Box Score

Yes, I'm ****** at the bad loss, so maybe I'm reaching here - But can someone please tell me why the AI auto-adjusted me from a (+1) to a (-1) in the second half?

In the first half, the opponent took 12 3-point attempts (out of 36) against my +1 defense. At the half, the defense was auto-adjusted to -1. In the second half, the opponent took 12 3-point attempts (out of 25).

I was under the impression that the halftime adjustments were based on how things played out in the first half, but I find it difficult to believe that 1/3 of the shots coming from the 3-point line justified the maximum possible shift inside (two spots) to start the second half at -1. Thoughts?

The opponent shot 44% from the field and only made 1 out of 12 three pointers.   I find it very justifiable that the engine would shift to a -1 based on the ineptitude from beyond the arc.

Despite giving up more treys the defense held the opponent to 40% shooting in the 2nd half.   BTW, I dont think the defensive performance had anything to do with this loss.  You lost because you gave up 10 net offensive posessions due to rebounds and turnovers.

Posted by mullycj on 2/3/2012 3:35:00 PM (view original):
The opponent shot 44% from the field and only made 1 out of 12 three pointers.   I find it very justifiable that the engine would shift to a -1 based on the ineptitude from beyond the arc.

Despite giving up more treys the defense held the opponent to 40% shooting in the 2nd half.   BTW, I dont think the defensive performance had anything to do with this loss.  You lost because you gave up 10 net offensive posessions due to rebounds and turnovers.

I suppose it is possible that the shift is a result of opponent's efficiency from 3-point land, but that doesn't seem consistent with my past experience.

In effect, that can be seen as working against the game-plan, since presumably a game-plan designed to limit 3s would reduce their efficiency. So, because the team shot a lot of three and failed to do so effectively (the defense did what it was trying to do), the defense should be changed to allow them a greater opportunity to perform better in the second half?

Looking through some other box scores, you may be right though. Like this.

That seems a horrible way to have it work.

Coach before the game: OK guys, we're gonna really press the perimeter because they shoot a lot of threes and we don't want them to beat us that way.

Coach at halftime: OK guys, great job out there. They're tossing up bricks from beyond the arc, but they keep gunning. Now stop what you were doing and sag back inside.

Really?

You should have set your halftime adjustment to only when you are losing by 5 or more. 

I think that's exactly how it works unfortunately.  Agree that it's not really worth much, but I'd rather the sim try something if my gameplan is getting beat badly.

Yes, tianyi, I could, but under my (apparently incorrect) understanding of how it worked, it didn't really make sense to do so.

And killbatman, I agree I'd like the sim to try something, but here is what I *thought* it worked (and is the *something* I think it would make sense to try):

Coach before the game: OK guys, we're gonna really press the perimeter because they shoot a lot of threes and we don't want them to beat us that way.

Coach at halftime: Well damn, they keep feeding that post players and they're killing us down low. Who could have predicted they would only go 2/2 from 3-point land? They normally shoot 25 3-pointers per game! OK well guess we better sag in and defend the shots they are actually taking.

Not that much different, I guess, because it is the same idea only based on how the other team is trying to score, rather than making any judgments about efficiency. If they are gunning from 3-point land, I want to guard that even more closely, not sag in and give them the chance to do better than 1-for-12 in the second half.

By and large (I don't know, what maybe 75% of the time), half time adjustments get it right, it is a computer program afterall, plus I have made more than my share of bad half time choices real life, it isn't like coaches get it perfect.

I wonder what the response would be to a ticket, asking for the adjustment to be weighed more heavily toward types of shots tried, vs shooting %'s?

Posted by oldresorter on 2/6/2012 8:11:00 AM (view original):
By and large (I don't know, what maybe 75% of the time), half time adjustments get it right, it is a computer program afterall, plus I have made more than my share of bad half time choices real life, it isn't like coaches get it perfect.

I wonder what the response would be to a ticket, asking for the adjustment to be weighed more heavily toward types of shots tried, vs shooting %'s?

i wonder too - i was with you zbrent, i thought it was based on attempts. there is obviously so much luck involved in the making or missing of 10 or so 3 point shots in a half, it seems ridiculous to base the decision on that. maybe its some of each?

Why does this part of the sim have to be a mystery?  Coaches should absolutely understand this part of the equation in order to have some level of control during a game.  In a perfect world, the first half would sim, then there would be a time window for a coach to make any adjustments before the second half is played. 

it doesn't. you could probably spend a half hour and look at a hundred games and know for sure what the answer was (only 3pta, only 3pt%, or a mix - i suspect its a mix). what that mix is, of course, could take further refinement.

the real problem here is, if CS told you the answer, would you believe it? i guess i am much more of a skeptic than most of the coaches around here, but i sure as heck would not take it on faith. id still have to go out and do the study to know, to me CS can never be more than a verification. but then again, neither can the belief of a veteran coach, even if its OR or lostmyth or any other of the coaching elite. even when CS and the vets agree, its happened a number of times on major issues, they can still ALL be wrong, and someone who everyone writes off like colonels, furry, and tmac (all BEFORE they turned things around - which they all did - no disrespect guys in the slightest) could be right. that exact scenario happened with pulldowns... it was tmacfan and furry_nips vs all the games greats AND cs. and guess who was right? yup. tmac and furry.

so, i guess all im saying is, no it doesn't have to be a mystery. but someone telling you the answer doesn't make it any less of a mystery. got to go check it yourself.