League Forums

Yeah .....I am only interested in your test parameters and results. Knowledge of successful play combos should be posted. Granted you were allowing for that knowledge in the audibles, but I cannot wait. Even though these results were only tested to prove public theory.

One other test test should be equally rated guys, but one technical team and one athletic. Speed is not as valuable under this code.

As of right now wr/dB is messed up with knock downs, deep zone and YAC is over the top. Catch overs seem better along with running the ball. Others have noted the return game, DTs rule pass pressure and LBS are having cover issues also.

So, time spent on this is another layer to the game. As other parts of code are tightened hopefully audibles , like familiarity , like the effect of coaches and crowd noise are kept low. Cause other issues will still be here, but I like this to be tied to coaching staff.

Right now it would come down to which coach has more money plays . ....not to cool, but if "elite" coaching staffs could have in game adjustments rating we might be talking.

I would like the order of plays calls to have an effect also. Set up a draw play or play action fake. Or if they are blitzing inside a pitch will work better or if they blitz the outside a delay will be more successful.

Each offensive play was run against each defensive play 10 times with everything being reset and as many neutral inputs as possible (so, neutral defensive play calling, fixed experience and fatigue values, all attributes equally set at 90, etc). There were only 11 plays that had an overall average of less than 2 yards, but all plays had some combination of defensive plays that resulted in both a large average gains as well as consistent losses.

My idea of pushing you the ability to audible into a play that is naturally an effective counter to a play the other team is overusing is based on the fact that no matter how good your team is, there are some plays that will not succeed against certain other plays, period, but there are some plays that can beat those plays that are being overused.

I agree with raymattison21 on a lil more importance to our coaching staffs. Coordinators especially. It could add more depth to the game for sure.

I still think playbooks should come with the coordinators and not the HC. And if OC's and DC's match the HC's playbook the players would get some sort of familiarity bonus.

Players themselves could play better under certain systems than they do in others. Say a QB that favors a West Coast offense gets a familiarity bonus if playing in that type of offense and has a penalty in other offenses.

I still do like the overuse penalty to some refined degree and used with audible. Which I think audibles is a great idea to enrich the game as a whole. But I dont know if relying on audibles alone will curtail exploiters and GM's who overuse the same handful of plays. On the surface it seems like it could invent a new way for them to exploit.

Maybe eventually combining overuse penalties, audibles, coaches, systems, limiting rules, and min amount of plays that have to be used in a game plan will have an effect. Either way one thing is for sure, this game just keeps evolving.

Ah, sorry... I misread the first post as saying each position was set to 90 overall, not 90 to all attributes across the board.

That still leaves the question as to whether this was with bleeding edge or the current game engine, unless I missed that somewhere too. If it's bleeding edge, everything I'm saying here can be ignored, 'cause I don't know how significant the changes have been.

Also, you're saying that each defensive play had offensive plays that averaged large gains... but of course they did. With an n of 10, there's no way that won't happen, especially with how frequently passing plays 'work' no matter the circumstances (every game you'll see at least a few examples of a back shoulder throw completed 30 yards down field into triple coverage while being sacked). It also only takes one busted run in an n of 10 with the rest getting stuffed behind the line to end with a large positive, and if you're testing 60+ running plays, naturally a handful of them will likely bust one (or two). I know it increases the test time ridiculously, but I think we need a sample of 100 minimum to establish any kind of average for a given pair of plays.

I think with that kind of simming done, you'll see that the effectiveness of most rushing plays goes down considerably against a certain subset of defenses (not coincidentally the ones being run by the majority of teams currently). Zone will be largely ineffective against most plays, and most importantly won't be any better at defeating those plays that may 'break' those aforementioned certain subset of defenses, meaning those will still be the best to play in any given circumstance. Maybe I'm wrong in my predictions, but I do think that 10 is not close to sufficient to telling us anything particularly useful about individual match-ups, and would be misleading to use in terms of telling users what should 'beat' a certain play.

Actually, one way to verify the results is to rerun the test 2 times with n=10. If the result is extremely close if not exactly the same, then you would establish repeatability of the results with the test and would effectively have an n of 30. An n of greater than or equal to 23 would be considered statistically significant given a standard distribution.

It's using the bleeding edge engine.

Let's all keep in mind that 0.1 is a net positive. It doesn't mean you'd ever use that shotgun run in a game plan.

The complaints that we have all had are about the efficacy of plays.

I'd like to see all the offensive plays run against ONLY the 1-Deep Man, 2-Deep Man, and Man defensive calls since that's what the majority of players run on defense. I run a lot some 3-Deep Man as well, but I bet I'm in the minority. I know of no one that runs a true zone defensive play.

'Efficacy'? Wow, I had to ask the wife what the definition of that word is...

Mr JDB, I think the idea has great merit, as long as it was simple enough to use and the ability to 'exploit' audible was kept to a minimum. It sounds great.

Would it be difficult to run the play test with the present build? It would be interesting to see if there are more extremes with the success/failure of the plays. It would also show if the bleeding edge build is fixing the problems that we are presently experiencing, which would be awesome...

What I've actually done is run smaller-scale tests (i.e. just pass plays) and dump the passing stats to see completion percentage, int percentage, etc, with various predetermined attribute values (i.e. passing accuracy at 100, 90, 80, 70, 60) - this did reveal some issues that were already somewhat expected based on you guys' observations, but helped me fine-tune what was causing it. So, I do know the changes are fixing issues.

The 0.4.1 version has diverged from the 0.4 game engine too far to efficiently backport the testing part into that version, though.

Play overuse and hot reads need to be rethought. Some times hot reads get called in a way that actually punish a team and some times play overuse actually punishes a team, and other times your opponent can just run the Nickle Strong Double LB CB3 blitz all game and completely shut down your game plan: https://paydirt.myfootballnow.com/log/328

I actually lost this game because of turnovers, so that's on me, but I'm struggling to understand how hot reads can sometimes be called 8-12 times in a game against a team that is running the send the sink blitz plays over and over. Other times, we're back to playing 0.4.0 all over again, and no one wants that. No one.

To be very clear for a minute, the fact that play overuse does not have a more dramatic effect on game plan (i.e. gets called enough) is the clearest evidence that positional attributes mean very little in the game right now. Play overuse only really "worked" in testing when it just made players stop moving so the other team could run away for TDs. This was terrible, and we shouldn't go back to it, but I think that play overuse tells a lot more about the value of positional skills than it acts as a deterrent for players to not run the same plays.