New Sticky Note

NOTICE: This league is using the BLEEDING EDGE game engine. For more information, click here.

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 - Beta Chat

Re: [Update] Game engine version fbecf8b

By jdavidbakr - Site Admin
11/08/2018 2:43 pm
I didn't make any changes for last night games but I did add tracking for play distance and gameplan rules called. When I looked at Setherick's game earlier and noticed his distribution was so heavily toward medium and long passes when his gameplan didn't seem to call for it, I got a little concerned. Did you guys experience the same thing where later in the game your gameplan seemed to get abandoned for no apparent reason? If so, I'd like to look to see what logic fired during that time.

Re: [Update] Game engine version fbecf8b

By setherick
11/08/2018 2:46 pm
jdavidbakr wrote:
I didn't make any changes for last night games but I did add tracking for play distance and gameplan rules called. When I looked at Setherick's game earlier and noticed his distribution was so heavily toward medium and long passes when his gameplan didn't seem to call for it, I got a little concerned. Did you guys experience the same thing where later in the game your gameplan seemed to get abandoned for no apparent reason? If so, I'd like to look to see what logic fired during that time.


Infinity has repeatedly seen this for the past year. We brought this up multiple times when we noticed the base defense bug.

I wouldn't have been shocked if my play calling had been explained by the 35 GP adjustment that I use since that pushes things to longer distances on second down.
Last edited at 11/08/2018 2:47 pm

Re: [Update] Game engine version fbecf8b

By Infinity on Trial
11/08/2018 2:54 pm
I don't have an example in front of me at the moment, but offensive playcalling is frequently outside of statistical probability of matrix settings (I rarely use rules on offense).

I rarely have this problem with defense, but I only use rules on defense. However, there are rare instances in which a particular defensive play will be called obsessively for no apparent reason.

I'll post examples moving forward, but I've dramatically cut back my leagues/participation. As setherick said, I've been beating the drum about problems with offensive matrix for a long time.

Re: [Update] Game engine version fbecf8b

By raymattison21
11/08/2018 9:22 pm
https://mfn1.myfootballnow.com/log/7651

Here's a log from when I was glitching the long pass and play abuse. We had only 3 plays offensively in the matrix and this game bailed on that plan in the 4th. Defense calls were good, but I need to double check .

It sought out plays in my book , but out of formation /set 90% of the time , and mostly long passes were called(like my matrix )and a few medium passes (not in my matrix).

It sounds like the bug. Once I took the plays out of my book and or setting slow down and hurry up to zero that fixed it. Or worked around the bug.

If I can find that plan I will run it next game to see if I can get it to reproduce similar results .

Some other games I noticed this late in the 2nd quarter as well. My blame perhaps was directed to the hurry up mode for that one. As some games these settings were set to the max.

And this is all with no rules as well.
Last edited at 11/08/2018 9:23 pm

Re: [Update] Game engine version fbecf8b

By raymattison21
11/10/2018 7:01 am
raymattison21 wrote:
https://mfn1.myfootballnow.com/log/7651

Here's a log from when I was glitching the long pass and play abuse. We had only 3 plays offensively in the matrix and this game bailed on that plan in the 4th. Defense calls were good, but I need to double check .

It sought out plays in my book , but out of formation /set 90% of the time , and mostly long passes were called(like my matrix )and a few medium passes (not in my matrix).

It sounds like the bug. Once I took the plays out of my book and or setting slow down and hurry up to zero that fixed it. Or worked around the bug.

If I can find that plan I will run it next game to see if I can get it to reproduce similar results .

Some other games I noticed this late in the 2nd quarter as well. My blame perhaps was directed to the hurry up mode for that one. As some games these settings were set to the max.

And this is all with no rules as well.


No calls out of my game plan . ...

Re: [Update] Game engine version fbecf8b

By setherick
11/10/2018 7:44 am
General Observations

* Pitches are fixed, and they make outside running more viable without any changes to the OL. That's a bonus.

1) https://mfn1.myfootballnow.com/watch/8902#1625366
a) Something is off with the running algorithm. The best way for the RB to turn here is to the sideline and trying to get up to the marker. Instead the RB turns back to the middle of the field and loses yards.

2) https://mfn1.myfootballnow.com/watch/8902#1625375
a) Same thing. Better example.

3) https://mfn1.myfootballnow.com/watch/8902#1625392
a) Another weird, super fast pass into a bunch of people.

4) https://mfn1.myfootballnow.com/watch/8902#1625394
a) Is this because the TE is very slow and has terrible route? The TE definitely does something weird here. He tries to stutter, but then freezes in place.

5) https://mfn1.myfootballnow.com/watch/8902#1625400
a) Another "throw away" where it looks like the QB is throwing at the WR.

