All Forums > MLB > Defensive Metrics
9/3/2013 9:37 PM
>> Expected out rate: How often balls hit to that area are turned into outs (data through Aug. 28)
9/3/2013 10:23 PM
Posted by lad_buck on 9/3/2013 9:37:00 PM (view original):
>> Expected out rate: How often balls hit to that area are turned into outs (data through Aug. 28)
this sentence again can be reduced to the first 3 words, (expected out rate), proving when tested that it does not meet the expected number related to something called an out rate, when applied to reviewing baseball numbers and how to categorize them.

u are not close here, in prioritizing number values, in standard equations.

the priority is pitching first. your numbers presume no change in the predominant, first application of innings pitched, by right-handers and left-handers, and the chances each fielder has when lefties bat, or when righties bat. your numbers do not have any direction when u give mike scioscia power to place his outfielders differently than how joe torre would play them. walter alston and tommy lasorda went with pitching, and kept their jobs for a long, long time.

pitchers alone give every manager the expected out rate, in their simple numbers, that determine a teams success. why set expectations with a grid, with vast mlb experience of managers doing it differently? when will u quit with these expected expectations?

9/3/2013 10:26 PM
Posted by lad_buck on 9/3/2013 9:37:00 PM (view original):
>> Expected out rate: How often balls hit to that area are turned into outs (data through Aug. 28)
defensive metrics dismiss these matters. bonds in the batters box with 3 infielders on the first base side, and a 3rd baseman at shortstop. your numbers reflect no mark mcgwire, facing 3 infielders blocking his view of the left field fence. your numbers do not reflect a first baseman covering second base. and that your numbers entirely ignore pull-hitters, dead-pull-hitters, spray hitters, flyball hitters, and the most simple numbers of all, that say everything typed below.

has anyone ever showed u where u could find the same evaluative info, contained in simply measuring 3 simple things. righties hit with greater predominance to left field and center field, and less frequency to right field. opposite with left handed batters hitting to the first base side. an example would be joey votto, and his errors only when lefties are on the mound, and his errors with right handers on the mound. joey votto is having some error tallies based on who is pitching and who is batting. lefties or righties. it tells everything. not everyone uses numbers to categorize joey votto.

simple numbers create simple expectations. your complex numbers cause u to expect differently, how outs should be credited, as each season is not an assembly line of programmable robots. left, center, and right. very simple to track when simplified. right handers and left handers are the only numbers that tell the whole story and what goes into the simple equation determining defense. chances and errors is still a fundamental battle explained by the predominance of pitching.

i dont even see a "predicted, and expected" out rate as even having a 30% to 40% success rate at determining whether a ball could be caught, or not. u are simply trapped in a whirlpool. u pretend that its solid ground, even while standing in quicksand with numbers that weigh u down, and leave u empty.
9/4/2013 12:44 AM
Posted by bad_luck on 9/3/2013 4:18:00 PM (view original):
Posted by burnsy483 on 9/3/2013 2:32:00 PM (view original):
Posted by bad_luck on 9/3/2013 1:55:00 PM (view original):
Of course there are tons of variables. All chances in the 5-50% range aren't counted the same. The chart is just a basic summary.

Right.  I'd like to see it broken down more.  Because when 50 of those 83 chances last year are in the 40-50% range, and 20 of those 60 chances are in the same range this year, it may explain a lot.

The idea that he had to write an article to explain why his defensive numbers are down suggest that it isn't obvious to the fan who watches Angels games why the number is lower.  Meaning there's a good chance he's playing just as well this year as he did last year, and the hitters are simply "hitting them where he ain't."

I'm not confident that even an avid Angels fan who watches all the games on TV would be able to tell the difference just by watching and remembering.

Just in general, People tend to remember things that are remarkable (Trout making a diving catch as opposed to pulling up and letting a ball drop) and that confirm their existing beliefs.

Specific to this situation, I don't know how well you can judge an outfielder's defense on TV. You only see the end result of all but the most spectacular plays and you never have a frame of reference. I don't think I can tell by watching what balls turn into outs 20% of the time and which ones turn into outs 40% of the time. Yet there's a difference in defensive value between the outfielder who can get to both and the outfielder who can only get to the 40% ball.
Specific to this solution, the last paragraph begins... Hoped that
1-day, a denouncing of the 'Web-Gem" analysis that didn't prove
a thing, in nightly lavish replay sessions for 7 months last year...

Yes, one of the 12-Steps of recovery might be happening, B-4 my
my very eyes... Confessing finally that he does NOT think he can
evaluate outfielders defense on TV...

Well, it's been a laff, if nothing else... The odds of relapse loom...
9/4/2013 10:39 AM
Posted by bad_luck on 9/3/2013 4:18:00 PM (view original):
Posted by burnsy483 on 9/3/2013 2:32:00 PM (view original):
Posted by bad_luck on 9/3/2013 1:55:00 PM (view original):
Of course there are tons of variables. All chances in the 5-50% range aren't counted the same. The chart is just a basic summary.

Right.  I'd like to see it broken down more.  Because when 50 of those 83 chances last year are in the 40-50% range, and 20 of those 60 chances are in the same range this year, it may explain a lot.

The idea that he had to write an article to explain why his defensive numbers are down suggest that it isn't obvious to the fan who watches Angels games why the number is lower.  Meaning there's a good chance he's playing just as well this year as he did last year, and the hitters are simply "hitting them where he ain't."

