Question for those who like math. Topic

alykaramazov
Posts: 121 (0)
Hall of Famer
5 9 2 2 2 175 11
Prospect
Posted by deanod on 12/6/2011 1:59:00 PM (view original):
i think it's nearly impossible to do any sort of analysis that isolates the value of PC in a meaningful sample.
Excellent point above BTW.

In terms of PC.  Someone told me that they had come up with .1 of ERA per 10 points of PC , but I don't remember who.  I have been using that number.
12/6/2011 2:01 PM
deanod
Posts: 3150 (0)
Hall of Famer
115 18 1 37 1 51
Prospect
and if you just wanted to look for correlations between PC and cERA in a vaccuum w/o isolating external variables, you'd probably need something retarded like 10,000 seasons worth of data for any conclusion to be considered remotely relevant.  there's just way too much noise that clusterfucks everything.
12/6/2011 2:02 PM
MikeT23
Posts: 58012 (0)
Hall of Famer
87 1 47 1 189
Prospect
There weren't many outliers.   I built all my teams pretty much the same way.   In those 20+ seasons, I think there was one example of the higher PC having the higher CERA.    I wouldn't say it was perfect as we're talking about 35-45 catchers but, as I said, they all fell pretty much in line.
12/6/2011 2:02 PM
boogerlips
Posts: 3731 (0)
Hall of Famer
204 45
Hall of Famer
Given enough sample size, all other variables cancel each other out. I've never bothered to go through the effort, but if you looked at teams with more than one catcher and look at their cERA, prorate their innings caught to total relevance, you'll eventually get a solid number.
12/6/2011 2:03 PM
boogerlips
Posts: 3731 (0)
Hall of Famer
204 45
Hall of Famer
Posted by deanod on 12/6/2011 2:02:00 PM (view original):
and if you just wanted to look for correlations between PC and cERA in a vaccuum w/o isolating external variables, you'd probably need something retarded like 10,000 seasons worth of data for any conclusion to be considered remotely relevant.  there's just way too much noise that clusterfucks everything.
This simply isn't true.
12/6/2011 2:04 PM
deanod
Posts: 3150 (0)
Hall of Famer
115 18 1 37 1 51
Prospect
Posted by alykaramazov on 12/6/2011 2:01:00 PM (view original):
Posted by deanod on 12/6/2011 1:59:00 PM (view original):
i think it's nearly impossible to do any sort of analysis that isolates the value of PC in a meaningful sample.
Excellent point above BTW.

In terms of PC.  Someone told me that they had come up with .1 of ERA per 10 points of PC , but I don't remember who.  I have been using that number.

that's way too high imo

if i was going to pull a number out of my ***, i'd say 10 points of PC = .03 of ERA.

i have nothing to support that, other than the fact that the rating is 100% broken if it has that significant of an impact.

12/6/2011 2:05 PM
deanod
Posts: 3150 (0)
Hall of Famer
115 18 1 37 1 51
Prospect
Posted by boogerlips on 12/6/2011 2:04:00 PM (view original):
Posted by deanod on 12/6/2011 2:02:00 PM (view original):
and if you just wanted to look for correlations between PC and cERA in a vaccuum w/o isolating external variables, you'd probably need something retarded like 10,000 seasons worth of data for any conclusion to be considered remotely relevant.  there's just way too much noise that clusterfucks everything.
This simply isn't true.
maybe 5000 seasons?  2000?  1000?  i don't know, but i'm pretty sure any sample of like 20 seasons is going to be 100% worthless.
12/6/2011 2:06 PM
deathinahole
Posts: 4139 (0)
Hall of Famer
76 2 1 42 1 72
Prospect
Posted by deanod on 12/6/2011 1:48:00 PM (view original):
Posted by deathinahole on 12/6/2011 1:31:00 PM (view original):
That's part of why I think PC and arm for a catcher are underrated.

I'm positive that in a pitchers park, the difference between an 85 and 50 PC with an above average arm all in is about half a run in ERA, which is 81 runs a season.

So you bat around the math (assuming that catcher is going to generate less than 64 runs, since they are usually lighter hitters), and it works most time.

It's the other 8 positions that I need to grasp now.
lol wut

first of all, the worse your pitching staff/more hitter friendly your park is, the more defense saves runs.  PC is going to be less valuable in a pitcher's park.

secondly, to save half a run per game that means that you need to shave 18 points of your batting average.  there is no ******* way in **** that 2 points in pitch calling save 1 pt of batting average.

all wifs will say is that a 10 point difference is not significant but a 40 point difference is.  i have no idea what the **** that means- maybe that every 10 points is worth 1 pt of batting average?  Half a point?  Who knows.

