I've played the last couple games where opponents have doubled two of my guards who have been on the floor more than 20 minutes together. I've seen a lot of hoops and I'm not sure this would ever be a very effective strategy as it would open the floor and should be easily beaten by ball rotation. Both teams were pretty successful against me.

Is this a flaw in game logic or a flaw in my logic?

i feel like double teams are one of the things i understand least in this game, one of the only settings where i am like man, i have to be leaving money on the table here. i would just defer to others, but i find most of the opinions out there on double teams to be even less palatable than my own.

my overall take is that double teams aren't that powerful in this game, neither in the pros nor the cons. i definitely agree the 2x double team has too little impact on the remaining 3 players, and it doesn't seem to me like double teams are robustly linked to things like bh/pass, the number of scorers on the court, or any of that. it feels to me like merely, there is a small hit to efficiency and volume of the targeted player(s), and a small benefit to everyone else, that is done in a superficial way. that might just be me hedging against not really knowing what is going on though.

i generally will double team a high enough volume scorer (a clear team leader at 20ppg or higher). sometimes i do it automatically, but i guess that's really the regular season when i don't care. in the post season i kinda waffle a lot - if the guy isn't that good, or is already well defended, i do sometimes decline even on seemingly perfect opportunities. i tend to use double teams more to cover weaknesses than to double down on strengths, and i have some decently plausible rationale for doing so, but i'm not sure that's the right move. that is the main thrust of what i do though - 1 really high scorer most of the time (i avoid doubling multiple players in most cases, but not always), and to cover weaknesses. particularly if there is 1-2 good 3pt scorers, i will be inclined to double them to cover a weakness or allow me to cheat an extra 1-2 on the negative side. it often seems to do nothing, but it seems to help some. or if i have one really crappy defender, i will sometimes double team there against a good scorer, even not an excessive one, to try to cover.

importantly, i only use the always setting. the if leading scorer setting just seems so awful. supposedly, overly double teaming eventually works against you, according to stuff written about it somewhere. who knows if that's true. so maybe ILS could help with that - but if so, we'd see the double double teaming causing clear damage, one would think? anyway if you kinda think down the main line of how this stuff probably works, it just seems like a disaster. that said, if double teams aren't very meaningful, a double team setting that is a disaster may not hurt you much. in short, if a guy starts with a hot hand, now he's leading scorer, feedback is going to come smite him anyway. the worst thing you can do is reduce his volume, reduce the opportunities for a reversion to mean. so why would you double team at precisely that moment? just doesn't make sense to me.

there's a bunch of good and really good coaches out there who advocate for the ILS... but i would just say in general, there's about 1000 things to do right for a team, and the absolute best coaching ever done didn't get anywhere close to nailing them all. so certainly myself and other more successful coaches are screwing up constantly, and i'm just not sure double teams are meaningful enough for any of us to have any confidence we are moving the needle in the right direction, rather than the wrong one? i mean, i'm sure there are coaches who believe they are moving it in the right direction. but i remain skeptical?

My opinion is that yes, this is flawed game logic, or at least the set up is not as intelligent as it should be. Specifically yeah, when you double 2 players, of course there should be *lots* of easy buckets for other players unless you 1) have distribution set up to heavily favor the players doubled, and/or 2) you have bad passing and/or IQ. If you complain about this, CS will likely respond (as adlorenz did to me when he was still handling it) that while it can be powerful, it’s risky, and it doesn’t always work. But while it’s true they don’t always work, I have really not seen strong (or any) correlation at all with either of those previous conditions for when doubleteams don’t work. It honestly feels more like did your team get hit with an off day, bad biorhythms, etc? Oops, I guess your doubleteam gambit didn’t pay off, you got lit up. It feels more like a risk card you can play, usually pretty strong in reducing the effectiveness of the target, but sometimes backfires in that it just has little effect for no apparent reason; so again there should definitely be a more intelligent way to operate it, but that isn’t the game we have, unfortunately.

Something to consider regarding double team "if leading scorer" is a good way to set up a double team when you have 1 solid scorer you want to "always" double, but when he comes off the floor, the second lead scorer is now the primary scorer and likely will meet the criteria for "if leading scorer" at this point. Maybe I'm not thinking about this appropriately from the game design perspective, but logically, this is how I would apply it.

I use the "if leading scorer" a lot. I look at their players' season scoring, and if there are 1, 2, or 3 players that do most of the scoring, I put the "if leading scorer" on all of them. I rarely use "always", and I really can't see how you could put "always" on more than one player. But maybe that is effective with this game engine.

I agree there is some weirdness with double teams as a whole. I don't think it's flawed game logic or anything tho. Using the double team setting means that the engine attempts to double team always. Sometimes the defenders don't/can't get there in time so the double team doesn't come. That's completely reasonable in my opinion if you set two players to Always.