I'm not confident that even an avid Angels fan who watches all the games on TV would be able to tell the difference just by watching and remembering.

Just in general, People tend to remember things that are remarkable (Trout making a diving catch as opposed to pulling up and letting a ball drop) and that confirm their existing beliefs.

Specific to this situation, I don't know how well you can judge an outfielder's defense on TV. You only see the end result of all but the most spectacular plays and you never have a frame of reference. I don't think I can tell by watching what balls turn into outs 20% of the time and which ones turn into outs 40% of the time. Yet there's a difference in defensive value between the outfielder who can get to both and the outfielder who can only get to the 40% ball.
Correct, we don't have access to the data that many of us use to judge a player's performance.  Doesn't that seem off?  We have a relatively new stat that a lot of people are drooling over, and most of them don't know how the stat is generated, and we don't have access to the same data the "stat-makers" have.  The stat also seems off from time to time, and I'm essentially not allowed to figure out why.

Again, taking these numbers with a grain of salt.
9/4/2013 10:50 AM
But it's all we have!!!!
9/4/2013 11:08 AM
Well, we can actually WATCH the games.
9/4/2013 11:11 AM
No, that's not feasible.   And it's just crazy talk.
9/4/2013 11:12 AM
Posted by burnsy483 on 9/4/2013 10:39:00 AM (view original):
Posted by bad_luck on 9/3/2013 4:18:00 PM (view original):
Posted by burnsy483 on 9/3/2013 2:32:00 PM (view original):
Posted by bad_luck on 9/3/2013 1:55:00 PM (view original):
Of course there are tons of variables. All chances in the 5-50% range aren't counted the same. The chart is just a basic summary.

Right.  I'd like to see it broken down more.  Because when 50 of those 83 chances last year are in the 40-50% range, and 20 of those 60 chances are in the same range this year, it may explain a lot.

The idea that he had to write an article to explain why his defensive numbers are down suggest that it isn't obvious to the fan who watches Angels games why the number is lower.  Meaning there's a good chance he's playing just as well this year as he did last year, and the hitters are simply "hitting them where he ain't."

I'm not confident that even an avid Angels fan who watches all the games on TV would be able to tell the difference just by watching and remembering.

Just in general, People tend to remember things that are remarkable (Trout making a diving catch as opposed to pulling up and letting a ball drop) and that confirm their existing beliefs.

Specific to this situation, I don't know how well you can judge an outfielder's defense on TV. You only see the end result of all but the most spectacular plays and you never have a frame of reference. I don't think I can tell by watching what balls turn into outs 20% of the time and which ones turn into outs 40% of the time. Yet there's a difference in defensive value between the outfielder who can get to both and the outfielder who can only get to the 40% ball.
Correct, we don't have access to the data that many of us use to judge a player's performance.  Doesn't that seem off?  We have a relatively new stat that a lot of people are drooling over, and most of them don't know how the stat is generated, and we don't have access to the same data the "stat-makers" have.  The stat also seems off from time to time, and I'm essentially not allowed to figure out why.

Again, taking these numbers with a grain of salt.
Yes, the best we get is just a peak under the hood. ESPN/Fangraphs/etc will tell us exactly how they calculate UZR and DRS but they won't give us the Baseball Info Solutions play by play data because it is owned by BIS. We have to take it on faith that the data is reliable.

I don't have a huge problem with that. It's certainly better than using fielding percentage or relying on Gold Glove voters. But we have to accept it within context and know that, for example, one year of UZR isn't going to have much predictive power going forward and that to get a true idea of how good someone is in the field you need a three year average.
9/4/2013 11:12 AM
9/4/2013 11:13 AM
I do like the idea of trying to using stats to judge a fielder's performance.  This is a step in the right direction, but it needs to be improved significantly, in my opinion.
9/4/2013 11:13 AM
Posted by tecwrg on 9/4/2013 11:08:00 AM (view original):
Well, we can actually WATCH the games.
You can tell how much value an outfielder has defensively by watching games on TV?
9/4/2013 11:14 AM
"...for example, one year of UZR isn't going to have much predictive power going forward and that to get a true idea of how good someone is in the field you need a three year average."

Great, so let's ignore Trout's defensive stats until after next year.
9/4/2013 11:14 AM
Posted by bad_luck on 9/4/2013 11:13:00 AM (view original):
Posted by tecwrg on 9/4/2013 11:08:00 AM (view original):
Well, we can actually WATCH the games.
You can tell how much value an outfielder has defensively by watching games on TV?
I can tell you if he's a good or bad fielder(at least on that day).   You can't?
9/4/2013 11:15 AM
Posted by MikeT23 on 9/4/2013 11:14:00 AM (view original):
Posted by bad_luck on 9/4/2013 11:13:00 AM (view original):
Posted by tecwrg on 9/4/2013 11:08:00 AM (view original):
Well, we can actually WATCH the games.
You can tell how much value an outfielder has defensively by watching games on TV?
I can tell you if he's a good or bad fielder(at least on that day).   You can't?
Right, exactly.  I know a good fielder and a bad fielder when I see one.  I can tell a good defensive play from a bad defensive play.
of 23
All Forums > MLB > Defensive Metrics

Popular on WhatIfSports site: Baseball Simulation | College Basketball Game | College Football Game | Online Baseball Game | Hockey Simulation | NFL Picks | College Football Picks | Sports Games