Does anyone know exactly how this works? I had always thought it was based on what percentage of the other team's points come from behind the arc, but I just played a game where my opponent scored 9 of 24 points from beyond the arc (37.5%) and I auto-adjusted from +2 to +4. I then played a game where my opponent scored 12/35 (34.2%) from beyond the arc and I auto-adjusted from +3 to +2. Those figures seem close enough to me that I wouldn't expect such a huge difference in adjustment. Does anyone know exactly how it works? I suppose I can go through box scores and get more data, but I thought I'd ask first in case someone has an answer. 

If you are talking about the auto adjustment by the SIM, I think it would be percentage of shots taken and not points scored.

So look at FG3A/FGA ... and see what levels the auto adjust thinks about that ratio.

Now, if you are asking how I think you should actually set it (and not how the computer sets it), then I would say that FG3A/FGA is one part of the equation (it is 2x as important as the next part) ... and that %shots by guards as compared to %shots by bigs should be the other part.

I use this chart for FGA/FG3A:

Defensive Positioning
Recommended	Basis	Max	Min
Positioning
+3	NA	100.00%	48.00%
+2	40.92%	48.00%	38.81%
+1	36.71%	38.81%	35.13%
0	33.56%	35.13%	31.67%
-1	29.78%	31.67%	25.48%
-2	21.19%	25.48%	18.86%
-3	16.54%	18.86%	8.27%
-4	NA	8.27%	0.00%

I take that number and I multiply it by 2 ... then I add the factor in what I call d40.

My d40 is based on the fact that perfectly balanced scoring would be 40% Guards, 20% SF, and 40% Bigs.  I take (%Total_Points_Guards-40%) as my d40 for guards ... it will be a positive number for guards scoring > 40% of total points and a negative number if the guards score < 40% of the points.  My d40 for bigs is the opposite ... it is (40% - %points_bigs), which would be negative if bigs score more than 40% of the points and positive if bigs score less than 40% of the points.  I then add the d40's together and divide by 20 (I take the %'s above and multiple them by 100 before doing the math on them ... you could also use .2 instead of 20 as the divisor).  Using this, if guards score 100% of the points, you would have this:

100-40=60    d40 for guards

40-0=40    d40 for bigs

(60+40)/20=+5   d40 factor.

That would mean if guards score 100% of points the d40 would be +5 and if Bigs score 100% of the points it would be -5 ... and if it was 40, 40 for Bigs and Guards, it would be 0.

So, for a total example ... lets take a team that has these numbers:

FG3A/FGA is 546/1416 or 38.6% ... that would be +1 from my chart.

Total Points for the team are:  30.3 Guards, 15.8 SF, 21.9 Bigs ... so 30.3/68.0 = 44.5% Guards and 21.9/68.0=32.3% Bigs ... that is 4.5 + 7.7 = 12.2 and 12.2/20 = +0.61

Combining them ... [(2 x factor1) + (1 x d40)]/3  = [(2 x 1) + (0.61)]/3 = 2.61/3 =  +0.87 ... therefore, I would likely play +1 against that team.

That doesn't make a ton of sense to me. I mean, I can see the argument for it, but I'm not sure to what degree I buy it. I want my team to adjust if and only if the other team is taking advantage of a mismatch. Not if they're consistently trying and failing to take advantage of a non-mismatch. If a team goes 1/11 from deep and 8/10 inside, would you move your defense towards the perimeter because they're taking so many threes? 

Although it does explain this particular situation. The 9/24 team took 11 threes in the first half, the 12/35 only took six. 

On your point about how you should set it. . . awesome stuff. I'll have to take a few minutes to sift through it, but it looks very effective. Although, depending on the state of my team, I might try to take rebounding or fouls into the equation. 

Posted by tarvolon on 4/19/2013 6:34:00 PM (view original):
That doesn't make a ton of sense to me. I mean, I can see the argument for it, but I'm not sure to what degree I buy it. I want my team to adjust if and only if the other team is taking advantage of a mismatch. Not if they're consistently trying and failing to take advantage of a non-mismatch. If a team goes 1/11 from deep and 8/10 inside, would you move your defense towards the perimeter because they're taking so many threes? 

