There's a small chance I was being overly dramatic in my last update.
I put quite a few bonus keyboard hours in since then, and I think I have everything sorted out. Here are the heavy-hitters:
Slow WRs being useless: Even in Ex1, slow WRs were kind of useless. I thought I'd made enough changes to the engine so that it's possible to have a highly skilled slow WR be effective. It didn't work at all. The Oranges have a skilled slow WR crew and they couldn't do a thing. What I ended up doing was rethinking how I handle WR skill. I've tried a bunch of things to model WR skill, but they haven't been working very well. Now I'm using something I call "coverage radius" to model the difference between WR skill.
I'm assuming that any WR, no matter the skill, can run routes. What separates the good WRs from the bad ones is the ability to catch. Specifically, the great ones can catch in traffic.
This is a good visual. The WR on the left is rated 81, and the one on the right is about 60. The circle around them represents the distance they are considered 'uncovered' for catching calculations. The better WRs will have a much smaller coverage radius than worse ones. This means that a great WR could be covered fairly tightly and still have a high probability of making the catch. A really bad WR will have a lower chance of making the catch if any defenders are around. The net output of this is the ability to catch balls in traffic. Slow WRs with great skill can now catch balls when defenders are around. Bad WRs need to be wide open. The catching calculations also take skill into account, but in a pretty uncomplicated way.
I also implemented decision making time for the defense breaking on the pass. For lower skilled CB/S, they don't break on the ball immediately, and skilled WRs will sometimes get that little extra room.
Breakaway Plays: Still had waaaaaaay too many breakaway plays. I think the player pool for Ex2 made it worse. This one took a lot of experimenting, but the final solution ended up being relatively simple. It has to do with intercept points and decision making.
Here's the Greens RB breaking a run outside and turning upfield. That little yellow sphere is what the Orange player at the top calculated as the intercept point to hit the RB. It's actually correct. There's nothing wrong with that calculation. However, if the Greens player deviates at all from that path, the decision making time of the Orange player could cause him to continue to aim at that yellow sphere for too long. What usually happens then, is that the Orange player ends up aiming at a point behind the Green player and gets into chase mode. That almost always means a breakaway TD. I adjusted the intercept calculation so now the Orange player is aiming at the white sphere 3-5 yards upfield of the actual intercept point. That small change made a huge difference. The Orange player will take a conservative intercept angle until he's within a smaller distance to the Green player, at which point he'll change to the actual intercept point (tackle attempt).
And just like that, breakaway plays are not nearly the problem they were before!
QB Skills: This was easy. I tweaked the accuracy formula and got a better range of accuracy. While I was at it, I monkeyed with the progression reads and made it so highly skilled QBs can read progressions faster, and can cycle through them easily. Bad QBs will struggle to get through the progressions once. (But every QB can read the first progression)
Blocking: Made a lot of updates to the blocking system, and it's looking good now. Power offensive lines can move the DL at a good rate. Speed offensive lines can pass block like crazy. However, both need high OL skill to be effective or the DL will eat them alive.
Special Teams: (the Urquhart problem) The intercept angle update helped this a lot, but I also made changes to special teams coverage/blocking. Coverage players can break off blocks a lot easier now. Also, when the runner passes by, the defender can break free and pursue. This helps on those plays where Urquhart makes a ridiculous weave back through the blocks to get across the field. It always looked cool, but completely unrealistic that nobody hit him.
There were a lot of other small items I had to fix. I ended up taking care of them when I got stuck on these big items. In the end, it looks like Ex2 is still a go!