Honestly the only "relevant" bigs who stand out at SF after a quick glance in version 2 are Bill Russell, Amare Stoudemire, Karl Malone, David Lee, Kevin McHale, Moses Malone and Ben Wallace. Someone please point me out if I'm wrong, which I could be.

Posted by seble on 9/25/2010 1:04:00 AM (view original):
It's not possible to affect one position without affecting all positions.  For example, changing the odds of a center being able to play SF also affects the odds of a PG/SG/PF of playing SF.  I can find a middle ground between the last two versions that may work better, but there will have to be trade-offs.  Either we sacrifice a few guys being able to play something they maybe shouldn't, or vice versa.  I don't mind making a few manual changes where it's clearly necessary, but I'm not going to do that for 50-100 guys. 

Obviously there is a split over whether it's better to be more restrictive or less.  At this point I will play with it a little more to find a sweet spot, but it appears impossible to please everyone.

I don't understand how it's "not possible to affect one position without affecting all positions".  Aren't you defining the formula right now?  Make it so that it is possible.  As ashamael said, playing 1-2-3 should be more likely than playing 3-4-5 (since 4 and 5 are post positions and 1 through 3 are perimeter positions).  If you're looking for an easy, straightforward formula, I'm starting to think you should just make everyone who played guard eligible to play 1, 2, and 3, because really, what's the fear here?  If someone wants to play a real point guard at the 3, they'll be sacrificing rebounds anyway, and if someone wants to play a natural 2 or 3 at point guard, they'll be sacrificing assists.  Just make it more flexible (within a certain amount of reason, I guess) - people should be allowed to experiment with ridiculous lineups.

Exactly. detlef and ashamael have addressed the major concerns for what some users are calling "too flexible". Either they sacrifice the rebounds or assists with the 1s, 2s, and 3s playing out of position. Or the rebounding bigs will cost too much to play at the 3 due to the salary changes. FWIW, thats just the way I see it, and they seem to as well.

Seble, thanks for seeking input from the users.

Posted by seble on 9/25/2010 1:04:00 AM (view original):
It's not possible to affect one position without affecting all positions.  For example, changing the odds of a center being able to play SF also affects the odds of a PG/SG/PF of playing SF.  I can find a middle ground between the last two versions that may work better, but there will have to be trade-offs.  Either we sacrifice a few guys being able to play something they maybe shouldn't, or vice versa.  I don't mind making a few manual changes where it's clearly necessary, but I'm not going to do that for 50-100 guys. 

Obviously there is a split over whether it's better to be more restrictive or less.  At this point I will play with it a little more to find a sweet spot, but it appears impossible to please everyone.

If "changing the odds of a center being able to play SF also affects the odds of a PG/SG/PF playing SF" then you need to change the formula because this should not be.  I really don't see why this is being made to be so complicated when it isn't.  I can use an excel spreadsheet to come up with position effectiveness for all players that is consistent.

As I said I will tinker with it a little more, but that's probably the final review phase.  I think I have enough feedback to make a final version.  Expect the release sometime next week.

Posted by seble on 9/25/2010 4:27:00 PM (view original):
As I said I will tinker with it a little more, but that's probably the final review phase.  I think I have enough feedback to make a final version.  Expect the release sometime next week.

"Expect the release sometime next week."

Heh.

Posted by seble on 9/25/2010 4:27:00 PM (view original):
As I said I will tinker with it a little more, but that's probably the final review phase.  I think I have enough feedback to make a final version.  Expect the release sometime next week.

Thanks for the opportunity to provide several rounds of input.  And I think you're right - the bunch of us will never be satisfied because a) with an engine this complex trying to replicate reality we'll always find something to pick at, and b) we don't necessarily agree with each other.

Yes, Seble, thank you for the opportunity to provide feedback before the release and for listening to and using said feedback.  I can't speak for everyone, but I greatly appreciate it.

Yes thanks Seble..

Along the lines what Ash was saying how about something like this for the bigs:

1) Anyone with a PF designation cannot be any higher than 90% at SG
2) Anyone with a C designation cannot be any higher than 90% at SF

For guards/ small forwards:

1) Any SG 6'6" and over will be 100% at SG/SF.  With exceptions to 88-89 MJ at 100% at PG and Lebron being 100% at PF
2) Any SG 6'5" and under will be 100% at SG/PG.  With exception to Magic being 100% at SF 
3) Any SF 6"6" and over will be 100% at PF not at SG.  With exception to Pippen
4) Anf SF 6'5" and under will be 100% at SG. 

Let me know what you think and if anyone can think of any other exceptions

The above suggestions look great.

Moses should certainly not be 100% at SF.

Just so we're clear - are you also using resources like 82games.com besides just "listed position"?  If a guy played SG for 10% of his team's total minutes but is listed as a SF (even if he doesn't meet whatever other requirements you're using, like height, etc), he should get 100% SG.  I don't know what the right % of minutes at a position should be, but I'd be in favor of being fairly liberal with it.

Tyreke Evans walks into a bar ... "Wanna hear a joke?" says the bartender? ... "Sure," says Tyreke ... "Come back next week," says the bartender.

Nice improvements. THANKS SEBLE.

Don't understand why Larry Bird's base position is PF in 79-80 and 83-84, making him 100% at C instead of SF when the basic personnel on the Celtics did not change. 

Again, there really needs to be consistency over the years.