Posted by milwood on 5/15/2015 6:04:00 PM (view original):
Thanks TJ.  I will still try to get guys that can both block shots and play defense.  

It seems a bit confusing, though.  So when I'm playing a 2-3 my forwards are averaged together. However, my SF never has any kind of shot blocking at all and my power forwards still tend to have decent and way more blocked shots than my SF.  I understand it is widow dressing but in that scenario I would expect my PF to be closer statistically to my SF than to my C in blocked shots, but that is just never the case.

From my last year's team.  The guy that started every regular season game at SF had a 5 rating for SB and he averaged 0.1 blk/game. (again, all but 4-5 games playing 2-3)  The guy playing PF had a 75 rating for SB and averaged 1.7 blks/game.  If blk is averaged they would both have a 40 rating for BLK and you would expect their statistics to be at least in the same ball park.

I guess what CS is saying though is that the forwards together have an 80/2 rating for SB, and then after the play happens it assigns the blk statistic which my PF would get a 15/1 ratio of those blocks.  But then why does my C have just as many blocks (roughly) as the PF.

I'm glad I know I'm not smart enough to figure these things out otherwise I would surely screw everything up.  I like to just play by the eye test.....yup, that guy looks like he'd make a good PF.....

I think what happens is that the SIM decides if a block happens and then assigns it to the player....who gets it is probably a formula based on position and SB rating.

interesting thread. a few things

- first, i agree the notion of being able to hide a bad defender is misleading. the reason for this is that the bad defender is averaged in, so against your average team where the coach isn't trying to exploit m2m matchups, you really gain nothing here. the only gain is that a subpar defender cannot be specifically targeted by an intelligent opposing coach, like they would be in man.

- second, and this one is going to cause a big shift in thinking for many folks - we should dispense the current notion of how players are averaged in the zone. we say, in a 2-3, the pg/sg an sf/pf are averaged. this is actually not correct.the pg/sg use the same equation, and the sf/pf use the same equation, while the c uses his own equation. this is the only relation between those pairs of players. in terms of the averaging - all 5 zone players are averaged (not necessarily with equal weight) into the figuring of the defense on every shot.  i suspect, but it is not known for sure, that the weighting of how much each player factors in, depends on the distance of the shot.

what seble stated is that all 5 players were averaged, and that their equations (the ones where they are the same for pg/sg sf/pf and then c alone, in a 2-3) depend on the distance from the basket. this could have been put in place by old admin for 2 purposes, the way i see it. first, it would be to weigh ratings differently based on where the shot is taken - a center's speed would matter more defending a 3 than his blk, where as the opposite is very much the case for most cases. second, old admin could have set it up this way as a means of weighting the players differently when they are averaged together in defense of a shot. for example, on a 22 foot shot, the c could weigh in much less than the rest of the team. i suspect, both logically, and from (recent and somewhat limited) observation, that the second aspect there was tarek's primary goal, but i definitely could be wrong on that. basically, i read it as, even though all 5 players are averaged together on every shot, they aren't all averaged together at equal weights.

so, what does this mean for comparing the pairs of players we all reference? it means you can position players within the sets (pg/sg sf/pf in 2-3, pg/sg/sf and pf/c in the 3-2) arbitrarily, from a defensive standpoint. in the averaging of the 5 players on each shot, you can think of those 2 as being first averaged together, and then getting averaged with the rest of the team, because its logically equivalent to what actually happens - which is that those players are simply using the same equation, and then averaged together with the team. because their equations are identical, including the part that is based on difference, you know those players will factor in the exact same way to the team average which occurs on every shot. so if you want to think of them as being averaged, fine, just make sure you are realizing they are also being averaged (probably with unequal weights) with the rest of the team, too.

- third, shot blocking is absolutely more important in zone than man, and sb is more important in man than press. seble never said anything but from experience i am virtually positive on these counts. shot blocking in the center in zone is extremely important, and i consider it a top tier stat for centers in the 2-3, along with ath and reb (note the exclusion of def there). its possible shot blocking is the single most important rating for a center in zone, and its probably a top line stat for pfs and cs in the 3-2 as well.

edit: just want to confirm that if people are only thinking of the impact in BLK, in terms of the stat sheet blocks, this is a big misjudgement (albeit one i made myself, for years - i was just lucky enough to play press at a time when BLK was minimally important in the press - so i didn't really get burned). BLK also factors into fg defense, outside of the blocks themselves (obviously a block will cause a missed fg, so there is the direct impact. but good shot blockers also alter shots, and decrease the fg% of shooters even without registering a block. it has been confirmed numerous times by the game admins that altered shots are a significant part of the sim engine logic)

OK, how about assignment of fouls in a zone.  If the engine determines that my team has committed a foul (based on some averaging of ratings across all the players, which determines the outcome of the play), is a crappy defender guard more likely to commit the foul than an awesome defender guard (in a 2-3, let's say, where the two guys are playing PG and SG in some order).  This question is at the core of "can I hide a bad defender?"  

