consideration.