Return to Baseball Leftovers

Player Evaluation Formula

Moderator: Baseball Moderators

Postby shortsavage » Tue Dec 05, 2006 3:18 pm

Right. That's kind of how you do it unless you want to come up with some sort of complex derivation...which would take me quite some time.

I think running a formula across every potential big league starter and then taking a look is a decent method.

I bet a lot of systems work like this, but their creators don't show you exactly how things work because then they open up the gates for floods of criticsm.
Mike Pelfrey > Matt Garza
shortsavage
General Manager
General Manager

User avatar
Cafe WriterWeb Supporter
Posts: 2931
(Past Year: -15)
Joined: 25 Aug 2003
Home Cafe: Baseball
Location: UC Davis

Postby shortsavage » Tue Dec 05, 2006 8:26 pm

Alright, here are the lists that I got when I weighted the years:

Catcher:
1. Joe Mauer, 15.92
2. Brian McCann, 14.58
3. Victor Martinez, 11.78
4. Michael Barrett, 11.43
5. Jorge Posada, 9.56
6. Mike Piazza, 7.69
7. Greg Zaun, 7.51
8. Russell Martin, 7.03
9. Ramon Hernandez, 6.86
10. Mike Napoli, 6.18


First Base:
1. Albert Pujols, 18.41
2. Lance Berkman, 12.20
3. David Ortiz, 12.10
4. Ryan Howard, 11.50
5. Todd Helton, 10.60
6. Jason Giambi, 8.24
7. Nick Johnson, 7.53
8. Justin Morneau, 7.49
9. Paul Konerko, 6.69
10. Carlos Delgado, 6.46


How do these rankings compare to some of the other ones you have seen? Should I charge forward with 2B, or do you think I still need some major tweakage?
Mike Pelfrey > Matt Garza
shortsavage
General Manager
General Manager

User avatar
Cafe WriterWeb Supporter
Posts: 2931
(Past Year: -15)
Joined: 25 Aug 2003
Home Cafe: Baseball
Location: UC Davis

Postby cordscords » Tue Dec 05, 2006 8:29 pm

I'm a little bit lost when you say "weight the years". Are you using 3 year averages?
cordscords
General Manager
General Manager

User avatar
Mock(ing) Drafter
Posts: 4340
(Past Year: 5)
Joined: 23 Jan 2005
Home Cafe: Baseball

Postby RugbyD » Tue Dec 05, 2006 9:40 pm

shortsavage wrote:Alright, here are the lists that I got when I weighted the years:

Catcher:
1. Joe Mauer, 15.92
2. Brian McCann, 14.58
3. Victor Martinez, 11.78
4. Michael Barrett, 11.43
5. Jorge Posada, 9.56
6. Mike Piazza, 7.69
7. Greg Zaun, 7.51
8. Russell Martin, 7.03
9. Ramon Hernandez, 6.86
10. Mike Napoli, 6.18


First Base:
1. Albert Pujols, 18.41
2. Lance Berkman, 12.20
3. David Ortiz, 12.10
4. Ryan Howard, 11.50
5. Todd Helton, 10.60
6. Jason Giambi, 8.24
7. Nick Johnson, 7.53
8. Justin Morneau, 7.49
9. Paul Konerko, 6.69
10. Carlos Delgado, 6.46


How do these rankings compare to some of the other ones you have seen? Should I charge forward with 2B, or do you think I still need some major tweakage?

if you can wait, i might have my standard deviation mthod done by tonight (unadjusted for outliers) if you want to compare relative scores.
RugbyD
Hall of Fame Hero
Hall of Fame Hero

User avatar
Cafe Ranker
Posts: 5589
(Past Year: -2)
Joined: 7 Dec 2004
Home Cafe: Baseball
Location: punting small dogs and being surly

Postby shortsavage » Tue Dec 05, 2006 9:58 pm

cordscords wrote:I'm a little bit lost when you say "weight the years". Are you using 3 year averages?


