Edman85 Posted 13 hours ago Posted 13 hours ago (edited) Meg Rowley mentioned in passing in one of the Effectively Wild episodes they are changing over their defensive metrics but I forget the details. Probably trying to distance themselves from Lichtman given what was uncovered about him a few years ago. Edited 13 hours ago by Edman85 Quote
chasfh Posted 13 hours ago Posted 13 hours ago 11 hours ago, Tenacious D said: Can someone explain this to me? Julio is a premier defender in all facets. Riley is average-ish. Fangraphs has Riley up on Julio 2.9 to 2.4. Quote
gehringer_2 Posted 13 hours ago Posted 13 hours ago 9 hours ago, Tiger337 said: It's due to defense. Julio is a center fielder, so WAR gives him more positional value. He also has a much higher DRS (which is what b-ref uses for the defensive component) than Greene this year. My confidence in DRS has been falling annually. Quote
Edman85 Posted 13 hours ago Posted 13 hours ago 11 hours ago, Tenacious D said: Can someone explain this to me? Go down to the value section of their pages and click the glossary drop down. Their value calculation for position players is made up of five components (batting, gidp, baserunning, fielding, positional scarcity). Greene has a 15 run batting advantage, but loses 2 in baserunning, 1 in GIDP avoidance, 14 in fielding, and 6 in positional scarcity. 1 Quote
chasfh Posted 13 hours ago Posted 13 hours ago DRS, OAA, FRV, which ever letters you like, Julio is just better. He positions better, he covers more ground, and he has a better arm. That’s OK, Riley is outhitting Julio by a mile, and the Tigers are ten games better than the M’s. Quote
gehringer_2 Posted 13 hours ago Posted 13 hours ago 3 minutes ago, chasfh said: DRS, OAA, FRV, which ever letters you like, Julio is just better. He positions better, he covers more ground, and he has a better arm. That’s OK, Riley is outhitting Julio by a mile, and the Tigers are ten games better than the M’s. So it makes sense that positional value of LF is lower in general, but does a guy get dinged for being a LF at COPA, which has higher positional value than most other parks? Aside from that, can't disagree Riley has a noodle arm. 1 Quote
Tiger337 Posted 13 hours ago Posted 13 hours ago 12 minutes ago, gehringer_2 said: My confidence in DRS has been falling annually. Defensive statistics are less reliable than hitting stats in general, but I trust DRS more than most defensive measures. I like that is continually updated and transparent. I usually look at statcast and DRS, but it takes a long time for defensive numbers to stabilize. I don't think even a full season is enough. Quote
gehringer_2 Posted 13 hours ago Posted 13 hours ago (edited) 11 minutes ago, Tiger337 said: Defensive statistics are less reliable than hitting stats in general, but I trust DRS more than most defensive measures. I like that is continually updated and transparent. I usually look at statcast and DRS, but it takes a long time for defensive numbers to stabilize. I don't think even a full season is enough. I agree stability is part - I think it over measures - I don't believe player defensive values ping-pong around from season to season as much as DRS does. So I'd argue your syntax a bit - I wouldn't say DRS stabilizes over time, you're just taking an average over time to create a stability that isn't actually there in the measurement. So in a way that gives it more credit than I would. But the other thing I question is when you put it all together I don't think the total team aggregates pass the smell test. Now it's fine that knowledgeable people understand the limitation of defensive metrics; they aren't easy to get a good handle on, so you do the best you can. The problem is when the defensive piece gets rolled into WAR and the world take that number as gospel. Edited 13 hours ago by gehringer_2 Quote
Tiger337 Posted 13 hours ago Posted 13 hours ago 28 minutes ago, Edman85 said: Meg Rowley mentioned in passing in one of the Effectively Wild episodes they are changing over their defensive metrics but I forget the details. Probably trying to distance themselves from Lichtman given what was uncovered about him a few years ago. I hope they go with DRS so WAR will be more consistent across b-ref and fangraphs. Statcast might be better - I have kind of assumed that it is - but I don't think it has been proven. I don't know if WAR ever should have been mainstreamed, but since it already is happening, it should be more consistent. Those who want to dive deeper can plug in any numbers they want and make their own WAR. Quote
Tiger337 Posted 13 hours ago Posted 13 hours ago (edited) 5 minutes ago, gehringer_2 said: I agree stability is part - I think it over measures - I don't believe player defensive values ping-pong around from season to season as much as DRS does. So I'd argue your syntax a bit - I wouldn't say DRS stabilizes over time, you're just taking an average over time to create a stability that isn't actually there in the measurement. So in a way that gives it more credit than I would. But the other thing I question is when you put it all together I don't think the total team aggregates pass the smell test. Now it's fine that knowledgeable people understand the limitation of defensive metrics; they aren't easy to get a good handle on, so you do the best you can. The problem is when the defensive piece gets rolled into WAR and the world take that number as gospel. Yes, that is a problem. I don't think it's a good mainstream stat. To your other point, I don't think defensive ability necessarily changes so much year to year, but it's very possible that the small sample performance that we see does vary a lot. Defenders are much like relievers in that they don't have a large sample size of difficult plays over a season. Edited 12 hours ago by Tiger337 1 Quote