6) https://mfn1.myfootballnow.com/watch/8902#1625402
a) Two in a row.

7) https://mfn1.myfootballnow.com/watch/8902#1625419
a) Another one for the other team.

8) https://mfn1.myfootballnow.com/watch/8902#1625422
a) Why does this ball end up so short?

9) https://mfn1.myfootballnow.com/watch/8902#1625423
a) Better question: what's wrong with accuracy in general?

10) https://mfn1.myfootballnow.com/watch/8902#1625426
a) Here's one where the QB should have actually thrown it away since he isn't going to run for the WIDE OPEN first down.

11) https://mfn1.myfootballnow.com/watch/8902#1625427
a) Something is definitely off with the running algoritm. Watch the RB here run head long into the defender without trying to readjust.

12) https://mfn1.myfootballnow.com/watch/8902#1625501
a) Still not sure what happens on these plays.

Re: [Update] Game engine version fbecf8b

By setherick
11/10/2018 11:32 am
Add this question: Why do WRs come back 5 yards (!!!) on a curl route?

https://mfn1.myfootballnow.com/watch/8902#1625462

Re: [Update] Game engine version fbecf8b

By jdavidbakr - Site Admin
11/10/2018 12:05 pm
Infinity on Trial wrote:
I don't have an example in front of me at the moment, but offensive playcalling is frequently outside of statistical probability of matrix settings (I rarely use rules on offense).

I rarely have this problem with defense, but I only use rules on defense. However, there are rare instances in which a particular defensive play will be called obsessively for no apparent reason.

I'll post examples moving forward, but I've dramatically cut back my leagues/participation. As setherick said, I've been beating the drum about problems with offensive matrix for a long time.


There were some fixes in 0.4.3 to the matrix, but I do smell that there might still be an issue. I need to have some examples in MFN-1 week 12 and beyond that illustrate it, though, so I can look into my more in-depth logs for the play selection logic to see if that assists in finding out what might be going wrong.

Re: [Update] Game engine version fbecf8b

By jdavidbakr - Site Admin
11/10/2018 1:47 pm
Couple changes for the next games, I'm tweaking the pocket positioning algorithm. May not look different but I hope the result is that QBs will stay in the pocket better and not float out so frequently. Also tweaking some of the throw-away reporting logic to be a little more realistic looking.

Re: [Update] Game engine version fbecf8b

By jdavidbakr - Site Admin
11/10/2018 1:59 pm
Wanted to comment on these as well, real quick:

setherick wrote:

1) https://mfn1.myfootballnow.com/watch/8902#1625366
a) Something is off with the running algorithm. The best way for the RB to turn here is to the sideline and trying to get up to the marker. Instead the RB turns back to the middle of the field and loses yards.

2) https://mfn1.myfootballnow.com/watch/8902#1625375
a) Same thing. Better example.


Noted, but I'm going to take this out of the scope of this update, this has more to do with the RunToEndZone logic and less with the route running logic. I believe what is happening is that he is favoring the direction his momentum is taking him, but tweaking this will also impact RB cutbacks (remember back when RBs would cut back over and over and the defense couldn't keep up? I think we're running into the code that prevents that here.)

setherick wrote:
3) https://mfn1.myfootballnow.com/watch/8902#1625392
a) Another weird, super fast pass into a bunch of people.


I think this is a quick shovel pass to the RB, but it does look weird and should be addressed. It also shouldn't really end up as an interception.

setherick wrote:
4) https://mfn1.myfootballnow.com/watch/8902#1625394
a) Is this because the TE is very slow and has terrible route? The TE definitely does something weird here. He tries to stutter, but then freezes in place.


I believe you are right, more because he has poor route running. He is stutter stepping, and his route skills determine how long he stutter steps before continuing. I think a few things are happening - 1, he is getting probably the far end of the random distribution for how long he is stuttering; and 2, he is just coming out of his stutter when the ball hits the ground, you can barely see that he is starting to run again right at the last split second of the play.

setherick wrote:
5) https://mfn1.myfootballnow.com/watch/8902#1625400
a) Another "throw away" where it looks like the QB is throwing at the WR.

6) https://mfn1.myfootballnow.com/watch/8902#1625402
a) Two in a row.

7) https://mfn1.myfootballnow.com/watch/8902#1625419
a) Another one for the other team.


I'm updating some of the throw away logic for the next games.

setherick wrote:
8) https://mfn1.myfootballnow.com/watch/8902#1625422
a) Why does this ball end up so short?


I believe this should have been a drop, I need to investigate why there's no play-by-play indicating why it was not complete.

setherick wrote:
9) https://mfn1.myfootballnow.com/watch/8902#1625423
a) Better question: what's wrong with accuracy in general?


Hoping that the next update addresses this.