"Circumstantially I think DT does a good job of denying access to the ball rather than creating turnovers, but I don't have any hard data to support that theory."

Me too. Best guess is that ALWAYS reduces a player shots by like 3-4 per 40 minutes but only increase TO's by like 1 per 40.

I use the ILS option as a crutch in case I make a mistake in gameplanning. For example if I sell out vs. the three and target a couple guards, but the other team also has a beast center who usually gets 10 a game but could easily go off for more against you, that's when I use the ILS. It's not often as I don't like leaving holes in a defense but I am not the best gameplanner so sometimes I need to hedge a bit.

Posted by Trentonjoe on 10/3/2019 1:34:00 PM (view original):
I almost never disagree with Gillispie but I do now, twice.

1. "the time when i think a double team makes the most sense is when there are 1-2 strong 3 point scorers doing all the 3pt scoring, and you can double them and be safer in running a significant minus." I have done this MANY times and it just doesn't work. The penalty for double teamnig is less thean the +/- adjustment. I have tried double teaming a teams one three point shooter and playing a heavy minus and the 3 point shooter almost alway crushes me.

2. 'here's why i think ILS is about the worst thing you can (use)'. While Gil's logic is spot on, it does have situational uses. My favorite use is when you have one substitute who shoots alot and one or two starters who shoot alot. in this case, you probably want to double the bench scorer when it's just him in but not if one of the other two scorers are in (and they are better than him). i pretty frequently ALWAYS a teams scorer and ILS the third guy to make sure if third option overall is the first option on the court, then he has the penalty.

good points.

on #1, i think there is some nuance there i may have left out. to be clear, my starting point is that DTs are almost useless and irrelevant, so i definitely agree +/- is a bigger factor. i probably wasn't that precise in my ramblings, let me try to tweak this one a bit. so, im not really trying to say go ahead and DT this 3pt scoring dude and that gives you the green light to drop from 0 to -3 or something like that. im more saying, if the data points to a -3 but you are scared this one guy will torch you, it makes it a bit easier to do that -3. i do think DT has some impact both on shooting % and on volume of shooting, and the combo is helpful particularly when some guy is liable to destroy you both in terms of quality of shooting and in volume. not a huge help, but it is definitely a case where i find DT to be clearly justified. any time some dude is scoring an insane amount it is also justified. so its more of a mitigation strategy for the case where you are trying to run - but 1-2 guys might wreck you, than it is a justification to go - when its not warranted without the DT.

on #2, very interesting! i didn't think about DT'ing a non-starter only when hes playing with other backups. in that case, it seems sort of hard to control what is going on, and unless its press you don't really see the hockey style rotations which are sort of conducive to what you are talking about. but still, it definitely is a middle ground between the all or nothing option you are otherwise afforded. pretty interesting idea you have there, i like it! especially because those 1-key-backup-scorer scenarios often feature a really good backup with some really garbage scorers so the negative of the DT, letting the other players shoot better, is probably not much of a concern.

..i just don't look at whether the guy i double-teamed scored less, did he have more to's? was he fouled more? etc..
(ex.) >

14:50

Wooster

James Keane is double-teamed and called for traveling

in this game Wooster's 2 main scorers were double-teamed..both avg'd over 20 pts/game with 2 to's during the season..in this game both scored 13 pts., with 9 to's between them..Wooster lost by 1 pt..i would say the double teams were successful in this game..

Posted by gillispie1 on 10/3/2019 12:03:00 PM (view original):
i agree with a bunch of this - first off, that doubling for turnovers is probably not something that works as one would expect, and likely is not something that exists at all in the sim engine. the whole DT thing is very half baked, it may be as simple as a small fg% penalty and fga penalty (fewer possessions) for the main scorer and a smaller fg% boost for the rest of the team. that said, ive never deeply studied it and don't know anyone who has. you may want to do a targeted study of double teams in the context of your situation, and see what you can uncover.

i also agree its really about usage % not shooting %, i just use that as shorthand - but definitely usage% is the better indicator.

