I hear it all the time, people citing fielding percentage and errors as an example of Detroit’s ineptitude with the glove. They have the 9th most errors in MLB and only two AL teams have more (the Indians and the Angels) and their fielding percentage ranks 23rd overall. Those numbers aren’t great, but what is more troubling is the fact that so many are still using those stats when there are so many betters ones easily available.
Play by play data that tracks hit direction and type of hit has been available for a number of years. This is data that has been collected by video scouts and stringers that aren’t making a nebulous judgment about whether or not the play should have been made. They are just looking at direction and how hard it was hit and whether or not the play was made. The data then tells the story of whether or not the play should have been made by comparing the play to all the other plays and how often they get made.
So let’s look at some of this objective data and see what the objective data says about the Tigers fielding. Baseball Prospectus lists the Tigers team defensive efficiency ratio (DER simply looks at all the balls in play and figures out the rate at which they were converted to outs) as 16th in MLB. On this measure they come up middle of the pack.
Ultimate Zone Rating (UZR) has the Tigers at 10th in MLB at 13 runs better than a typical team. A similar system called +/- has the Tigers at 11th in MLB at 25 defensive runs saved.
Revised Zone Rating is similar to fielding percentage in that it expresses things as a rate or percentage, but it uses a much more logical denominator. Instead of using the denominator as the number of balls that a fielder touches, it uses the number of balls hit to a player’s zone. Therefore it isn’t rewarding the slow of foot for not even getting to balls. In RZR the Tigers are 6th in MLB.
While there is variation in the systems, I just presented 4 defensive stats that aren’t subjective. They weren’t manipulated by me to make the team look one way or the other. The worst of those 4 said the Tigers were middle of the pack. The other 3 said the Tigers were above average to good.
The only metric that says the team isn’t good with the leather is the one that is frought with the most problems in methodology. It is the one that is most subjective. The one that doesn’t measure an ability to get to balls. Concluding that the Tigers are a bad team because of fielding percentage would be an error in judgment.
Wondering what everyone thinks of Guillen at second base so far? I think he has been okay, not terrible, which is what I was expecting. He looks a little awkward at times trying to turn the double play.
I don’t think Guillen is a long term solution, obviously. I don’t think the Tigers will be a good team until they upgrade the middle infield and catcher, too.
if you WORK the right stats he’s probably the best second baseman in baseball
Actually because of the volatility of defensive statistics drawing significant conclusions from Guillen’s stats at second base would be foolish. But if you want the numbers, not worked at all, just the numbers, he is -3 runs on +/- and -.9 runs in UZR.
I’ll have to agree with that, billfer. Excellent report.
heres the thing…if you have three sets of stats basically recording the same basic thing,and one set shows you near the top and one set shows you near the middle and one set shows you near the bottom……then there’s something wrong with all three sets!
thats why the old saying Stats can tell you whatever you want to hear, is very true
But I showed 5 sets of stats. Three of which showed the team was pretty good, 1 that showed they were pretty average, and 1 that showed they were bad. The one that shows they are bad is the one with the worst measurement methodology. I don’t see why there is any ambiguity here.
They aren’t recording the same thing… they’re all recording different things. The differences were all explained by Billfer in his post.
To use an analogy to the offensive side of the ball, if one team is terrific in batting avg., but awful in HRs, do we conclude that these are useless statistics? Is there “something wrong” with these sets of stats?
Trillo said that playing second base was more dangerous than ss. While at ss you go against the runner with the ball on hand , at 2b you have to put yourself in front of the runner with no defense from him. If you are injury prone you will hesitate when going to the play. And been around 35 wont help either, specially if you play there for the first time in 11 years.
Anyway , since Cabrera was a 3b and have a long range he can cover for him
Facts are meaningless. You could use facts to prove anything that’s even remotely true!
– Homer Simpson
And Guillen has been better at 2B than I expected. Has anyone heard a decent explanation for why he plays so far up the middle?
Because he’s more comfortable ranging to his left than he is to his right…
Because Cabrera seems to like ranging toward 2B?
I kind of wonder if defense kind of defies numerical quantification to a point where it’s almost useless to try. I suspect you’d learn more from watching a fielder for a month with your own eyes than you would from studying his stats for several years.
