Defensive Metrics Topic

Posted by bad_luck on 9/4/2013 4:39:00 PM (view original):
Posted by MikeT23 on 9/4/2013 4:35:00 PM (view original):
Pretty much every time you say "We need three years of data" or "It's the best we've got".      Do that sound like ringing endorsements to you?
You need three years of data, but you can still regress one year 50% and get a pretty good idea.
Why do we need 3 years of data?    Do we need three years of data to know Cabrera can hit the **** out of the ball?
9/4/2013 5:08 PM
Posted by tecwrg on 9/4/2013 4:49:00 PM (view original):
Posted by bad_luck on 9/4/2013 4:17:00 PM (view original):
Posted by tecwrg on 9/4/2013 3:08:00 PM (view original):
Adam Jones is generally regarded as anywhere from a somewhat above average to a very good defensive center fielder.

In 2012, his dWAR was -1.3.

Which is more likely to be wrong . . . all the folks who watch him play and regard him highly?  Or a number on a web page?
His UZR since he became a fulltime CF in Baltimore has ranged from a high of 6 in 2008 to a low of -8 in 2011. He has been in the range of -4 to -8 every year for the last 5 seasons (over 6000 innings). I'd say that, in reality, he is a below average center fielder and that the people that generally regard him as above average are wrong, which isn't unusual. How many gold gloves does Derek Jeter have?
So once again, your argument boils down to: people who watch baseball games and draw conclusions from what they see are wrong.

You are awesome.
You act like there are two groups, people who watch baseball games and people who like baseball stats. Not true.

The reality is that reputations and perceptions based on what people see are often wrong. This applies to a lot of things.

People see Adam Jones rob a home run and hear Harold Reynolds talk about what an amazing defender he is and they flag him in their mind as a great outfielder. Then they remember every great play they see him make, confirming the belief they already had.

Funny thing about this though, Fangraphs has a fan scouting report for every player. Every year they allow anyone that wants to to rate a player defensively based on what the fan sees. Jones has a negative total for the last four years. Maybe his reputation as a defender among the "people who watch baseball" isn't that great.
9/4/2013 5:10 PM
Posted by MikeT23 on 9/4/2013 5:08:00 PM (view original):
Posted by bad_luck on 9/4/2013 4:39:00 PM (view original):
Posted by MikeT23 on 9/4/2013 4:35:00 PM (view original):
Pretty much every time you say "We need three years of data" or "It's the best we've got".      Do that sound like ringing endorsements to you?
You need three years of data, but you can still regress one year 50% and get a pretty good idea.
Why do we need 3 years of data?    Do we need three years of data to know Cabrera can hit the **** out of the ball?
Nope, because he gets 700+ plate appearances. If he only got 150-200 PA a year, we'd need more than just one year.
9/4/2013 5:11 PM
Posted by bad_luck on 9/4/2013 4:32:00 PM (view original):
When did I use it then disregard it?
When U claimed MTrout must not be very good defensively,
because he was not viewed on ESPN web-gems, as much
as last season...

Best 2 keep disregard'n... The better path, 4-sure... Use no
ESPN-TV in no future math analysis, please...

Now, how much R those numbers, so that U can FINALLY
get around to disregard'n them, too ???... How much ???...
9/4/2013 6:20 PM (edited)
That's when your sales-pitch 1st made it 2 my site-mail box...
U got noticed... Lucky U...
9/4/2013 5:16 PM
How much ???
9/4/2013 5:17 PM
4 those proprietary numbers ???
9/4/2013 5:17 PM
How much ???... In American tacos, or similar something...
Value is value... How much ???
9/4/2013 5:19 PM
yawn.
9/4/2013 5:23 PM
thats the double xx man of wis. he speaks japanese, in spanish. his dollar here, is worth twenty thousand in other worlds.
9/4/2013 5:25 PM
ok. but if i ptch-in some bucks, then guarantees need to be dropped on the table here,
9/4/2013 5:26 PM
what do we all get for the money u need? exactly? i like exactly. no bull. no manure.
9/4/2013 5:28 PM
Posted by bad_luck on 9/4/2013 5:11:00 PM (view original):
Posted by MikeT23 on 9/4/2013 5:08:00 PM (view original):
Posted by bad_luck on 9/4/2013 4:39:00 PM (view original):
Posted by MikeT23 on 9/4/2013 4:35:00 PM (view original):
Pretty much every time you say "We need three years of data" or "It's the best we've got".      Do that sound like ringing endorsements to you?
You need three years of data, but you can still regress one year 50% and get a pretty good idea.
Why do we need 3 years of data?    Do we need three years of data to know Cabrera can hit the **** out of the ball?
Nope, because he gets 700+ plate appearances. If he only got 150-200 PA a year, we'd need more than just one year.
So it's a numbers game?  

Do all position players need 3 years of data?   Seems like certain positions get more opportunities.   Of course, because I'm not digging into a stat book when I say that, I'm just basing that on what I've seen.
9/4/2013 5:33 PM
Are you asking why sample sizes play a large role in the reliability of a stat?
9/4/2013 6:01 PM
What if the 'time-machine' only lets U go back to Babe Ruth's time,
back to the 1920's...

What if when U get there, the only baseball job u get is as a 'ump'
&, homeplate is all yours, evey 6th day (?)...

Will U '________________' ???

A.) Witness Babe Ruth hit monstrous amounts of homers, on balls
.     that weren't strikes ???...

B.) Stain your drawers ???

C.) Flash a hitch-hiking thumb @ a U.F.O. space-ship ???

D.) All of the above, in THAT order ???

9/4/2013 6:33 PM
◂ Prev 1...8|9|10|11|12...23 Next ▸
Defensive Metrics Topic

Search Criteria

Terms of Use Customer Support Privacy Statement

© 1999-2024 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.