(80% accuracy attribute + 20% experience*) x (reduction based on scramble skill/movement) x (reduction based on duress) x (reduction based on distance) = accuracy for pass

FOV is not used in the error calculation, it is used to determine whether the QB sees an alternate open receiver or sees a defender in the area.

* "experience" is actually a more calculated formula that takes into account the player's positional experience, his experience with both the offensive and defensive plays, his fatigue, and crowd noise. His positional experience (what you see on his player card) carries the greatest weight, followed by his fatigue, followed by his play experience for each play, followed by the crowd noise.

To be as specific as possible, the accuracy of the pass? Or the accuracy of the best placement of the ball? It seems like placement of the ball since that was logical step number one before.

Also, I'm more convinced the bolded parameter should be commented out until pass blocking is fixed since it will come into play almost every passing play until then.

The optimal location to throw the ball with the highest chance of it being completed and the lowest chance of it being intercepted is calculated equally among all players. The likelihood of the ball actually arriving at that location is a function of the accuracy.

Probably an easier illustration is looking at field goals, since that's more obvious to see and the calculation it much simpler. The optimal location for a kick to arrive at the goal post is in the center of the post. All kickers aim for that spot. Their launch vector is adjusted based on their accuracy. That one there is no penalty for distance because a few degrees of error become magnified over a longer distance; the only elements that come into play are the kicker's accuracy and the holder's skill. (Kick strength determines how far the kick will travel and does not impact its direction).

Passes follow the same logic flow, there are just more factors that are considered that can reduce the accuracy.

This makes sense to me. I need to adjust my QB weights quite a bit based on this conversation then. I always started with Accuracy, but the code flow starts FOV and then Arm. It's only after those two are used that Accuracy comes into play, and then it's only as important as Scramble in reducing a penalty to the throw.

If I'm understanding the code flow correctly now.

100 FOV, 80 ARM, 60 Scramble (until pass blocking is fixed), 40 accuracy, 1 everything else should give me the best approximation of what a good QB is.

It would be nice if there was better documentation for the game physics, so these questions didn't keep coming up. I always thought that Accuracy meant how well the QB could throw the ball, not a factor in how well the QB could throw the ball to a particular point chosen by a combination of two other attributes and then impacted by a host of other attributes. Boy, was I stupid.

I must really not understand how Weights work.

Just as you, my 'Useless' Weights go to 0 (or 1, if I want to see something on the Player Card, like Long Snap on all my OL in case I need to find a Snapper not my Center quickly due to Injury).

But you seem to use a 20 Drop on each 'step'... does it require that much, or is it different for different things? Do you sometimes use a 10 Drop instead, if you feel the Abilities are closer?

For my DE, which for me is the #1 Position on the field, I demand 100 SPD, 100 STR and 100 ACC (rare, I know, but not so rare you can't find them). So I start with a 100 Weight on all 3.

Then I look at Run Stop and Pass Rush, which you'll need to use to get in position to use your Tackle... and thus Tackle follows. I believe I have all 3 of these Weighed at 95.

For myself, that's it... that's the 6 Abilities that matter as a DE in MFN-7 (from my understanding). Now I do use some Zone Blitz, and my main DE was a Converted MLB, so he happens to have really good M2M, Zone and Punish Receiver... and obviously Strip is a good Bonus Ability but not one I'm going to 'rely' on.

In fact, I may have changed my Weights so my Big 6 are all 100, and everything else is 1 or 0 (I think my Coverages and Punish are 1 so I can see them).

I use different approaches. I usually start with a 20 point drop between attributes I feel are essential to the ones I feel are useful. Sometimes I use a much bigger drop between these - I think my RBs are currently weighted something like 100 SP, 60 Break Tackle, 60 Carry, 40 Avoid Fumble.

I do a lot of things with weights to figure out how players work, and I don't often recommend people using the strategy that I do. It works really well for some positions, but not all. I definitely effed up my weights in the 75 draft tho.

New question: How likely is it for a player to have 100 experience? What is the average experience for QBs?

The way the formula currently works a player has a negative applied to their shown accuracy as soon as their accuracy attribute equals their experience. The question is how great that negative is. For instance, if we cap experience at 80, a 100 shown accuracy QB has an adjusted accuracy of 96. But if we cap experience at 66, a 100 shown accuracy QB has an adjusted accuracy of 93.2. That's a big difference.

EDIT: I've been thinking about this more. And I really would like to know how big the penalties are for having a defender in the QBs face and the penalty for distance.

I'm pretty confident how that first parameter will work. So let's start there.

QB has 100 Accuracy and a ceiling of 80 play knowledge. (This knowledge is probably generous since its nearly impossible to gain experience with defensive plays and very slowly with offensive plays, but we'll be generous and assume that the QB was pregerated with 100 knowledge in the play he's running, so 80 is a good ceiling.) With this setup the QB has an adjusted accuracy of 96.

For ease sake, let's assume that the QB has 100 Scramble or is not moving so that we can ignore the second parameter in the accuracy calculation.

Now we get to the wildcards player in the face and distance.

If the player in the face is a 20 point penalty that would mean on most plays a non-moving QB with 100 listed accuracy has an adjusted accuracy of 76 before distance penalties are applied.

Now, let's add back in the penalty while moving and assume that the QB is moving because QBs are often moving because of pass blocking code. Before we assumed a scramble of 100, but that's not common, so let's assume a scramble of 60. If the scramble is up to a 20 point penalty, let's guess that the QBs scramble negates 60% of that penalty. That would mean our QB with a listed accuracy of 100 would have an adjusted accuracy of 68 while moving and a player in the face before applying distance penalties.

Does that sound about right? Or are these penalties even worse than that?

I'm beginning to see why good QBs barely 50% passing for a season.

There are a lot of moving pieces to the equation, and I am continually trying to find the balance to make the passing game effective without making it too effective (obviously same with the running game).

One point I do want to make ... none of this is to be considered it its final form. I doubt it ever will be, even after I decide it's at least good enough to make it a 1.0 release.

We all understand that.

I just want to know how the physics and penalties work so that I can set my QB weights correctly and know that when he throws a 5/25 game with 4 INTs, which happens way more often then it should, I know it was my game plan that was the problem and not some effect of not having the right attributes for a QB.

I mean FOV and Arm determine the likelihood a QB is going to throw an INT? I had no idea. I would have thought it was Discipline, but I knew better.

Arm strength only limits the velocity of his passes - FOV influences the likelihood of him throwing an INT, as does his experience, as well as his accuracy, as well as the DB's man coverage skills and zone skills (if he is playing zone), and the WR's route running skills ... there's not a simple formula and that's intentional, but also adds to the apparent randomness. As the game matures I hope that it becomes more intuitive and also am able to document better the impact of each attribute on various results.