M2M pass coverage logic is absolute GARBAGE

[Dean-Atlanta]

Is this sim data from simming against one team each week for that data?

[Dean-Atlanta]

Below is your data with data from my testing added. My data is from simming 500 games (50 games each vs. 10 teams) simming against the best teams and game plans (re-created from scouting) in the league. This is much more rigorous than simming against just one team, and it does tend top drive the numbers up. But a play that still puts up low stats by this testing will be much better against any given team. This is a very new approach I started doing near the end of the season, and the defense I used vs Minn at the end of the season was based partially on this data and testing.

AZ32pm20 441 4.6 5.8
PEPMB33S 414 4.6 6.3
AF30pm5z 395 4.8 6
DN23PM14 413 4.9 5.7
SF23PM4S 400 4.9 6.9
DB31PMR3 424 5.0 6.9
DB31pmM3 424 5.1 5.7
DN23PM10 431 5.1 5.1
DN23PM12 399 5.1 5.9
DN23PM05 441 5.2 5.6
SF30PM4S 450 5.2 5.9
DN23PM13 437 5.3 5.4
OR22PM2R 392 5.5 5.9
PEPME24L 449 5.5 6.2
SF33PM2C 154 4.3 6.3
DB41pmRZ 145 4.4 7.7
AZ2PMIAZ 172 2.1 6.7
DB2PM01 155 2.9 5.3
LA32PMIA 101 3.1 6
SF23PM4S 164 4.0 6.9
PEPME24L 173 4.1 6.2
MN32MB2 184 4.4 7.2
DN23PM09 203 4.5 5.8
MN32MB6 162 4.5 6.3
SF32PM2D 183 4.5 6

[Dean-Atlanta]

Last season I was doing the kind of simming and testing I think you have been doing, against one team. This approach only has limited effectiveness. Your defense gave up 335 points last season, mine gave up 388. Both defenses were broken. The Jets gave up 215 over the season.

Better data = better game plans.

"Better" data is not always lower numbers but data produced by more rigorous testing, testing against the best teams.

[Mitch-Oilers]

Good Luck!