one thing i disagree with fairly strongly (as strongly as one can in the context of a half baked checkbox that barely does anything), is the use of 'if leading scorer'. ill state this strongly and let other folks throw darts at it. ILS would potentially make sense if and only if you are unsure of the strategy of your opponent. however, as wildly changing distros are very rare in this game, the logical conclusion of that statement is that ILS very rarely makes sense, which is my position.

here's why i think ILS is about the worst thing you can. suppose you are targeting the lead scorer on the floor in terms of normal scoring, which is the typical usage. now, your double team is working, the guy is missing shots, and hes not the leader. so what do you do - stop doing the exact thing that is working for you! that makes a lot of sense, right? (not). even more, if you think about feedback, you are getting double ******. suppose the guy is getting screwed by RNG, so he's not leading scorer. now, for the rest of the game, hes going to be getting a fg% boost via the feedback - so its PRECISELY the time you really need to be focusing on him. instead, you abandon the strategy and let him have at it. conversely, suppose RNG is helping him a lot. now hes the leading scorer and feedback is going to wreck him going forward. at this point, its a lot less important to double team him, because you want to let him shoot and get the penalty of all that feedback. instead, you are going to push on with the double team.

so in summary, i think ILS works consistently against your interests, both because when it works, you abandon it, and because of how feedback is part of this game. i'd definitely be interested to hear a counter argument though.

1. I’m not convinced distribution needs to be “wildly” changed in order for DTILS to make sense. It just has to be unpredictable enough for me to see some utility in hedging my game plan.
2. I’m not convinced distribution changes are as rare and negligible as you state. Among the humans I play in my active conferences, I’d say it’s pretty close to 50% guys who will absolutely change up their distribution if they think they can predict what you’re going to do defensively.
3. As I stated, I will consider DT-always if I am up against a team with only one guy I’m worried about. The fact is that most human teams have more than one.
4. TJ’s point 2 is correct.
5. Unlike TJ, I have had good success with the scenario I described above, which is probably the type of situation that will most likely cause me to use DTILS - where I’m up against a team that rarely (under 20%) shoots the 3, but does have guys who could hurt me if I play heavy minus, and they kick up the outside distribution. Maybe it’s my typical 3-2 base, but I have not experienced the perimeter torching TJ describes when I play those teams 3-2 at -4, and DTILS their 3pt threat(s).
6. Like all gameplan choices, it’s really about having a coherent plan to exploit your opponent when possible, and mitigate your own vulnerabilities where applicable. It doesn’t have to represent an all-in type approach to be valid, conservative gameplans can win.

i am not sure there is much daylight between TJ, you, and myself on the negative with DT of top 3 point scorer - i think TJ is saying, more or less, if you should be running a 0 based on the overall ratio of 3pt shooting and paint scoring or per vs lp ratings, whatever (there are several mostly legit ways to evaluate that), then you can't drop to a -3 by DTing the 1-2 key scorers. i was trying to say, if you are going to run a -3 but the concern is their 1 3 point scorer, you may as well double and it can enhance your strategy. maybe you can go -1 more than you were or something, a -2 without a DT could go to -3 with a DT, something minor - which i don't think is what i conveyed, that TJ was objecting to.

i could be wrong but i think you are roughly saying the same thing as me. 3-2 -4 is not a super heavy negative for teams that are shooting 20% or fewer 3s, it actually is a very reasonable way to go. so i think you are complementing a reasonable strategy with a DT to form a more comprehensive whole, which is also what i was trying to say - and i think TJ is more saying you couldn't do that 3-2 -4 against a 30% 3s team just because you DT, which is probably a response to my not-so-greatly worded first post on the subject. so not sure there is any conflict in any of that? what do you think?

i am curious what you are talking about in #2. minor distro changes are common, major ones are not, is how i see it - with major meaning you are doing stuff like, say you are on a 10 max scheme, guys going from 10 to 6-7 and 4-5 to 8, stuff like that. not +1 here, -1 there, maybe 2 on occasion. i am not generally aware of any coach in HD history who does major changes to distro without shooting themselves in the foot, myself included. i consider myself on the extreme end of game planners, i'll walk into a NT championship and not hesitate to run a rotation i've never even thought about using - so i'm not trying to knock fairly extreme game planning in general. i do think major distro changes in particular are unlikely to be effective except in very rare cases, because you more or less have to play your team first and the opponent second, and your good scorers are usually so much better than your poor ones.