FWIW if you set 10 pts = 1 point of average, that means that the difference between a 50 and 85 is ~ 15 runs a season.
first of all, the worse your pitching staff/more hitter friendly your park is, the more defense saves runs.  PC is going to be less valuable in a pitcher's park - Correct, thus the qualifier. I have little data outside of that because I don't see many samples of the light hitting defensive catcher in a hitter's park. Don't think people can force themselves to do it. Counter inuitive, and all that.

secondly, to save half a run per game that means that you need to shave 18 points of your batting average.  there is no ******* way in **** that 2 points in pitch calling save 1 pt of batting average. - I'm saying that a 4.00 ERA becomes a 3.50 ERA IF it's a combination of PC and arm. Apples to apples.  What part affects what I can't separate. But I have enough that I'm confident in the difference +- .03.
12/6/2011 2:09 PM
deanod
Posts: 3150 (0)
Hall of Famer
115 18 1 37 1 51
Prospect
Posted by deathinahole on 12/6/2011 2:09:00 PM (view original):
Posted by deanod on 12/6/2011 1:48:00 PM (view original):
Posted by deathinahole on 12/6/2011 1:31:00 PM (view original):
That's part of why I think PC and arm for a catcher are underrated.

I'm positive that in a pitchers park, the difference between an 85 and 50 PC with an above average arm all in is about half a run in ERA, which is 81 runs a season.

So you bat around the math (assuming that catcher is going to generate less than 64 runs, since they are usually lighter hitters), and it works most time.

It's the other 8 positions that I need to grasp now.
lol wut

first of all, the worse your pitching staff/more hitter friendly your park is, the more defense saves runs.  PC is going to be less valuable in a pitcher's park.

secondly, to save half a run per game that means that you need to shave 18 points of your batting average.  there is no ******* way in **** that 2 points in pitch calling save 1 pt of batting average.

all wifs will say is that a 10 point difference is not significant but a 40 point difference is.  i have no idea what the **** that means- maybe that every 10 points is worth 1 pt of batting average?  Half a point?  Who knows.

FWIW if you set 10 pts = 1 point of average, that means that the difference between a 50 and 85 is ~ 15 runs a season.
first of all, the worse your pitching staff/more hitter friendly your park is, the more defense saves runs.  PC is going to be less valuable in a pitcher's park - Correct, thus the qualifier. I have little data outside of that because I don't see many samples of the light hitting defensive catcher in a hitter's park. Don't think people can force themselves to do it. Counter inuitive, and all that.

secondly, to save half a run per game that means that you need to shave 18 points of your batting average.  there is no ******* way in **** that 2 points in pitch calling save 1 pt of batting average. - I'm saying that a 4.00 ERA becomes a 3.50 ERA IF it's a combination of PC and arm. Apples to apples.  What part affects what I can't separate. But I have enough that I'm confident in the difference +- .03.
There's no way that you're nearly adept enough with statistics to assign yourself such a small margin for error.  i don't think you're even in the ballpark.

if you really believe that to be true, show your work.
12/6/2011 2:12 PM
MikeT23
Posts: 58012 (0)
Hall of Famer
87 1 47 1 189
Prospect
Posted by deanod on 12/6/2011 2:06:00 PM (view original):
Posted by boogerlips on 12/6/2011 2:04:00 PM (view original):
Posted by deanod on 12/6/2011 2:02:00 PM (view original):
and if you just wanted to look for correlations between PC and cERA in a vaccuum w/o isolating external variables, you'd probably need something retarded like 10,000 seasons worth of data for any conclusion to be considered remotely relevant.  there's just way too much noise that clusterfucks everything.
This simply isn't true.
maybe 5000 seasons?  2000?  1000?  i don't know, but i'm pretty sure any sample of like 20 seasons is going to be 100% worthless.
Patterns are patterns.   I'm pretty sure you've said "If I guy is hitting .300 on his career, he's probably a .300 hitter so you can look at past performance to see what a player is."   How many players ahave 20+ seasons under their belt?

Do you always get pissy when someone runs some formulas that appear to show a pattern before you do? 
12/6/2011 2:13 PM
deanod
Posts: 3150 (0)
Hall of Famer
115 18 1 37 1 51
Prospect
Posted by MikeT23 on 12/6/2011 2:13:00 PM (view original):
Posted by deanod on 12/6/2011 2:06:00 PM (view original):
Posted by boogerlips on 12/6/2011 2:04:00 PM (view original):
Posted by deanod on 12/6/2011 2:02:00 PM (view original):
and if you just wanted to look for correlations between PC and cERA in a vaccuum w/o isolating external variables, you'd probably need something retarded like 10,000 seasons worth of data for any conclusion to be considered remotely relevant.  there's just way too much noise that clusterfucks everything.
This simply isn't true.
maybe 5000 seasons?  2000?  1000?  i don't know, but i'm pretty sure any sample of like 20 seasons is going to be 100% worthless.
Patterns are patterns.   I'm pretty sure you've said "If I guy is hitting .300 on his career, he's probably a .300 hitter so you can look at past performance to see what a player is."   How many players ahave 20+ seasons under their belt?

