I didn't catch this during testing because I wasn't throwing Long passes, but the decision clock doesn't appear to be working. Every long pass ends up as a sack.

Also, every incomplete pass is being counted as a drop.

I know I'm right now. Look at this pass: https://rivals.myfootballnow.com/watch/79#14401

The QB waits forever to throw it and then throws it well short of the target. THIS MUST BE FIXED.

QBs throw long passes early in a WRs route and let the WR run under the ball. In MFN, QBs are waiting until receivers finish long routes.

Can you run it with the Rivals teams, and post the numbers?

I added an important edit to this post that I didn't want to get missed.

Also, another example: https://rivals.myfootballnow.com/watch/78#14265

Two things that I finally noticed that may break open the problems with sacks.

1) If the defense sends more defenders than there are OL, the OL doesn't know who to block and therefore blocks no one. The OL sits there and spins like it can't figure out which man to pick up.

I'm pretty sure something similar is happening to OTs. If the DE lines up on the edge, the OT does not know whether to block the DE or block inside. Since all OL block inside now because of how DTs used to overpower Gs, the OT blocks inside and allows the edge to be free. Remove the everyone must block inside logic, and I bet you see a dramatic reduction in sacks.

Here's an example of what I mean. The RDE lines up over the TE, the LT should take the DE, but he doesn't know who to block, so he blocks inside: https://usflwfl.myfootballnow.com/watch/2373#444997

Here's an an even worse example. The RDE is lined up just outside the LT, but the LT spins around like an idiot trying to find someone to block: https://usflwfl.myfootballnow.com/watch/2373#445016

Here's the same thing as above: https://usflwfl.myfootballnow.com/watch/2373#445023

Any time a defender is outside the tackle, it's a sack. This didn't happen in 0.4.3. Why is it happening now?

Fix #1 here, and you've fixed the game.

2) Sacks happen more frequently later in drives or later in the game. I'm not sure if this is a similar buffer problem as with catches, but it happens.


Other things about sacks.

1) OT blocking bug where the OT blocks inside
https://usflwfl.myfootballnow.com/watch/2373#444979
https://usflwfl.myfootballnow.com/watch/2373#444981

2) OL in general sometimes turn away from their man at the snap of the ball and then jump to another person to block. I noticed this more than a year ago. I thought it had been fixed:
https://usflwfl.myfootballnow.com/watch/2373#444953
https://usflwfl.myfootballnow.com/watch/2373#444969

Here's a really bad example of this. On this play the TE and RT should double the LDE, and the RG should take the DT. No one blocks at all: https://usflwfl.myfootballnow.com/watch/2373#444994

3) QB decision clocks taking way too long on downfield passes. I'm certain that medium and long passes are just broken right now. I would remove all distance penalties from all route distances and let the QB decide who to throw to.

4) QB taking too long to decide to throw it away: https://usflwfl.myfootballnow.com/watch/2373#444967

5) OL are getting overrun by inferior DL. I want to say that at one point OL blocking got reduced in effectiveness. Was that ever changed?
https://usflwfl.myfootballnow.com/watch/2373#444972

6) Whatever happens on plays like this: https://usflwfl.myfootballnow.com/watch/2373#444998

Decision clock

At 04:51 the QB MUST THROW THIS BALL: https://usflwfl.myfootballnow.com/watch/2373#444881

Weirdness

Why is this ball "thrown away": https://usflwfl.myfootballnow.com/watch/2373#444983

Ball placement on med/long passes is consistently short: https://usflwfl.myfootballnow.com/watch/2373#444976

This is more of a long term thing, but on plays like this, the RB needs to go into pass protection: https://usflwfl.myfootballnow.com/watch/2373#444977

---

OL DO NOT BLOCK THE MAN OVER? WHY?

Take this play for example. The DL is lined up in a tight bunch over the two Gs and C. The RG ignores the DE that is lined up over him and blocks inside against the DT. The RT doesn't have a man over him, and can't get to the DE in time to block him, so this is a sack: https://usflwfl.myfootballnow.com/watch/2373#445002

Seriously, it can't be that difficult of a change to make OL not move around so much for no reason: https://usflwfl.myfootballnow.com/watch/2373#445011

It's like you have it set up so OL always block the same guy.

After you have sacks figured out, can you re-run this game? It's a great example of the problems with the OL: https://usflwfl.myfootballnow.com/box/2373

11 of those sacks came on long pass plays, of 27 long pass plays called in the game (40%). I think that if you have a game with almost 30 long pass plays, you should see a lot of sacks ... maybe that's just me. Is 40% excessive? Well, probably. 12 of those sacks came from 39 medium pass plays (30%). Note the play-by-play also when it says 'So-and-so completely missed the block' and separate those out (when that appears, the dice roll was so in favor of the defense that the block didn't even happen, so if that is excessive without cause that would be a separate issue).

I'm curious if the teams in this game intentionally had a long-pass-heavy game planning, but I'm not opposed to long-pass-heavy games having lots of sacks. I do, however, agree that the long passes are probably not getting thrown as soon as they should which may be why it is as high as it is.

Here is the play selection distribution in that game:

Outside Run - 10
Inside Run - 29
Short Pass - 32
Medium Pass - 39
Long Pass - 27

To me that seems excessive with medium and long passes, and 39 runs to 98 passes also seems excessive. 25% sacks is still a lot, and probably too much, but only one of those sacks came on short pass plays. Remember, too, that if you are extremely out of balance, the defense starts to key on the pass and that is going to increase sacks and interceptions. So to have 2.5x as many passes as runs again is going to make those passes ineffective.

Just to be clear, I'm not blowing off what is being said about sacks being high, but I do want to make sure this isn't just a result of overuse of passing. (And no, there was no modification to the pass rushing algorithm at all in 0.4.4, only to the QB pocket logic - which may indeed be increasing sacks just there.)

One thing to note also as far as sacks go, is that in the long-term testing, sacks were in line with NFL averages. And I did look to see if those sacks came in excessive blitzing schemes, and they don't really seem to be, so I feel like the issue is that the play calling was just too far out of balance. If we decide that's to aggressive then we certainly can explore that, but I'd like to make that a matter of a newer version rather than a hot patch. I may change my mind if we see some games with high sack numbers and more reasonable balance in the play calling.

In other news, I'm going to release a patch tonight that:

* Fixes the pitch play
* Fixes the 'drop' attribute for WRs (I did find a condition that was counting drops that shouldn't have)
* Shortened the pass point for streak routes - I'm treading lightly on this one since I'm making hot patches to non-beta leagues, but am curious if there is a noticeable decrease in streak route sacks

What time will this patch be, we sim in 16 minutes it won't be done by then right

P.S. Thanks for getting for work on this

Looks like it compiled and uploaded just in time.

Great thanks. I'll put a few plays in to see if there is still problems