Yes. I'm looking at the last three years and using a 60/30/10 formula right now (75/25 for two years).
Mike Pelfrey > Matt Garza
shortsavage
General Manager
General Manager

User avatar
Cafe WriterWeb Supporter
Posts: 2931
(Past Year: -15)
Joined: 25 Aug 2003
Home Cafe: Baseball
Location: UC Davis

Postby RugbyD » Tue Dec 05, 2006 11:20 pm

RugbyD wrote:if you can wait, i might have my standard deviation mthod done by tonight (unadjusted for outliers) if you want to compare relative scores.

no gonna get to it tonight. forgot about Boston Legal
RugbyD
Hall of Fame Hero
Hall of Fame Hero

User avatar
Cafe Ranker
Posts: 5589
(Past Year: -2)
Joined: 7 Dec 2004
Home Cafe: Baseball
Location: punting small dogs and being surly

Postby RynMan » Wed Dec 06, 2006 3:25 am

RugbyD wrote:
shortsavage wrote:
RugbyD wrote:i'd be more interested to hear your reasoning behind the coefficients than to see the results. i'm also curious as to why you went with this kind of approach as opposed to using a simple sum of standard deviations across the catagories that will count for your league.


Kind of lame but the coefficients are mainly trial and error. I came up with a formula that I'm using for minor leaguers, so I've been running data through that for a while. I think OPS and BB/K are pretty important for major league success.

Ratios can be a lot more telling than countable stats. They don't vary as much, so I figured if I could cash them into a value, I'd be in good shape. Again, I got the idea for this approach after trying to find ways to rank prospects.

i guess that begs the question of how do you know if one trial is better or worse than the next aside from if it gives you a clear outlier result? it might lead to finding a formula the confirms your opinions of how rankings should look vs. one that answers a question of what they are.


This was my initial thought aswell.
Image
RynMan
Hall of Fame Hero
Hall of Fame Hero

User avatar
CafeholicFantasy ExpertCafe WriterCafe RankerAward-Winning Graphics ExpertMock(ing) DrafterEagle Eye
Posts: 8664
(Past Year: -227)
Joined: 4 Feb 2004
Home Cafe: Baseball
Location: Oztailia

Postby shortsavage » Wed Dec 06, 2006 1:06 pm

Have you guys heard of derived formulas?

I mean, it's not like I came up with them randomly...had a pretty good idea of what I wanted the formula to do.
Mike Pelfrey > Matt Garza
shortsavage
General Manager
General Manager

User avatar
Cafe WriterWeb Supporter
Posts: 2931
(Past Year: -15)
Joined: 25 Aug 2003
Home Cafe: Baseball
Location: UC Davis

Postby RugbyD » Wed Dec 06, 2006 1:26 pm

shortsavage wrote:Have you guys heard of derived formulas?

I mean, it's not like I came up with them randomly...had a pretty good idea of what I wanted the formula to do.

its just that i have trouble finding a reason to validate an equation without a statement like, "The coefficient for SB should be X because Y", without you having seen a result first.

If you were going to derive an equation to replicate past actual rankings over multiple years and back-test its predictive ability, that would be a different story. Without the prior validation, I can't find a reason to give it credibility without a particular reason for a coefficient being what it is.
RugbyD
Hall of Fame Hero
Hall of Fame Hero

User avatar
Cafe Ranker
Posts: 5589
(Past Year: -2)
Joined: 7 Dec 2004
Home Cafe: Baseball
Location: punting small dogs and being surly

Postby Tavish » Wed Dec 06, 2006 1:47 pm

shortsavage wrote:Have you guys heard of derived formulas?

I mean, it's not like I came up with them randomly...had a pretty good idea of what I wanted the formula to do.


What is it you want the formula to do?
Bury me a Royal.
Tavish
Mod in Retirement
Mod in Retirement

User avatar
CafeholicFantasy ExpertCafe WriterCafe RankerMock(ing) DrafterEagle EyeCafe SpotterWeb Supporter
Posts: 10553
(Past Year: -511)
Joined: 3 May 2004
Home Cafe: Baseball

PreviousNext

Return to Baseball Leftovers

Who is online

Users browsing this forum: Yahoo [Bot] and 17 guests

cron