Total Zone with Location Data
A couple weeks ago Sean Smith released a revamped Total Zone system which included Gameday location data. You can read all about it on baseballprojection.com.
On FanGraphs you can now find the new Total Zone numbers for the 2005-2009 season in the player pages under the heading TZL in the Advanced Fielding sections. Here’s what Carl Crawford’s numbers look like:
David Appelman is the creator of FanGraphs.
For those of us newbies who enjoy the differing defensive metrics, what TZL scores would be considered “awful”, “very poor”, “poor”, “below average”, “average”, “above average”, “good”, “very good” and excellent!
And are we still looking at 3 seasons of data to really get a statistical feel for a players value on defense, such as with UZR?
Cheers,
Matt
For 2009 the distribution for UZL and TZR are very similar.
I think doing something like the following is pretty reasonable. It’s similar to what we did for the fan projections.
-12.5 to -17.5 or below awful
-7.5 to -12.5 poor
-2.5 to -7.5 below average
-2.5 to 2.5 average
2.5 to 7.5 above average
7.5 to 12.5 good
12.5 to 17.5 or above excellent
are these scaled the same way UZR are?
like a 1 = 1
or is it
1 TZL is equal to 3 UZR
It’s all based on runs saved so my guess would be that they’re on the same scale.
they’re not scaled the same way… so it’s not easy to relate them like that. here’s crawford’s data, if interested:
http://img175.imageshack.us/img175/2635/uzrtlzcrawford.jpg
It looks like they’re scaled the same way to me.
TZL and UZR measure the same thing: runs saved. However, because they use different methods to determine that, the results will not be exactly the same.
Hmm, perhaps my post was errantly worded. What I meant to suggest was that, of course, these metrics measure the same thing, so they (presumably) are highly correlated. However, since they use different data and apply it in different ways, there does not appear to be a simple 1->1 mapping of UZR to TZL and vice versa. i.e. there is no nice way to say X units of UZR equals Y units of TZL.
198d, that still doesn’t quite make sense to me. If both are based on runs above/below positional average, isn’t that the same scale? Surely they may arrive at different results due to the differences between the way they are calculated, but aren’t runs, runs?
I have a question about TZL. if a player is 8.5, is that 8.5 above average for his position or for all players?
For example, if a shortstop is 8.5 TZL and a first basemen is 8.5 TZL, are they considered to have the same value defensively?
Scott,
it means they have prevented the same number of runs above average. People will say that doesn’t mean they have the same value defensively because the 1B can’t play SS, but I largely consider that irrelevant. The SS isn’t likely to be any good at 1B.
But it means they have prevented the same number of runs, and wrt value in that dept, they are equal.