Here's a different look at how plays perform based on the 2041 season:

All Aggregated Data: https://drive.google.com/file/d/12EJraFRlW_PdBmoN-ctL7ZBk5h8ln_dU/view?usp=sharing
Aggregated Data Stratified by Redzone: https://drive.google.com/file/d/1NbgQ36fAV42zoRiTL3vW0R7WciTC_MyU/view?usp=sharing

What's This Thing
Other than the WeightPoints column you're looking at probabilities (based on a Bayesian statistical model) for various categories. These probabilities were than weighted against every defensive play a team in this league called against the play last season (2041).

Some specifics to know:
* WeightExpectedYards: This is an average of plays excluding sacks, which would make this column go very negative for a lot of plays and mess up the WeightSuccess column.
* WeightSuccess: I have changed this back to a linear formula.
** Yards Gained for Successful Play
*** Inside and Outside Run: 4 or TD
*** Short Pass: 6 or TD
*** Medium Pass: 8 or TD
*** Long Pass: 12 or TD <-- Just don't call these.

(The yards for Medium and Long passes are stupidly low, but that's just the nature of MFN right now.)

* WeightBigPlay: The probability that the yards gained by the play will be 2x what the value for success is.

Because a few folks have asked how to best use this. Here are some pointers.

* if you want the plays that provide the most consistent yards, look for plays with a greater than 0.4 weighted success

* if you want the plays that score the most points, look for passing plays with around 0.5 expected points and running plays with around 0.25 expected points

* if you want to minimize negative plays look for passing plays with less than 0.2 sack probability and the lowest negative play probability for running plays

* if you want to minimize interceptions, look for plays with less than 0.1 interception probability

Since sacks and interceptions are largely dependent on defense, the numbers may be a bit higher than 0.2 and 0.1 in the aggregate

While we're talking about expected points, it's good to remember that the game penalizes QB that throw more than 50 passes in a game where their accuracy drops wildly after 50. That means if you want to hit that magical 50 number, you have to balance your passes to be no more than 60% if you assume you're going to run 80 plays in a game. If you plan on fewer total plays you can go higher but then risk teams keying your passes.

Updated the data sheets after re-running the probability model with plays in the red zone. Let me know if you see anything interesting in the data. I have known for a while that passing plays are more effective in the red zone than running, but I haven't really explored this much.

Fixed the link. Now it's the right data.

Would love to see the maths, but would then need to learn to understand it...

I assume for the values such as WeightCompletion, just need to multiply by 100 to get a percentage, ie 0.495 = 49.5%

Not really. They aren't percentages. They are probabilities. So .495 means that there is a 50/50 chance of the play happening. Not an observed 50%.

Well, that just shows my maths ignorance.

Statistics show what happened. Probabilities show what could happen.

Semantics, schemantics.|

Adding that on my to do list to learn.