New Sticky Note

The new user interface is in preview!

Want to check it out? Click here! (If you don't like it, you can still switch back)

League Forums

Main - General MFN Discussion

Re: The Playcalling Matrix Randomizer

By raymattison21
9/17/2016 8:49 am
setherick wrote:
I could try something similar in 75, but I've been dicking around with my playbooks so much in the other leagues, I thought I'd do something more normal.


It will take a while to get anything consistent out of that league for a bit. QBs rocked preseason. I attribute it to defenses learning new schemes .

Re: The Playcalling Matrix Randomizer

By setherick - League Admin
9/17/2016 8:58 am
raymattison21 wrote:
setherick wrote:
I could try something similar in 75, but I've been dicking around with my playbooks so much in the other leagues, I thought I'd do something more normal.


It will take a while to get anything consistent out of that league for a bit. QBs rocked preseason. I attribute it to defenses learning new schemes .


My secondary actually held my opponent to 10/31 with 2 interceptions. (Fingers crossed that that continues.)

Re: The Playcalling Matrix Randomizer

By raymattison21
9/17/2016 9:10 am
setherick wrote:
raymattison21 wrote:
setherick wrote:
I could try something similar in 75, but I've been dicking around with my playbooks so much in the other leagues, I thought I'd do something more normal.


It will take a while to get anything consistent out of that league for a bit. QBs rocked preseason. I attribute it to defenses learning new schemes .


My secondary actually held my opponent to 10/31 with 2 interceptions. (Fingers crossed that that continues.)


Wow! There were a hand full that did bad. I think SAT (san antonio?) Both qb had a rough day. My back up CBs both missed tackles and got beat deep twice after catch overs, but i though my D did as expected. I faced young QBs with legit weapons. So i am not too worried yet.

I also attribute the better qb play to future drafts end up flooding the league with talent. I have only done one other allocation, and it seems talent is weak in allocations in comparison. . I picked DBs late but they are all 60's in rating.

Re: The Playcalling Matrix Randomizer

By setherick - League Admin
9/17/2016 9:22 am
OL and DB are the two positions I've started targeting early in the draft unless there is a can't miss player there.

Re: The Playcalling Matrix Randomizer

By MistbornJedi
9/17/2016 10:16 am
I've wondered about the impact of the percentages as well. Haven't done the math, but there have been several times when I've had a run-heavy gameplan that still results in more than 50% pass plays. The best way I've found so far to control it is by what plays are the in the playbook, but I don't want to go so far as only having 3 plays as in the example talked about here.

Re: The Playcalling Matrix Randomizer

By raymattison21
9/17/2016 10:25 am
setherick wrote:
OL and DB are the two positions I've started targeting early in the draft unless there is a can't miss player there.


Grade A oline! And your DBs have nice experience. I would not say the same for the rookie QB Huff. A simple blend of straight man, cover 1, and 2 with blitzing. A good one vs. a young QB, but i would prefer to see this work in the playoffs against a vet QB. I see Huff only making to the playoffs at best. I doubt your QBs will be doing well in the playoff either but if so it will be because of the oline and that pass D.

Things will be different as teams learn there scheme. My team only knows like 6 plays and i have fully developed guys on half of my offense. My D knows none. Or a few guys are vets. I think that will be the difference this season.

We will know by week 8 who is in and who will be competing for high picks. My team is on the fence right now, but i will not be in that race. If anything my team will gain steam.

Re: The Playcalling Matrix Randomizer

By jdavidbakr - Site Admin
9/17/2016 11:45 am
setherick wrote:
Does the play calling matrix even work? I spent several hours last night building a new game plan. Only to watch the matrix NOT CALL A SINGLE PLAY THAT SHOULD HAVE HAD PRIORITY.


The main place to look is 'view plays' on your matrix to see the weight of each play. I've seen times where users spend a lot of time setting their personnel and play call values but don't have plays in their active gameplan that match what they're trying to do. (Not saying this is the case with what you did, but checking the 'view plays' gives you the logic the AI uses).

The AI chooses a play thusly:

1. Choose inside run/outside run/short pass/medium pass/long pass
2. Assembles the plays that match this play type and assigns the weights you see on the 'view plays' screen.
3. Chooses a play based on weights (larger weights = higher probability)
4. If no play was available (i.e. you don't have any inside running plays with non-zero personnel and the selection was an inside running play), a play is randomly selected.

Re: The Playcalling Matrix Randomizer

By setherick - League Admin
9/17/2016 2:32 pm
jdavidbakr wrote:
setherick wrote:
Does the play calling matrix even work? I spent several hours last night building a new game plan. Only to watch the matrix NOT CALL A SINGLE PLAY THAT SHOULD HAVE HAD PRIORITY.


The main place to look is 'view plays' on your matrix to see the weight of each play. I've seen times where users spend a lot of time setting their personnel and play call values but don't have plays in their active gameplan that match what they're trying to do. (Not saying this is the case with what you did, but checking the 'view plays' gives you the logic the AI uses).

The AI chooses a play thusly:

1. Choose inside run/outside run/short pass/medium pass/long pass
2. Assembles the plays that match this play type and assigns the weights you see on the 'view plays' screen.
3. Chooses a play based on weights (larger weights = higher probability)
4. If no play was available (i.e. you don't have any inside running plays with non-zero personnel and the selection was an inside running play), a play is randomly selected.


I know how to do this. And I use the View Plays all the time to make sure that my plays line up with my game plan. Do I need to post a screenshot with the HB Counter out of the 221 as the top play that should be selected?

The point was that the AI choose from the available plays, but not the ones that were weighted the highest except the short pass out off the 221. And that play is only weighted at about 75% of the Counter, which was not called.

It doesn't matter in the long run because this game plan idea was a colossal failure.
Last edited at 9/17/2016 2:33 pm

Re: The Playcalling Matrix Randomizer

By setherick - League Admin
9/17/2016 2:34 pm
raymattison21 wrote:
setherick wrote:
OL and DB are the two positions I've started targeting early in the draft unless there is a can't miss player there.

I doubt your QBs will be doing well in the playoff either but if so it will be because of the oline and that pass D.


You mean you don't like my "oh **** what happened to all of the QBs" pick at QB? My team will make the playoffs if I can run for 3500+ yards. If I have to pass, at all, forget it.

Re: The Playcalling Matrix Randomizer

By eyeballll
9/17/2016 3:49 pm
setherick wrote:

I mean that's what Infinity did - one rule to select from 3 plays on first down only and another rule to select from 3 plays on second down and one rule for defending the 113 and pass rush the rest of the time. He dominated league after league with that simplified approach. I guess that's the only way to do it. Stupid.


This is a pet peeve of mine, guys that limit their plays to a ridiculous unreasonable degree. Not even a team of 8 year olds runs only 3 plays on 1st and then 2nd down. The fact that Infinity 'dominated league after league' with this approach doesn't fill me with awe of his success, it's not something I can respect at all. It's undeniably successful, but it's a shortcut, it's lazy.

I've seen this strategy for years, going back to the old online FBPro96 leagues. My understanding is that the new game engine will adapt to the this strategy, and that makes me happy.