-- (1) If the crappy defender is more likely to be assigned the foul, then he is in no sense really being hidden, except in the pure matchup exploitation sense.  He would be increasing the team foul rate, increasing the opponent FG% rate, and drawing more fouls himself.  

-- (2) But if he is not more likely to be assigned the foul -- if it is randomly assigned to one of the two guards, say -- then he is increasing fouls and opp. FG% -- but he himself is no more likely to get into foul trouble than the other guy.  

If you're sufficiently better than your opponent, the number of fouls in a zone is small enough that it might not matter.  But if you're not, the game is uptempo, the opponent is a heavy foul drawer, etc., then the amazing scorer you had hoped to "hide" in a zone has an increased chance of getting into foul trouble if (1), but not if (2).  Whether (1) or (2) is true will determine how valuable that amazing scorer is, because it will determine his offensive production (by virtue of being in foul trouble and thus potentially out of the game or not) as well as team defensive ability.  

On my Redlands team, the bad defenders seem to draw about 1/3 more fouls per minute than my better defenders......my slower (55 compared to 75) SF draw more fouls than I thought he would as well.....I did the math quickly and in my head, it may not be right.....

good question, I'm really not sure about the fouls.

"what seble stated is that all 5 players were averaged, and that their equations (the ones where [pg/sg], [sf/pf] and then [c] alone, in a 2-3) depend on the distance from the basket."

Has anyone thought much about what determins the 'distance from the basket' of a shot?

What governs whether a hypothethical [35LP, 70PER, -1] PG's fg attempt is:
          -an open 3?
          -a jumper from the left block?
          -an easy layup?
          -a floater from the left side?
          -pull-up jumper from just outside the lane?
          -jumper in the paint?
          ***(I realize these descriptors are probably broken down into a few basic 'distances' - maybe close/mid/long-range?)

Is it reasonable to think that 'distance from the basket' is determined something like: (a RNG dice-roll using a players LP&PER rating) X (player 3pt +/- setting) X (opp. defensive +/- position)?

Here is the answer, which is not an answer, but It gives some input into what to expect.

Antoine,

In both types of zone defenses, the impact of the defense is an average of the 5 players on the floor. Each player's ability (as part of that average) is calculated differently based on what position they're playing. So the weighting of ratings will differ based on the player's position and the type of defense that the team is playing.

As to what ratings are most important, we don't give away all those details, but it's pretty much what you'd expect if you watch much real-life basketball. For example, a player positioned mostly on the perimeter in a certain defense will need to have some speed and athleticism. On the other side, a player who is positioned near the basket would benefit from shot blocking. Of course the Defense rating is always the most important when it comes to defensive ability.

Thanks for posting that response. I think the big takeaway again is that BLK is likely to be more valuable for bigs in the zone than it is for other positions and other sets. And it is still not as important as defense (nor athleticism, I'd suggest) as an attribute, even for bigs in a zone.

Yeah, I see zone like this now,

3 - 2 PG-SG-SF (#1DEF, #2SPD, #3ATH), PF-C (#1DEF, #2ATH, #3 BLK)

2-3 PG-SG (#1DEF, #2SPD, #3ATH), SF-PF (#1DEF, #2ATH, #???), C (#1DEF, #2BLK, #3ATH)

What do you guys think?

I think that is a good starting point, but there is another side of the ball to worry about.  Some bigs are super slow and some have virtually no bh and passing.  That type guy will be a better fit at C.  Then there are bigs that have some speed and good to excellent bh and pa which will help the offense flow better.  I would prefer to use that guy at pf even if his defensive skills more align with the C position (in a 2-3, I'm just not good at running a 3-2)

Shot blocking is important but I wouldn't give up a bunch of defense to get some extra blocking.  There are often guys with 80 block and 40ish defense.  I wouldnt even give that guy a second look.  But I would take the 80 defense 40 block without much question.  Well, I would if I couldnt find a guy that was 70s defense and 70s + block (DII).  I guess thats another reason why I dont really get this thread.  There are a lot of those type guys out there so I dont know why its so important to figure out which one is more important.  Get both!

I dont want to be misunderstood, however.  I think this is a very good and informative thread

Posted by beefburglar on 5/18/2015 7:16:00 AM (view original):
"what seble stated is that all 5 players were averaged, and that their equations (the ones where [pg/sg], [sf/pf] and then [c] alone, in a 2-3) depend on the distance from the basket."

Has anyone thought much about what determins the 'distance from the basket' of a shot?

