Posted by dahsdebater on 7/17/2012 3:01:00 PM (view original):

Posted by jtt8355 on 7/16/2012 7:44:00 PM (view original):
i only play d2, so keep that in mind if comparing to posts above. I +1 in 4 situations (NEVER +2):

1. if i know (read: can make an educated guess) that a team is going to cheat back (-3 or more) against me and i want to keep them honest, i'll +1 one of my guards.
2. in a similar vein, if i want to avoid or take advantage of a potential shift in half time settings (i.e., my bigs are loaded with distro, and i know that either opp adjusts every half time or if i'm winning, etc), i may be able to exploit a 2nd half opening), so i'll +1 my best 3 pt shooter.
3. when i have a player that's unathletic (< 40) but fast (70+) with good per (70+) matched up against a more athletic but slower defender, i'll +1.
4. Against the 2-3 zone i like to put in a per shooting sf and +1 one or both of my guards.

I can't believe multiple people have already agreed with #3 on this list.  I - and many coaches who are much better than me and whose opinions you actually care about - absolutely do not believe that individual ratings are compared for defensive matchups.  Telling people to compare ath or speed is reinforcing arguably the most prevalent major misunderstanding of how the sim works out there.

Even for m2m? So what you're saying that even in defenses other than zone, the def rating is averaged correct?

No, not even close to correct... What I'm saying is that the sim engine doesn't compare the ATH of the guy shooting to the ATH of the guy defending him, nor the speed.  Instead it looks at the guy taking the shot and comes up with some overall offensive score, which very likely is just an expectation value for making the shot in question.  So for a 3-point shooter, that's probably a big dose of PER, a healthy helping of SPD and BH, and a little bit of ATH, season with teammates' passing ratings, and go.  Then for the defender you'll get some perimeter defensive rating which likely looks like an adjustment to the expectation value or perhaps even an expectation value for an "average" offensive player against him, which likely is some combination of ATH, SPD, and DEF with maybe a little BLK in there (or maybe more, but I don't want to deal with that debate right now).  Then the two cumulative values are used to come up with some overall probability of the shot being made and the RNG outputs a value that determines whether the shot is made based on that probability.  But I don't think there's any stage of the calculation in which the ATH or SPD of the two players involved (or more, depending on offense, defense, etc.) are directly compared.

The upshot of that is that against two equivalently talented perimeter defenders the same offensive player will have the same expected 3 pt %.  It doesn't matter if one of them is very athletic and not so fast and the other is very fast and not so athletic.  Trying to read into things like that is giving the engine credit for more complexity than it actually has and potentially putting your players into shooting patterns that are less than ideal to try to exploit non-existent advantages.  I do think that faster guys tend to make better perimeter defenders, but that doesn't mean that a D3 guard with 65ATH, 55 SPD, and 60 DEF is an invitation to pump your guy up to +2 and go to town.

Each shot is an opposed check (OFF player vs DEF player) but the sum of the OFF player's attributes are used vs the sum of the DEF player's attributes as opposed to multiple checks vs each individual attribute?

Oh and I use +2 all the time, esp when  the sum of the offensive players PER and SPD is greater than 140 at DIII.  Once that sum gets over 120 I take it off of -2 and keep raising it.   If his 3 pt % is over 40% I make his setting higher, under 35% I lower it.  

Posted by Trentonjoe on 7/18/2012 5:54:00 AM (view original):
Each shot is an opposed check (OFF player vs DEF player) but the sum of the OFF player's attributes are used vs the sum of the DEF player's attributes as opposed to multiple checks vs each individual attribute?

Oh and I use +2 all the time, esp when  the sum of the offensive players PER and SPD is greater than 140 at DIII.  Once that sum gets over 120 I take it off of -2 and keep raising it.   If his 3 pt % is over 40% I make his setting higher, under 35% I lower it.  

interesting formula, and you base it off just those 2 stats? or do you vary it by defense played/matchups.

yeah more or less.... i dont think ath has a big part in the equation......not sure on bh.....i think bh deals more with getting the shot off

i leave the setting alone and adjust distro based on matchups

One issue that no one has mentioned is team balance. 

The best way to avoid being susceptible to +5 and -5 defenses is to balance your interior and exterior scoring.  If I have a team that looks like it will take too many inside shots, and thus be susceptible to a -5, I'll use the +2 on a few players.  The inverse is, of course, true for a team loaded with guards.