it may be a frame of reference issue. i only generally concern myself with what play against a top 10 opponent looks like, because its where you get your substantial NT championship share increases. for example, pushing a team with 2% odds to 1% (your rd 1 opponents, presumably) is a lot less important than pushing your 60% chance of beating the #2 team to 70% (in the f4 or title?), and the former is more difficult. anyway, you do see a LOT more crazy stuff against lesser opponents, but my view is that rarely works in their favor and can more or less be safely ignored (plus i dont care at all about the regular season, where i play more of those teams). also, i generally don't care about game planning unless my team is a significant title contender, so im not only assuming the opponent is a top team, but roughly that my team is the favorite. as the favorite, my best scorers are always way better than my mid scorers, so i can't do a lot distro wise to shift things around. if you have a very evenly balanced team, which i don't do because its less efficient, i could see some wild distro changes - but that's basically out of scope for me. this really may come down to scope, i sometimes need to keep in mind everyone else isn't limiting themselves to the same scenarios i am.

that said - more or less, if you care about optimizing stuff in the championship context, all that really matters is top 10 play. i generally encourage folks who can comfortably build A+ programs but struggle with consistent championships, to basically go to the hardest conference they can find, schedule a ball busting non conference, and to really limit their scope to that top 10 play. this game has so many variables and situations, to really optimize at a fine granularity, i think a reduction in scope, eliminating as many variables as you can, is essential. what is optimal vs the overall pool of teams is very different from what is optimal against top competition, so IMO it is very helpful to really focus on the top competition and treat the rest as noise.

I think I agree with what billyg is saying about what I am saying. It’s a lot of words though.....

“you couldn't do that 3-2 -4 against a 30% 3s team just because you DT, which is probably a response to my not-so-greatly worded first post on the subject. so not sure there is any conflict in any of that? what do you think?”

Right. The idea here is that I default to 3-2 at -2 against a team around 33% FGA from 3. I dip farther negative on a scale based on how rarely they shoot 3s. 20% is borderline -3/-4, 10% is borderline -4/-5. I don’t hesitate to dip that far in - but some of those teams have one or two guys who can shoot 3s. And some of those coaches know me. So I *may* hedge that move inside with a DTILS.

FWIW, I move distribution more than a point here or there. In the 6 Tark tournament games, D*kes, Reeves, and Breeden were between 7 and 10, but all of them had at least one game on both ends. As you hint, when you have a number of scoring options, and the difference between them is not significant, sliding distribution is a good weapon. That’s certainly the kind of thing I look for in opponents in the course of my (very limited) game planning routine.

sounds good shoe, and if you had a bit more balanced team than me, we seem to be in a similar range - i'd move that 9 distro lead up to 10 or down to 8 with my less balanced team - so its not like we are that far apart (7-10 vs 8-10). i think d2 is a little more forgiving on having balanced teams, i had more balanced teams in d2 and d3 than in d1, which may be counter intuitive due to the greater disparity of players. but the d2/d3 reality is that defenses are much weaker giving you more room to have high end offensive players than you do in d1 (i think top scorers in d2 vs top teams average defense is basically a bigger gap than d1).

in d1, the top pretty many teams are so good defensively that you have to confine yourself to the truly elite scorers to be truly elite, and big men are so disadvantaged that there like, aren't physically enough slots at the 1-3 to have much balance. i've been there in d2 though where i had 5 guys in a tight band in the lower double digits range. i had one team once that was so balanced in d2 it sort of pushed my sanity, i ended up going with all 0s for the final 4 and championship (which we won by 40 and 26, which sounds awesome but actually was relatively bad for my mental health at the time)... never come close to that in d1.

appreciate the opportunity to procrastinate! its performance review season, my least favorite thing of the year and/or millennium. holding out hope i get run over by a bus before my procrastination runs its course.