What governs whether a hypothethical [35LP, 70PER, -1] PG's fg attempt is:
          -an open 3?
          -a jumper from the left block?
          -an easy layup?
          -a floater from the left side?
          -pull-up jumper from just outside the lane?
          -jumper in the paint?
          ***(I realize these descriptors are probably broken down into a few basic 'distances' - maybe close/mid/long-range?)

Is it reasonable to think that 'distance from the basket' is determined something like: (a RNG dice-roll using a players LP&PER rating) X (player 3pt +/- setting) X (opp. defensive +/- position)?

yeah, i mean i wouldnt express those 3 factors as multiplication necessarily, but its most likely mainly a RNG dice roll, weighted by the lp/per rating, 3 point setting, the defense itself, and opponent defensive positioning. double teams will affect overall 3pta by the team but i doubt it will for the player. also the defensive ratings of the player (or team) in question will affect 3pta by the team, but i suspect it will not for the player, either.

end game settings, rebounds for put backs/tip ins, etc, all work differently, obviously.

in the game itself, distance is just an integer (i believe, it could be floating point, but i doubt it). there is no concept of left or right side, or middle. the game may categorize distances into types of shots (3s, jumpers, layups/dunks), but i kind of get the impression it doesn't work that way. i think the equation just depends on the distance from the basket and is smart enough to take it into account. i definitely could be wrong on that though. 

Posted by zorzii on 5/18/2015 7:46:00 PM (view original):
Yeah, I see zone like this now,

3 - 2 PG-SG-SF (#1DEF, #2SPD, #3ATH), PF-C (#1DEF, #2ATH, #3 BLK)

2-3 PG-SG (#1DEF, #2SPD, #3ATH), SF-PF (#1DEF, #2ATH, #???), C (#1DEF, #2BLK, #3ATH)

What do you guys think?

you mean just for defensive ratings, i presume?

in that case, for the 3-2, i'd probably put blk over ath. for the 2-3, you probably have it right, but i honestly wonder if blk might be 1 for the center. i realize that seble just clearly stated in the response posted above that defense is always #1, but im not confident i would take that always to mean always. it wasn't always before and seble did change the defensive weightings considerably, and i suspect defense is #1 in affecting fg% always, but is it overall? that is a much tougher question and im not sure seble is really looking at all the ways a rating can impact defense and aggregating them when he makes that statement (assuming he made that statement - probably a safe bet). blocks are big for blocks themselves but also in fg%. the combination of the two is quite potent. at this point i personally value blk over def for my c for 2-3 teams, but that certainly could be a mistake. especially if you have depth at the 5, i value blk over defense by a fairly significant amount.

I play press/zone, and I try to configure my teams to be able to move seamlessly between 2-3 and 3-2. As for overall priorities, on a 5-point scale:

Guards 
5-Def, Spd
4-Ath, BH
3-Per, Pass

Forwards
5-Def, Ath
4-Reb, LP
3-Blk, Spd, Pass

Center
5-Def, Ath
4-Reb, Blk
3-LP, Pass

Posted by pkoopman on 5/18/2015 9:43:00 PM (view original):

I play press/zone, and I try to configure my teams to be able to move seamlessly between 2-3 and 3-2. As for overall priorities, on a 5-point scale:

Guards 
5-Def, Spd
4-Ath, BH
3-Per, Pass

Forwards
5-Def, Ath
4-Reb, LP
3-Blk, Spd, Pass

Center
5-Def, Ath
4-Reb, Blk
3-LP, Pass

quick comment, this is a great start, but if you adhere to that strictly, its not nearly flexible enough. its all about roles, and abilities, in this game. for example, per is easily the #1 rating for a leading perimeter based scorer at the 2. for most point guards, passing is easily the #1 rating. so, weighting those as a 3 for those players, compared to the other ratings, would be a big mistake. as a starting point, still kind of suspect, but not nearly as dramatically so. in my examples, they are not even in the realm of reasonable, while on average, they are just a bit low.

forwards is another example, for a per scoring sf who is a lead scorer type, per is generally the #1 rating. you don't even have it listed. for a non scoring sf, those ratings look pretty good-ish. also the lp/per spread is way different for a 3/4 so its kind of tough to lump them. similarly for your c, those ratings are pretty solid for a non scoring c, but for a scoring c, are off by a good bit.

not trying to be a dick, i don't know how you use those ratings! some people use guidelines to quickly find the best players which they evaluate by eyes at that point. but others strictly follow their rating system in spread sheets and such. in the latter case, my comments very much apply. in the former, just generally a word of caution. as coaches move to the level of trying to make final 4s and such, its critical to evaluate players by roles, and the limited roles listed there are not nearly rich enough for higher level team planning!

HA! OK, for clarification, those aren't weights (and they're not strict). They're overall priorities. As I had said earlier, one of the biggest reasons I play zone is so that I can take advantage of players with elite skills - like passing pgs, rebounding bigs, defensive specialists, etc.