Analyzing a league for parity, balance and realism

[superben21]

You might want to change each individual coach's RB1/RB2 setting, as well as sub settings and ratings. Also, try changing the coach's run/pass percentages.

Thanks RevanFan, I had forgotten about the coach's run/pass percentages those should make a big difference.

[torontogrudlies]

You might want to change each individual coach's RB1/RB2 setting, as well as sub settings and ratings. Also, try changing the coach's run/pass percentages.

Hmmmmm...would you recommend changing these for the HC, the OC, or both?

[torontogrudlies]

Okay so I simmed a season and the defensive stats came back nicely.
Leaders were:
Tackles: J. Vilma - 171
Sacks: J. Peppers - 18
Picks: D. Revis - 7

Offensive stats on the other hand didn't quite work out to well.
Completion percentage: A. Rogers - 66%
Passing Yards: R. Fitzpatrick - 3,528
QB Rating: B. Roethlisburger - 102.7
Passing TD's: B Roethlisburger and P. Rivers - 30
Passing Int's: T. Brady and P. Rivers - 5
Rushing Yards: R. Rice - 2455
Rushing TD's: R. Rice 21
Rushing Attempts: 543
Catches: K. Boss - 113
Receiving Yards: A Johnson - 1611
Receiving TD's: C. Johnson - 17

There were 7 runners over 1500 yards and 2 over 2000 yards. After Ray Rice the next closest had 420 attempts. That's 123 less than Rice! The QB stats seem to be way low while the rb stats are way to high. The receiver stats don't seem to bad though. What sliders would you guys suggest changing? Or would it be attributes? Thanks!

My planned M.O. on this will be to first adjust the player and coach ratings/attributes, till a simmed season closely mirrors the typical results of a real-life season. This will include scrutiny of the scores as well.... if for example it's the early 2000's Bucs vs. the early 2000's Ravens (or their equivalents in my League) the final score shouldn't be 48-41, but rather something like 13-10.

Then I'd go to work on the sliders--the CPU ones first. Perhaps pick a team and total control simulate it thru 16 weeks, see if their standings, stats and results closely approximate the result when I straight simmed. Then finally working on the human sliders, till MY results are close to the other results.

[torontogrudlies]

I'm thinking it might be time to get busy with Visual Basic again. Anyone want to help?

I might want some script which will go into the various player stats tables in the .fra file, and automate the process of totaling various stats and averages. My overall idea was to sim an entire season, then compare the results against typical results found in a real-life season. The end part, tallying up all these totals, could be tedious, especially if I'm adjusting ratings, then re-simulating the season.

Possible example output:

FRANCHISE RESULTS, SEASON 6: 5 BLOCKED FGS
NFL AVERAGE: 21 BLOCKED FGS
RECOMMENDATION: DECREASE KICKER SPD/ACC ATTRIBUTE

FRANCHISE RESULTS, SEASON 6: 11.1% FUMBLES BY RB
NFL AVERAGE: 5.35%
RECOMMENDATION: GLOBAL INCREASE RB CARRY ATTRIBUTE

Initially, here's a rough list of what to track:

Total pass yards
total rush yards
pass attempts (coach pass/run sliders will affect)
rush attempts (coach pass/run sliders will affect)
sacks (LT PBK? QB AGI/BTK? DL various attributes?)
ints (QB throw accuracy/AWR? DB CAT/ACC/AGI?)
penalties (linemen AWR?)
completions
fumbles by rushers (CAR)
fumbles by receivers (CAR)
yards after 1st hit (BTK)
yards after catch (SPD/ACC/BTK)
drops (CAT)
fg block (K ACC/SPD?)
fg attempt
fg made
fg% 50+ yards
punts blocked
kr avg
kr td
pr avg
pr td
# blowouts by 28 points or more (gauges parity between teams....more than 24 would indicate too much disparity between teams)

These would all seem to be items which can be pulled from the franchise file, can be compared to real-life NFL stats, and can be adjusted via changes to coach sliders and player attributes.

I figured that rush TDs, pass TDs aren't as important, they're more of a result of the other factors. 200 rushing or passing yards is usually going to also translate into 3 or 4 TDs, etc.

Need other thoughts on this, and if there's anything I'm overlooking....

[superben21]

I defiantly be willing to help. I'm getting a little busy with school starting back today and having a speech I have to give to the city in a couple of weeks. But I'm willing to give as much time to this as I have. Let me know what I need to do.

[torontogrudlies]

I defiantly be willing to help. I'm getting a little busy with school starting back today and having a speech I have to give to the city in a couple of weeks. But I'm willing to give as much time to this as I have. Let me know what I need to do.

Do you do any programming in VB?

[superben21]

I defiantly be willing to help. I'm getting a little busy with school starting back today and having a speech I have to give to the city in a couple of weeks. But I'm willing to give as much time to this as I have. Let me know what I need to do.

Do you do any programming in VB?

No, however, I was reading up on it and might be able to learn some of it? I don't know how possible that would be, but I'd be willing to give it a try.

[torontogrudlies]

A good way to learn is by looking at other code and what it's doing. In one of the other topics I posted the code I used for the box score generator, gives a good general idea of how to manipulate stuff. Do you have access to anything like Visual Studio or tools like that?

[superben21]

I checked around and I could get a copy of Visual Studio 2010 Express, its a free trial but you can always just re-install those. Also IDK but my college may have some good software available for use. I'm willing to give it a try if you think it would be worth it.

[torontogrudlies]

One thing I need is a stats site which tells us "yards after first hit/contact" for a RB, and "yards after catch" for a receiver.

I may move this topic. Rather than a franchise guide, I think this is going to turn into a developed app which will scan a franchise, and compare a season's worth of such stats with typical NFL numbers (probably kept in a spreadsheet.)

So far I'm envisioning a piece which tests the league's parity: how many blowout games, avg # wins by division champs, # of shutouts, and # of "cellar dweller" teams (3-13 or worse).

Then a piece which gives info about the pass-run orientation: how many teams are fairly even (59/41 or closer) and how many are pass heavy/run heavy.... And a tally of 4000 yard passers and 1000 yard rushers.

Then, a bunch of checks of various average and total stats, for comparison against the real-life numbers: such as sacks, carry/fumble ratio, # of kickoff TDs, # blocked FGs, etc.