Also I feel like as coaches we feel like when we look at a box score, we're looking for the doubled player to go 1-18 with 8 TOs. When that's not realistic. Scorers still score. A double team on MJ will not keep him in single digits for an entire month. So I look at the goal of double teams to 1) hold the player to a shot or two less attempts/makes. 2) one or two more TOs. 3) and maybe even a bit more fatigue or end game failures.

If double teams on a player can accomplish that (in comparison to the players normal performances against the elite teams), then I feel like it's done it's job. And to me it seems like that is very often

My big issue (and others it seems) is not knowing how to measure the impact of double teams and not knowing how the Sim engine interprets a double-team (how does it figure into the equation). To address the latter…

If I try thinking through how the sim accounts for a double team… first, my understanding is the double-team will reduce shot attempts, reduce efficiency and potentially cause more turnovers on the double-teamed player. And in doing so, the non-double teamed players on the offensive team will generally be more efficient and the defensive team will pick up more fouls. Knowing this, I think the two easiest variables for the Sim engine to manipulate the formula/game logic to create these intended results would be the Def ratings of the defensive players on the court and the shot distribution created by the opposing offensive coach (the one with the player to be double-teamed).

With that in mind, my guess is some variation of the following two things occur before the Sim engine performs the calculation:

1. The Sim engine would give a boost to the Def rating of the player guarding the double-teamed player (assume man-to-man) and reduce the Def rating of the other players on the court. Essentially it’s a reallocation of the Def rating among the players on the court but it probably can’t be zero sum because one of the effects is the defensive team picks up more fouls (and lowering the overall Def rating of the five players on the court would achieve this).
   1. Using made up numbers, let’s assume the player guarding the scorer gets an additional 15 points added to his defensive rating to plug into the sim engine formula, and each of the other four defenders reduces their Def rating by 6 (6x4=24, net loss of 9 Def rating in total). This would reduce efficiency and potentially cause more turnovers for the double-teamed player, while also making it “easier” for the other players on the team to score and causes more fouls for the defense.
2. The Sim engine would manipulate the shot distribution created by the offensive coach by reducing the shot distribution for the double-teamed player by “some amount”.
   1. Using made up numbers, let’s assume the coach has allocated 30 to their star player and 70 to the rest of the players. The Sim engine would reduce the star players distribution from 30 to 21, so instead of 30 out of 100, the star player now gets 21 out of 91 (23% of total allocation instead of 30%). This step would reduce shot attempts for the double-teamed player.

After making the above two adjustments to the inputs for the formula, the Sim engine would run the calculation and see what the results are. Also, it’s possible the adjustments in 1 & 2 above are based on a percentage of each player’s rating (i.e. Def rating is increased by 15% or some other percent, not a set amount of 15, same for the reduction to Def rating and shot distribution). If so, this could be a flaw in the logic because this could defeat the purpose for double-teaming to help out a poor defender—the percentage increase to Def rating for a low-rated defender might not be enough to offset the percentage decrease to the Def rating for the other defenders if they have higher Def ratings.

Also, same concept could apply with a zone defense but the Def rating reallocation would skew the overall zone defensive ability in some way. I don’t know exactly how the zone defensive ability is calculated—I just know Gil has clarified in a few posts that the defensive abilities of the five defenders on the court are blended together to create a single defensive ability. But I wonder whether it’s a single defensive ability for LP shots and a separate defensive ability for PE shots or a different combined defensive ability for each position… I’m not sure how this works and I don’t run zone.

I have no data to support the above theory, but it’s how I’ve tried to make sense of it in my head based on what I’ve read and seeing the results when I double team an opposing player. I only double team in limited circumstances as I think the negative effects outweigh the benefits against most teams.

Another thought on man defense double teams… maybe it makes more sense that the Def rating boost for the on ball defender is a product of the other four defenders Def rating. Say the other four defenders have a combined Def rating of 240 (4x60). The four defenders would lose say 10% so a total of 24, and so the on ball defender would only get a Def boost based on a percentage of the 24 (so maybe 75% or 18). This would make more sense to me than my previous examples for how to reallocate the Def rating among the defenders. And makes sense that the boost is based on the “help” defense to come, doesn’t factor in the Def rating of the on ball defender.

I double multiple players all the time. I do not read the box scores lol. It could be hurting me or helping me I have no idea.

I use it mostly in the way Gill described: put the double team on the 3pt shooters and then cheat negative. Let's me win rebounding battles against man teams. I've also tried the inverse (double the lp threats and cheat positive) but in my typical 2-3 zone I haven't noticed many benefits to this.

I will add that if a team has 1-2 primary 3pt shooters with heavy distribution, I'll "always" double both and then, since one of the ways an offensive player can get a turnover is a foul, I'll throw an "ILS" double team on their low distribution 3pt shooters, who will probably not be the leading scorer except for if the primary guys are in foul trouble. This maintains the scheme through some wonky-ness and also, protects from a surprise high distro to bench 3pts if the other coach is targeting my perimeter D.