I asked something about this a long time ago, but I can’t find it now and never got a clear answer. So I will try again to clarify things.

When the AI dice rolls choose plays in the matrix, does the defense counter according to the chosen formation or the actual players in the offensive huddle?

It’s stands to reason that (especially in 4.6) if the defense automatically counters a 3 wide set with a nickel or dime defense that the owner on offense could choose to override the players in the formation with heavy goal line style personnel and basically bulldoze the undersize defense with the running game.

In 4.5 it certainly seemed trendy to go the other way with as much speed as possible with WR in the backfield to mismatch against slower/heavier base defenders. By the same token, it seems the teams that play this way like to eliminate DTs from their roster and likely substitute lighter across defensive positions as well. Is this strategy based on the way plays are chosen in the matrix?

To my knowledge the only thing that decides what defensive play is picked from the play matrix is the weights chosen by the sliders. I would think rules are intended for the exact scenario you're saying, choosing a defensive play based off the offensive formation. Along with that, I don't believe the defense has any way of changing the play based on the actual players at the positions(WR at RB/TE)

I may need to dig through some play by play logs, but if it is strictly the sliders there would likely be a lot more cases of goal line def against 3 wide sets or nickel/dime against base or heavier offenses, I think there is another level to matching up on AI. I know I need to start making rules, but I haven’t gone that route as I have mostly been pretty successful without them, and I figured it would probably mess me up more, at least at first.

To call any of the 'Attack #X' plays your sliders would just need a decent chance to roll DB man coverage and LB blitz 2/blitz 2+ along with the actual 'Attack #X' play being selected(assuming there are other plays that could be called) and to have your opponent call a play out of two, albeit sets with several strong plays, offensive sets. Not saying you are wrong in thinking there is more going into the defensive play matrix but maybe not as likely as you would think.

If anything I will be trying this out in a league to see what my results are with an equal chance to call a play from a base defense, nickel, etc. Will update :)

Yeah there’s a default AI selection based on formations. Or personnel, but for the 113 it’s nickle. There’s is downnfalls to the logic based the code we’re running but you can manipulate it. It’s not the best but I will run a strict 34 with a safety at WLB. You can mismatch it in the passing game but in general it’s solid vs the run and pass.

Nickle has a huge gap over the LG a lot . Very vulnerable to weakside runs if blitzes are not called. Really any D that doesn’t call defensive plays (non blitz) with proper gap protection will get toasted in the run game. Weak or strong you can beat them all with a different strategy as the 34 can get run on up the middle.

So take what you want but the regular 113 or 203 do well vs base nickel cause blocks had to be held longer to make up for differences in speed. Now it’s just overpowered....you can really see it on kick returns. Using big guys could be a thing. Size does matter a bit more at times but guys are still too slow based on weight

I don’t do rulees but 4.6 sees a need for them. Positioning and presnap alignments make for a set of mismatches. Blitzes do shut down the run if applied right but pass defense can suffer so to me there’s a balance. And I don’t think too small or too big is overpowered either. But mismatches are key.

TIL

So, if you don't use a Nickel D in your selection, would the AI go to base D (3-4 or 4-3) or Dime vs 113. Or would it be Coach dependant?

It’s been a minute but I think it goes dime. Like it always goes smaller. GL will get 34 and 43 split up by the matrix as they are equal. 212 will run nickle before dime/quarter.