Although it does explain this particular situation. The 9/24 team took 11 threes in the first half, the 12/35 only took six. 

I never have my team self adjust ... well, maybe if losing by 10 or more.

Also, I am not saying anyone else should use my method ... I am just putting it out there as what I do.

There are MANY other coaches out there who's jock strap I can't even see, let alone hold ... so I'm sure there is a better way to do this

Posted by hughesjr on 4/19/2013 6:44:00 PM (view original):

Posted by tarvolon on 4/19/2013 6:34:00 PM (view original):
That doesn't make a ton of sense to me. I mean, I can see the argument for it, but I'm not sure to what degree I buy it. I want my team to adjust if and only if the other team is taking advantage of a mismatch. Not if they're consistently trying and failing to take advantage of a non-mismatch. If a team goes 1/11 from deep and 8/10 inside, would you move your defense towards the perimeter because they're taking so many threes? 

Although it does explain this particular situation. The 9/24 team took 11 threes in the first half, the 12/35 only took six. 

I never have my team self adjust ... well, maybe if losing by 10 or more.

Also, I am not saying anyone else should use my method ... I am just putting it out there as what I do.

There are MANY other coaches out there who's jock strap I can't even see, let alone hold ... so I'm sure there is a better way to do this

I do but have frequently considered stopping. I do sometimes set it to only self-adjust if I'm doing worse than I expect to be doing. What I really wish is an option where you can set an auto-adjust to +/- 2 or to +/- 1. There are frequently cases where I'm not too confident in my gameplan and would like to adjust it a little bit either way if it's not working but am extremely confident that I don't want it adjusting by 2. This would seem like a pretty easy addition to me, but maybe I'm the only one who wants it. 

Posted by tarvolon on 4/19/2013 6:34:00 PM (view original):
That doesn't make a ton of sense to me. I mean, I can see the argument for it, but I'm not sure to what degree I buy it. I want my team to adjust if and only if the other team is taking advantage of a mismatch. Not if they're consistently trying and failing to take advantage of a non-mismatch. If a team goes 1/11 from deep and 8/10 inside, would you move your defense towards the perimeter because they're taking so many threes? 

Although it does explain this particular situation. The 9/24 team took 11 threes in the first half, the 12/35 only took six. 

If they are shooting a bunch of 3's ... and if you are playing to far back, they are more likely to make them (that is, shoot a higher percentage).  If you are playing more towards the perimeter, then they should make a lower percentage of 3's.

So, your adjustment impacts 2 things ... how many 3's they will take (a higher plus means they will take fewer 3's .. a more negative number, they will take more 3's) ... and how many they hit (they are more likely to shoot a percentage that is higher than their average for 3's the closer to the basket that you play ... they are more likely to shoot worse the farther out you play).

Of course, how well you play defense also impacts their percentages as well.

Posted by hughesjr on 4/19/2013 6:50:00 PM (view original):

Posted by tarvolon on 4/19/2013 6:34:00 PM (view original):
That doesn't make a ton of sense to me. I mean, I can see the argument for it, but I'm not sure to what degree I buy it. I want my team to adjust if and only if the other team is taking advantage of a mismatch. Not if they're consistently trying and failing to take advantage of a non-mismatch. If a team goes 1/11 from deep and 8/10 inside, would you move your defense towards the perimeter because they're taking so many threes? 

Although it does explain this particular situation. The 9/24 team took 11 threes in the first half, the 12/35 only took six. 

If they are shooting a bunch of 3's ... and if you are playing to far back, they are more likely to make them (that is, shoot a higher percentage).  If you are playing more towards the perimeter, then they should make a lower percentage of 3's.

So, your adjustment impacts 2 things ... how many 3's they will take (a higher plus means they will take fewer 3's .. a more negative number, they will take more 3's) ... and how many they hit (they are more likely to shoot a percentage that is higher than their average for 3's the closer to the basket that you play ... they are more likely to shoot worse the farther out you play).