Do you always get pissy when someone runs some formulas that appear to show a pattern before you do? 
lol at that comparison.

the more apt comparison is that if we didn't have access to any hitting stats, just total runs scored with each respective hitter in the lineup.  and then you had to guess somebody's batting average from THAT.

i'm really not ahead of any curve in making formulas, i just realized that my defensive estimates were full of **** and that chuck jr had done a better analysis.  the analysis that people are throwing about on pitch calling are complete ***.  if there was a way to estimate it without spending 100 hours, i would have already done it and much better so than you or death.
12/6/2011 2:18 PM
deathinahole
Posts: 4139 (0)
Hall of Famer
76 2 1 42 1 72
Prospect
Posted by deanod on 12/6/2011 2:12:00 PM (view original):
Posted by deathinahole on 12/6/2011 2:09:00 PM (view original):
Posted by deanod on 12/6/2011 1:48:00 PM (view original):
Posted by deathinahole on 12/6/2011 1:31:00 PM (view original):
That's part of why I think PC and arm for a catcher are underrated.

I'm positive that in a pitchers park, the difference between an 85 and 50 PC with an above average arm all in is about half a run in ERA, which is 81 runs a season.

So you bat around the math (assuming that catcher is going to generate less than 64 runs, since they are usually lighter hitters), and it works most time.

It's the other 8 positions that I need to grasp now.
lol wut

first of all, the worse your pitching staff/more hitter friendly your park is, the more defense saves runs.  PC is going to be less valuable in a pitcher's park.

secondly, to save half a run per game that means that you need to shave 18 points of your batting average.  there is no ******* way in **** that 2 points in pitch calling save 1 pt of batting average.

all wifs will say is that a 10 point difference is not significant but a 40 point difference is.  i have no idea what the **** that means- maybe that every 10 points is worth 1 pt of batting average?  Half a point?  Who knows.

FWIW if you set 10 pts = 1 point of average, that means that the difference between a 50 and 85 is ~ 15 runs a season.
first of all, the worse your pitching staff/more hitter friendly your park is, the more defense saves runs.  PC is going to be less valuable in a pitcher's park - Correct, thus the qualifier. I have little data outside of that because I don't see many samples of the light hitting defensive catcher in a hitter's park. Don't think people can force themselves to do it. Counter inuitive, and all that.

secondly, to save half a run per game that means that you need to shave 18 points of your batting average.  there is no ******* way in **** that 2 points in pitch calling save 1 pt of batting average. - I'm saying that a 4.00 ERA becomes a 3.50 ERA IF it's a combination of PC and arm. Apples to apples.  What part affects what I can't separate. But I have enough that I'm confident in the difference +- .03.
There's no way that you're nearly adept enough with statistics to assign yourself such a small margin for error.  i don't think you're even in the ballpark.

if you really believe that to be true, show your work.
Well, the BMath says yea, I am adept enough, and the show your work says "do it yourself". Proprietary, *****.
12/6/2011 2:18 PM
deathinahole
Posts: 4139 (0)
Hall of Famer
76 2 1 42 1 72
Prospect
Now, I'm careful to separate that from "I know what I'm doing with this game", because there are plenty better at me at it. Thus, I don't open my yap with your other analysis because I just don't know, and frankly, my free time is limited to how much I pour over this game.

That one thing, I'm sure of.
12/6/2011 2:21 PM
deanod
Posts: 3150 (0)
Hall of Famer
115 18 1 37 1 51
Prospect
death, if your math is correct that means that pitch calling alone (not even counting throwing runners out and blocking balls in the dirt) is worth more than the difference between a GG SS and like a slightly above average 3B playing SS.

i don't have any empirical evidence to suggest that it can't be the case, but if the developers stressed pitch calling that much they are completely braindead.  a lot of things are broken about the game but i feel like even if it started out that extreme, at some point they'd look at the formula and say "well, this makes no ******* sense, we should tone it down a bit."

edit: yes, i do think soulreading how much the weight the developers should have assigned the ratings is worth more than any analysis that somebody whipped together in an hour or two.
12/6/2011 2:23 PM (edited)
deathinahole
Posts: 4139 (0)
Hall of Famer
76 2 1 42 1 72
Prospect
Yea, I'm saying that. That's why I try to jump in cheap on them. And I try to have two so I get PC and arm, pinch hitter, PC and arm again to take away one ****** at bat.

One day, I'll sit down and figure out the rest of the game.
12/6/2011 2:24 PM
◂ Prev 1|2|3|4|5|6...9 Next ▸
Question for those who like math. Topic
△Top△Bottom

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.