Re: Old school stats vs. new school stats
posted at 1/27/2013 10:34 AM EST
In response to georom4's comment:
In response to SonicsMonksLyresVicars' comment:
In response to georom4's comment:
Im going to stick with <flush>
the first thing i learned in statistics class is that if you cant control all factors that influence the said activity, then your measurement is not valid...
is the wind being measured? the sunny field? the wet, heavy air? the velocity of the ball hit? are all ballparks the same size? do all players have the same reflexes (wouldnt that make as much sense as tracking their positioning?) the variables are endless which makes UZR moot...
Remember, this is the stat that said Cameron was a better center fielder than Elllsbury...why would anyone even consider this stat useful?
Using your eyes is a much more accurate way of judging a fielder....your brain is the greatest computer ever created and it is quite capable of judging a ball player defensivily
Geo, that makes no sense at all. What baseball activities have completely controllable variables and therefore are, in your eyes, valid?
- BA? - nope, sun in the batter's eyes
- HR? - nope, wind can carry or hold back a fly ball
- Any pitching activity? - nope, pitcher could have a hangnail
- Any batting activity? - nope, batter could have a hangover
A good statistic should minimise the noise around the measurement of a specific activity. That's why an individual's runs, RBIs and wins are such poor statistics....they are team activities (other than a home run) and come nowhere near isolating an individual's performance.
No one is claiming UZR or any statistic is perfect....only "anti-stat" people like you are claiming that "pro-stat" people are making that claim! It clearly has some validity....just look at the top and bottom of the UZR list posted earlier and compare it to your and most people's observation. How do you propose to evaluate the range of all MLB SSs when you only watch most of them on television a handful of times over a season?
i agree with both you and Notin that ANY stat can be incomplete....but the worst of the lot have to be defensive stats - they do not monitor production as batting stats do...im simply saying you are better off taking a holistic view of defense than relying on any metric....there is such a thing as a Triple crown winner - are you really comparing UZR to HRs? RBI's? BA?
Its funny that the argument that UZR proponents use is that errors are not a good stat to guage defense...really?
I came across an article in 2010 about UZR and it was quite funny that Sox players like Youk and Cameron had no idea what theirs were...they didnt even take it seriously....then of course the article jumped the shark by saying this is why cameron has to play centerfield...But it also ended with francona and the theo saying that they have their own ddefensive metrics that they rely on/not uzr (they didnt specify) and that a 7th inning defensive replacement would not be based on uzr...
Part of why UZR has yet to take hold is that it's very creator has already stated that it has to be weighted in 3 year cycles. When one seasons worth of data is considered a small sample. Even the most ardent supporters of the system can't emphatically state that UZR should be the gold standard for how to measure defensive metrics.
So we're left with incomplete means to truly value defense and the gold glove. the standard for defensive excellence continues to be a popularity contest vs going to the best defensive player at each position.
As for whether errors are or are not a good measure of defensive excellence. A case could be made that not all errors are alike and the circumstances of the game dictate the risk reward of an infielder trying to make a play vs eating the ball. This is where the eye test is the best measure And it also is where having a quality first baseman can mean the difference between throwing the ball and eating it. If you have a first baseman that's adept at picking the ball you're more likely to let it fly knowing that if you get it anywhere near the bag that he'll make the play. Conversely if he has hands of stone, your lack of confidence in him making the play can cause an infielder to try to make the perfect throw and could be the difference between making the out and not...every 10th of a second = about three strides.
Great defensive players rarely make mental errors which are the ones that to me are the most aggregous. They also make the routine play look routine. If your SS as an example has sure hands and during the coarse of a season makes the routine play 99 times out of 100 with one mental error accounting for the error I would think we'd all take that. During the coarse of a season how many times does a SS have to make the do or die play and of those what's his pct of success vs making an error? If you subtract the do or die errors from a players total errors taking into account how many of those were balls he should have eaten vs trying to make the play. I think you could arrive at a better statistical value on errors.
While range is an important tangible for measuring a SS and most of the all time greats Ozzie, Apparicio, Belanger and Vizquel due to having the lateral quickness got to allot of balls that mere humans had to dive for. range alone is not the penultimate measure. If you don 't have the arm and the footwork to make a strong and accurate throw after fielding the ball it's moot. David Eckstein had very good range, but a water pistole for an arm. He made himself into a ML SS by perfecting a quick release. He was also very good at turning the DP and made the routine play look routine.
In the end one has to take the total sum of the parts to arrive at the proper evaluation of a players defensive ability. The SS is the guy that's responsible for more than just fielding ground balls and throwing it to first. So when evaluating who's the best of the best. It's not that tough, but to arrive at how best to evaluate Jeter's impact on the entire defense is where UZR IMHO falls short. The intangible of leadership can't be measured.