Of course, how well you play defense also impacts their percentages as well.

Right. I got that. But if they're taking a whole ton of threes because they can't solve my interior defense, but they're really bad at three-point shooting, I'm kinda okay with that and don't want to adjust out towards the perimeter. Now if they're making a decent percentage, I might want to adjust. But if they're not, I'm fine where I am (even understanding that taking a missing a bunch of threes may be a statistical anomaly that could be corrected in the second half. If that's the case, I'll roll with the strategy I started with). 

Of course, on the other side of the coin, if a team with a couple 90+ PER guards starts 1/10 from deep, I sure as hell don't want to bring my defense in tight either. Maybe a combination of the two is the way to go. 

Posted by hughesjr on 4/19/2013 6:25:00 PM (view original):
If you are talking about the auto adjustment by the SIM, I think it would be percentage of shots taken and not points scored.

So look at FG3A/FGA ... and see what levels the auto adjust thinks about that ratio.

Now, if you are asking how I think you should actually set it (and not how the computer sets it), then I would say that FG3A/FGA is one part of the equation (it is 2x as important as the next part) ... and that %shots by guards as compared to %shots by bigs should be the other part.

I use this chart for FGA/FG3A:

Defensive Positioning
Recommended	Basis	Max	Min
Positioning
+3	NA	100.00%	48.00%
+2	40.92%	48.00%	38.81%
+1	36.71%	38.81%	35.13%
0	33.56%	35.13%	31.67%
-1	29.78%	31.67%	25.48%
-2	21.19%	25.48%	18.86%
-3	16.54%	18.86%	8.27%
-4	NA	8.27%	0.00%

I take that number and I multiply it by 2 ... then I add the factor in what I call d40.

My d40 is based on the fact that perfectly balanced scoring would be 40% Guards, 20% SF, and 40% Bigs.  I take (%Total_Points_Guards-40%) as my d40 for guards ... it will be a positive number for guards scoring > 40% of total points and a negative number if the guards score < 40% of the points.  My d40 for bigs is the opposite ... it is (40% - %points_bigs), which would be negative if bigs score more than 40% of the points and positive if bigs score less than 40% of the points.  I then add the d40's together and divide by 20 (I take the %'s above and multiple them by 100 before doing the math on them ... you could also use .2 instead of 20 as the divisor).  Using this, if guards score 100% of the points, you would have this:

100-40=60    d40 for guards

40-0=40    d40 for bigs

(60+40)/20=+5   d40 factor.

That would mean if guards score 100% of points the d40 would be +5 and if Bigs score 100% of the points it would be -5 ... and if it was 40, 40 for Bigs and Guards, it would be 0.

So, for a total example ... lets take a team that has these numbers:

FG3A/FGA is 546/1416 or 38.6% ... that would be +1 from my chart.

Total Points for the team are:  30.3 Guards, 15.8 SF, 21.9 Bigs ... so 30.3/68.0 = 44.5% Guards and 21.9/68.0=32.3% Bigs ... that is 4.5 + 7.7 = 12.2 and 12.2/20 = +0.61

Combining them ... [(2 x factor1) + (1 x d40)]/3  = [(2 x 1) + (0.61)]/3 = 2.61/3 =  +0.87 ... therefore, I would likely play +1 against that team.

Seems awfully complicated for a +/- setting.

emy, that is what spreadsheets are for

(you should see what I use to decide distribution settings :-D)

Posted by hughesjr on 4/20/2013 4:15:00 PM (view original):
emy, that is what spreadsheets are for

(you should see what I use to decide distribution settings :-D)

Didn't mean anything personal by it, just a little more complicated than the "system" I use, ha ha!

Posted by hughesjr on 4/20/2013 4:15:00 PM (view original):
emy, that is what spreadsheets are for

(you should see what I use to decide distribution settings :-D)

I would be very interested to see what you use for distribution settings - I like that system for plus-minus.