Thank you Bilfer. Great to hear you telling it like it is.
I guess we need to buy Panas’ book to get a better understanding as to why BA, ERA and Fld Pct really don’t mean anything.
billfer didn’t say – and I highly doubt Lee’s book says – that the stats you listed “don’t mean anything”. On the contrary, batting average is extremely useful for telling us how often a batter gets a hit. Any hit. And that is one flaw with batting average, it treats all hits as equal even though they clearly are not. The stats you listed provide information and do have meaning, it’s just that they have a limited meaning that isn’t always obvious until you really think about it. The purpose of all these other “new” stats is simply to capture information that the other traditional stats don’t.
Actually I was only being slightly facetious. BA alone cannot be counted on to be a good indicator of a player’s value (e.g. Nick Punto or Danny Worth), and the additional new hitting stats can be helpful. ERA is notoriously unreliable, for example in the case of a team with iron gloves backing up the staff, and especially in regards to closers, whose ERA can shoot through the moon with only one bad outing. I’m more skeptical of the new fielding metrics mostly because there are so many different ones and the results vary so widely. Perhaps the assessments will become more reliable with future refinements, but in the meantime I am not convinced even though I admit Fld Pct is a pretty wobbly way to go too.
To summarize my thoughts on the stats:
BA is actually not a bad stat. The problem I have with it is that people have traditionally used it to summarize a player’s performance when it only describes a small portion of his performance. I don’t see that happening as much in recent years which is a good thing.’
ERA is a pretty good stat too, as long as you understand its limitations and also look at peripherals.
FPCT really doesn’t tell you a whole lot about a player’s defense or a team’s defense. The new defensive stats are a lot better but still somewhat limited. You need to look at multiple stats and multiple years of data to get get a good picture.
Thanks Billfer — so the difference between UZR and RZR would indicate that the Tigers are getting to more balls than some of the teams rated above them in UZR, right? Can you tell from the numbers who might be most responsible for that — Jackson? Inge? The SS position? Also interested in Boesch’s numbers, though the sample size will still be pretty small — my even smaller sample of anecdotes says he gets a late jump on a lot of balls but seems to cover a lot of ground for such a big guy.
And here’s the link to another set of data over at Primer — spreadsheet won’t open for me at work.
http://www.baseballthinkfactory.org/files/dialed_in/discussion/weekly_drs_update_defensive_stats_thru_jun_20_2010/
I don’t have the stats, but it seems like whenever I listen or watch a game, Laird is bouncing another throw to second. With his .180 average, he better kept back to throwing them out.
Although, I agree that fielding percentage is not a good stat…Use your eyes and knowledge of baseball…to say the Tigers are even an average defensive team..is suspect. There are no good stats..you need to have a working knowledge of baseball..seen many games and use all types of data. The Tigers are a team, that has a CF that can get to so many balls..that he may in the long run save the ineptness of the corner outfielders..We all know that Jackson saves many runs in any give week. We don’t need stats to tell us that…we also know that guys like Raburn, Damon, Maggs, and Boesch (with his bad jumps) probably cost the tigers some runs every month. I guess it just depends where the balls are hit and at what stadium, you are playing…Don Kelly cost the Tigers a division title, because he couldn’t catch a popup in the metrodome..what did the stats say about him? Now the infield…other than Inge..that is a pretty poor defensive group. Guillen’s range is laughable and the SS is average at best. I would entertain the idea, that the Tigers are better than their error total (which is still important because errors are easy plays, not made)…but, to say they are a good defensive team…defies all logic and everything that I have learned over the years from watching 200 plus games per year for over 20 years…This is a team w/ two very good defensive players (not counting Laird) and a bunch of guys below average. Miggy does the job, but, he strays too far off 1st base compared to any of the greats. They also commit more errors than anyone…I know errors are not a great stat (mainly because all teams commit more errors than what are scored) so…to diminish the fact that they fail to make more routine plays than any other team…severely damages any credibility. You could say “Tigers better defensively than the numbers indicate”but, to make that leap and say they are good…well, drink some more Kool-Aid. They are a great defensive team, when the ball is hit to Inge or Jackson.