This post could not be converted. To view the original post's thread, click here.

Ath is undervalued for a sg. Speed is underated for a pf... per undervalued for a pf... I feel defense is much more important at the pg position then any other.

aaarg sorry cant get format to include last row

I might suggest putting a bit more weight into athleticism at SG. It's otherwise fairly solid, but it doesn't account for SG's with high speed and athleticism, but low PER ratings, doing well. This happens, especially at D3 and D2, and IMO, it's due to the ATH-speed simulating an effective dribble-drive player.

Quote: Originally Posted By fd343ny on 12/29/2009aaarg sorry cant get format to include last ro

If only they'd roll out the new forums...

Quote: Originally Posted By achampa1 on 12/29/2009I might suggest putting a bit more weight into athleticism at SG. It's otherwise fairly solid, but it doesn't account for SG's with high speed and athleticism, but low PER ratings, doing well. This happens, especially at D3 and D2, and IMO, it's due to the ATH-speed simulating an effective dribble-drive player

I agree fully. The problem is that then one gets into roster construction issues. Would one want a roster filled just with that sort of guard? probably not. But one or two guards like that are fine, so long as you have some other guys who can hit from outside.......which is why I think this kind of approach to measuring team strength is not productive.

Quote: Originally Posted By fd343ny on 12/29/2009

Quote: Originally Posted By achampa1 on 12/29/2009
I might suggest putting a bit more weight into athleticism at SG. It's otherwise fairly solid, but it doesn't account for SG's with high speed and athleticism, but low PER ratings, doing well. This happens, especially at D3 and D2, and IMO, it's due to the ATH-speed simulating an effective dribble-drive player.

I agree fully. The problem is that then one gets into roster construction issues. Would one want a roster filled just with that sort of guard? probably not. But one or two guards like that are fine, so long as you have some other guys who can hit from outside.......which is why I think this kind of approach to measuring team strength is not productive.

Great observation! haha. I like the idea of trying to do it but it just doesn't work out well when it is done. Also you'd need numerous more tables for each of the O/D sets...

Quote: Originally Posted By fd343ny on 12/29/2009

Quote: Originally Posted By achampa1 on 12/29/2009
I might suggest putting a bit more weight into athleticism at SG. It's otherwise fairly solid, but it doesn't account for SG's with high speed and athleticism, but low PER ratings, doing well. This happens, especially at D3 and D2, and IMO, it's due to the ATH-speed simulating an effective dribble-drive player.

I agree fully. The problem is that then one gets into roster construction issues. Would one want a roster filled just with that sort of guard? probably not. But one or two guards like that are fine, so long as you have some other guys who can hit from outside.......which is why I think this kind of approach to measuring team strength is not productive

Exactly.

Speed is more important for bigs if pressing. Def less important in a 2-3, etc. I don't think this is something that can be done in the abstract.

If you are doing this, you might want to look up Zhawk's post on the average ratings for division one and try plugging some of those in to see what happens, and also, at least based on what the averages for each position in the power six conferences are - which are going to be heavily human populated with experienced coaches - you can get sort of a consensus idea of what coaches consider most important at each position, in the rough.

I think ath is woefully under represented for guards,

I have done this sort of thing before, it is best to throw out a trial balloon, let excel have at a couple of doz diff teams and adjust as ratings / results come in.

BUT, i think most ratings work by division, with the formula something like this for d1:

ATH for PG = (ATH-50)*.05 or speed (SP-70)*.15, etc

I have tried to use math to model ratings to performance, the trouble is without the opponent ratings for each game, your own team models only paint a part of the picture. For those who play strat o matic, i often theorize that offensive player ratings account for exactly one half of the touches a player makes, the defenders for the other half - not sure if that makes sense to others, essentially roll one red dice, if 1-3 comes up I look at the off guy, if 4-6 comes up I look at the defense. This explains why sometimes some pretty great guys / teams don't do so well, if they play off the defenders card an abnormal number of times??????

ah. strat!

and I am convinced that ATH is too low in what I did - should be more like .08 maybe - but then what to cut? maybe WE?

indeed, should WE be relevant at all since this is not a depiction of roster quality but a snapshot of current strength? so maybe zero out the WE entries?