Arlington Posted 12 hours ago Posted 12 hours ago 13 hours ago, casimir said: I guess I look at it differently. I think tiers of teams rather than an absolute order. Both the Tigers and Dodgers are top tier, and I’d probably add the Astros, Phillies, Cubs, Mets, and Blue Jays in some semblance, although I could see where maybe the Mets and Blue Jays fall into the next tier. The Brewers are around in there, too. You're telling a different story. These are the power rankings and should reflect the best based on performance to date. As for being able to predict the future yours is more intelligent. It's like playoff odds. 162 games is a long season and it gives you a pretty good indication of how good a team is. When you take a playoff series for example, and you have the league's best team which won a little over .630 of their games and pit them against a team that won .520 of their games you end up with a result statistically of the better team winning the series less than 6 times out of ten and the lesser team winning close to 5 times out of ten, so yea. The lesser team has a much better chance than people tend to assign them. Anyone in your tier could beat anyone else in the tier and in any series the outcome pretty much could go either way. 1 Quote
gehringer_2 Posted 12 hours ago Posted 12 hours ago (edited) 20 minutes ago, Tiger337 said: To your other point, I don't think defensive ability necessarily changes so much year to year, but it's very possible that the small sample performance that we see does vary a lot. Defenders are much like relievers in that they don't have a large sample size of difficult plays over a season. Which also means it matters how you want to use the statistics. So player X in RF ends up having the opportunity to make some plays in a given season that elevate his DRS compared to player Y, who didn't have those chances, but (for the sake of argument) we know has better tools. If you look at the stats as a measure of accomplishment - like a hitter's counting stats, then we are on solid ground - the guy made the plays and no reason not to credit that. But if you want to predict (say for team building purposes) which player is going to give you better D in the *future*, you may find yourself looking at a number that is not the best predictor. Which is in line with your point about relievers - the parallel being that looking at their secondary stats is generally more predictive of what you are going to get going forward than their cumulative stats like ERA or number of Saves in any given point. Edited 12 hours ago by gehringer_2 Quote
chasfh Posted 12 hours ago Posted 12 hours ago 40 minutes ago, gehringer_2 said: So it makes sense that positional value of LF is lower in general, but does a guy get dinged for being a LF at COPA, which has higher positional value than most other parks? Aside from that, can't disagree Riley has a noodle arm. Yeah, probably, but the difference between Riley and Julio is so great that the Comerica Park left field alone almost certainly couldn't be enough to explain it. Quote
Tiger337 Posted 12 hours ago Posted 12 hours ago 32 minutes ago, gehringer_2 said: Which also means it matters how you want to use the statistics. So player X in RF ends up having the opportunity to make some plays in a given season that elevate his DRS compared to player Y, who didn't have those chances, but (for the sake of argument) we know has better tools. If you look at the stats as a measure of accomplishment - like a hitter's counting stats, then we are on solid ground - the guy made the plays and no reason not to credit that. But if you want to predict (say for team building purposes) which player is going to give you better D in the *future*, you may find yourself looking at a number that is not the best predictor. Which is in line with your point about relievers - the parallel being that looking at their secondary stats is generally more predictive of what you are going to get going forward than their cumulative stats like ERA or number of Saves in any given point. That kind of explains the difference between b-ref and fangraphs pitching WAR. b-ref focuses on past performannce while fangraphs is morte predictive. Both philosophies have their uses. But what if fielding performance does in fact change because a player loses a step? Then we don't know if the change is due to performance or variability. It has been shown that fielding performance peaks earlier than hitting. Quote
gehringer_2 Posted 11 hours ago Posted 11 hours ago (edited) 11 minutes ago, Tiger337 said: It has been shown that fielding performance peaks earlier than hitting. that's an interesting point, and I wonder if that's a general physiological thing, or just an artifact because if your hitting falls off first, you are out of the league, whereas if you lose a step but still hit you keep your job. So you might end up with an asymmetric distribution for fielding wrt hitting. And of course as guys train for strength over the years for hitting they can lose some elasticity for catching and throwing. I don't know if with modern training that's the issue it once was, but I'd guess it remains in play to some degree. Edited 11 hours ago by gehringer_2 Quote
Arlington Posted 9 hours ago Posted 9 hours ago Riley has a strange stride, with a stiff upright back while landing on his heels. I see a little VMart in him. I wonder if he could be trained to run differently. I read someone's opinion that he will age fast and may be a risky long-term signing, something VMart was not. Who knows where he is going to peak. It seems like he just put it together last year and he's like 25? Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.