I'm trying to settle a question/debate that has come up in my online sim league. My teams are performing horribly on defense despite reasonably good talent. But what I'm hearing is advice from more experienced players like, don't bother with base defense (4-3 / 3-4), you should be using at least 80% 3-3-5 and dime. Quote from one player who's a two-time champion: "I'd say quarter is probably my most common formation right now and we stop the run pretty well." Read: The quarter defense stops the run. That just plain should not work. Another experienced player told me quite plainly (and I quote) "real football strategy isn't going to help a lot." I was absolutely FLOORED at those statements! I was also told very plainly that the coach defensive playbooks are useless and that I'll "always be in the bottom of the league" unless I use custom playbooks. I already do on offense but defense isn't my forte so I've tended to let the CPU/coach handle that. No no no, I'm told. You will be at the bottom of the league if you don't customize. That just seems wrong to me. I can understand that a truly dominant world-class player will need custom plays and playbooks but to have to do so just to avoid being in the basement tells me there's some flaw with the game's AI or logic.

So all the football I've been watching for 35 years, all the strategy I've picked up from watching pro coaches and players for 35 years is moot in this game? From the discussion it basically seems that the defense in the game is not realistic and if I'm getting advice to go almost exclusively nickel (NFL teams I looked up average about 27% in base formations). While it does make sense that base defenses are easier to pass on, this does seem extreme. It just seems to me that the advice I'm getting on how to set up my defensive playbook and game plan, should not work. In a different league I noticed even the best defensive teams have few defensive players above 65 OVR.

I understand the dev team is limited and programming AI to act like real life is extremely difficult. The Madden game is 30+ years old with a huge team of programmers and practically unlimited budget and they can't even get it to work. This game is despite its issues still has a very bright future ahead and is ahead of Madden in many respects. I've already purchased pre-access for the College game and plan to for Pro 24. But I'll be honest it does suck a LITTLE of the enjoyment out if I'm feeling like I'm learning how to beat the game rather than play defense. But I'd like to ask a few questions from the devs: Do you think this is a fair assessment? What is the general attitude is in terms of how much priority is given to realism. How much does the game's internal AI try to replicate real situations? For example, if I run successfully does that truly open up the pass? Should 4-3 at least work against the run? What do you say to the advice I'm getting that real football logic doesn't really matter?

For online leagues they're probably right. Some of those players have been data mining and playing for so long that they have highly customized offenses dialed in, and therefore you'd need a highly customized defense dialed in as well... Online leagues have always focused more on winning and beating each other via any means necessary (including finding any advantage or flaw in game logic)

We did a lot of testing for the recent versions of the games so that every offense and defense combination has some sort of counter, and a lot of the online commissioners were involved to try and balance things. It sounds like we still have work to do... if you find scenarios that you think need to be buffed or nerfed by all means let us know and when we're doing our tuning passes we can make adjustments.

Yeah I've always been annoyed when I saw what looked to be strategy tips in Madden only to find out they were "money" plays designed to exploit flaws in the game's AI. I always considered that borderline cheating. I guess if EA sports can't get it right, it must be tough to close those gaps with limited programming staff. There are a lot of people who complain about this on Madden but as a developer myself I know that making players in a computer game think and behave like actual players is next to impossible so there are bound to be holes. I haven't personally noticed specific scenarios, but when people are telling me they can effectively run just 3-3-5 and nickel (even quarter!) and still stop the run (I"ll put that to the test real soon), that's a red flag for me. If I notice anything I'll let you know. Thanks for replying.

I would also imagine that ANY type of AI tweak between versions would, even if they didn't stop this process would create different holes so that it would at least reset it so those expert players would have to data mine starting all over again. But when I asked them about it they say that data mining doesn't really take long at all.

In a related question, how much does the game's AI have... how can I put this... an overall internal logic to playcalling? Since I can only set run/pass ratio by down and distance vs. game situation, I might focus on run in the game plan. However one game down by 14 with 5 minutes left I looked on in horror as my team was running and throwing 6 yard passes. Now, I did set run-focus and pass-short in the game plan. But I would think there'd be something internal that might modify or override that in this situation. I did set the plan to "pass more" when losing. Also, is there any way to control my team's logic for things like when to go on 4th down, when to go for 2, when to onside kick or call timeouts, or how aggressive to be on offense at the end of the first half. I notice that when a team is behind, they seem to call time outs more on offense, but they never do so on defense to stop the clock. And does the clock represent the time left at the end of the last play or the beginning of the next one? If we ran a play at 2:00 and then called timeout I'd expect the clock to say 1:54 at the start of the next play vs. 1:25-1:40 depending on tempo. And does/should end of game logic override tempo settings? Even if I have fast tempo set with a pass heavy offense, I'd like to run 90% of the time if I'm up by >8 points inside 2 minutes.

I don't know if there's a universal logic for this or if there are settings we have which influence these things. I just sometimes see head-scratching plays like I describe above. My team, in ordinary situations, will go for 4th and 3 at our 40 but pass up 4th and inches at the plus 20. I've made a suggestion that instead of just down and distance which doesn't tell the whole story we should have "situational" strategies also or instead. Such as when up by 10+ points in 4th quarter set the run setting to 10. If down by 7+ in the 4th quarter set pass setting to 85. 1st and long I may have pass set to 60 but if it's 1st and 15 and I'm ahead with 20 minutes left to go I want 2TE Jumbo and running.

The AI does have logic that kicks in, but it's more of a nudge than a complete override. It's a hard balance to achieve where we want game planning and strategy and coach hiring to mean something and we don't just want the AI overwriting the player's wishes. We used to do a lot more AI adjustments, but we got a lot of feedback that players didn't want us messing with their settings. So this does lead to odd playcalls at times, when your coach may decide, either based on personality, or game settings, to do something odd. If you find specific scenarios where we need to intercede we can certainly take a look, but again, in most cases we'll default to allowing the coach to make the calls.

Tbh I use real world logic when I play offline and it does seem to work but online I'm preety sure a lot of that stuff won't work because of how people build their online playbooks. Tbh I might be wrong because I dont really play online because I like roleplaying realistic football and do stuff like running double nose tackle and a ton of 4-3 and 5-2 fronts cause I like physical football teams that teams are afraid to run on (Proudly a Titans fan)