Player Review Phase 2 - Position Effectiveness Topic

Just uploaded a new spreadsheet (www.whatifsports.com/db/nba_poss_eff.zip)...changes below.

- Instead of comparing a player's size to a historical average for the position, I'm now comparing them to the average for that position in the season they played.  That should work better, especially for the guys from the 50s and 60s, where players were much smaller and for the recent seasons, where players are larger.

- I increased the weighting on listed position, which should reduce the ability for most players to play 2-3 positions away from their listing.

I think this solved the problem of big guys playing guard for the most part. 


9/23/2010 3:08 PM
just go ahead and release it...the only way youre going to get it right is to do it manually....it will always be flawed if you use a system

magic doesnt play point some seasons....wilt can play pf but shaq cant...peja cant play sg....and just looking at my sixers team it doesnt have thad young able to play pf, which he played almost more then sf...thats one of im sure many many mistakes....youre never gonna get it right.....just release it already
9/23/2010 3:35 PM
Just because a guy is less than 100% doesn't mean he should never play a position.  You probably wouldn't see any difference in production for a guy in the upper 90s compared to 100%.  Also, Thaddeus Young is 100% at SF and PF for all 3 of his seasons.  There may be a rounding issue in what you see in the spreadsheet.
9/23/2010 4:16 PM
I really dig this for the most part.  Only a couple of glaring issues to me:

several versions of Magic Johnson need to be fixed (not 100% at pg - only one I can reasonably agree with him not being is his comeback in 95-96, and even then he should be upper 90s if not 100)

I don't understand how Shaq isn't 100% at PF when so many other centers (that weren't before) are now (Wilt, Kareem, Gilmore, David Robinson, Mutombo, ...)

I'm actually going to stop right there.  How is Mutombo justified but not Shaq?  Same era, relatively same height... Not like it would be overpowered in any way (Wilt being 100% at both is far more powerful than anyone else - and I'm not complaining about that just fyi).  My logical mind can understand it, yet I don't see how a system that produces Dikembe Mutombo at 100% PF will put Shaq in the mid to upper 90s.  The only thing I can figure is weight...

I've run through 50-something players I use a lot or want to use (but haven't been able to because they've been irrelevant for a while) and those are the only two issues I've seen.  shaq isn't that big of one since all the 100% center-only guys can now play PF... But the Magic one bothers me A LOT.



9/23/2010 4:48 PM
I'm combining height/weight when looking at size, so heavy guys like Shaq will be less likely to play at smaller positions than a skinny guy like Mutombo.  If necessary I don't mind making a few manual changes, such as making Magic 100% at PG.  If there are only a few guys like that just start compiling a list.
9/23/2010 5:09 PM
Hi Seble. Thanks for the hard work. I appreciate your time working on the position effectiveness issue.

I have one main concern after a brief look at the spreadsheet: Guys like Calvin Natt, Elgin Baylor, Shawn Marion, George McGinnis, Donyell Marshall and Red Robbins are all at 100 percent effectiveness at the SG position. Also Elvin Hayes is at % 95 at the SG position and Truck Robinson is at % 99 at the SG position.

My other concern is players like Buck Wiliams and Spencer Haywood are at %100 effectiveness at the SF position.

Thanks
9/23/2010 5:13 PM
This spreadsheet looks much more realistic and I think it's ready to be released finally.
9/23/2010 10:53 PM
Bill Russell should not be 100% at SF and Wes Unseld should not be 98% or 99% at SF.
9/24/2010 3:05 AM
Posted by vancem on 9/23/2010 5:13:00 PM (view original):
Hi Seble. Thanks for the hard work. I appreciate your time working on the position effectiveness issue.

I have one main concern after a brief look at the spreadsheet: Guys like Calvin Natt, Elgin Baylor, Shawn Marion, George McGinnis, Donyell Marshall and Red Robbins are all at 100 percent effectiveness at the SG position. Also Elvin Hayes is at % 95 at the SG position and Truck Robinson is at % 99 at the SG position.

My other concern is players like Buck Wiliams and Spencer Haywood are at %100 effectiveness at the SF position.

Thanks
Hey Seble

Having a more generous position effectiveness is great but we can't go over board and make it unrealistic.  I agree with all of this post plus if there's some seasons where Magic is not 100% PG that needs to be fixed
9/24/2010 5:55 AM
I agree with all that's been pointed out.  I'm a big one for simulations like this reflecting real life as much as possible, whether you focus on positions players actually played, or positions that players obviously could have played if needed without noticeable handicap. 

Very few point guards were capable of playing small forward - Magic and Oscar for sure, and I'll turn a blind eye to Kidd at SF.  Very few players were capable of playing both C and SF with reasonable effectiveness.  Rasheed Wallace and Cliff Robinson come to mind, for many or most seasons.  Also Kevin Garnett. 

Bigs like K. Malone, Brand and McHale being 100% at SF has been one of my pet peeves for a couple of years.  If it becomes easier to load up unrealistically on rebounds at the SF spot, then I worry that we'd see even less of many of history's best SFs (Barry, Havlicek, Wilkins, King, etc.).  That would make the game - to me - less interesting.
9/24/2010 7:51 AM
Version 3 released with new pinned thread.
9/24/2010 9:31 AM
Thanks seble. I like the positions being more restricted in the way that you did it. GREAT JOB!
9/29/2010 2:50 PM
Player Review Phase 2 - Position Effectiveness 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.