FOF9 Historical Player Generator

[Passacaglia]

Hey, I think my file is ready for anyone here. I've been working on it in the Off Topic forum, and if anyone wants to watch my progress on that, it's here:

Just a moment...

I'm actually on v4 of the file, with more versions to go. Here it is:

fof player generator 1920 2024 v4.xlsm - Google Drive

Edited to add version 4.1 here -- fof player generator 1920 2024 v4.1.xlsm - Google Sheets

And v5 is here -- fof player generator 1920 2024 v5.xlsm - Google Drive

Here is v6 -- fof player generator 1920 2024 v6.xlsm - Google Drive

v7 -- https://drive.google.com/file/d/16ef...ew?usp=sharing

This file will go back to whatever year you want, from 1920 to 2024, and create a universe that starts in that year, adds in everyone in the league that year and puts them on the teams they played for that year. Then it adds players for all future years, making them draft-eligible in the years they enter the league.

I'm still fine-tuning, so I'd love thoughts on it while I'm still doing that!

There's a handful of inputs, but hopefully it's not so bad. I'll go over them:

Start Year -- This is the year you want to start playing.
Salary Cap -- This is the salary cap your league is going to have, in hundred thousands. If you don't know what it is, it's probably 2248. If you do know what this is, it's the number that goes into your league_info and your xxxx_info file.
Delete Columns E and H -- These columns need to be filled in after you do the first step. If you've already filled them in and don't want to have Step 1 delete them, make this NO.
How many years to go back -- This is a weird one. The game needs more players than I have. So, to fill the FOF rosters, the file needs to either create fake players, or go back to previous years and get players who used to be in the league. Think Phillip Rivers this year, or the QB your team drafted in the 7th round two years ago who never saw the field. The question here is, how far back do you want the file to look for players before it gives up and starts creating fake players.
End Year -- This is the last year that real players will show up in your universe. Data only goes to 2024, so that's the highest year you can put here. Of course, you can still play the game after that, but when you get past this year, it will be standard FOF fictional.
Quality Balance Factor -- Lets you adjust how good the players are. Higher numbers will get you more good players.

Once this stuff is filled in, you can hit the "generate players" button.

Columns E and H -- This is a tricky one. You'll need to go into your default_teams.csv file to figure out which numbers go with which teams here. It's a pain, but it's done this way to accommodate multiple league structures, as well as ensuring continuity. This requires some knowledge of up to 4 different file types (default_teams.csv, league_info.csv, xxxx_info.csv, and x_y_z_schedule.csv), but I'm here to help you through that if you don't know what to do. Also, the team names here are wonky, but should be pretty self-explanatory -- for example, if you see BRAVESREDSKINSCOMMANDERS, you'll know that that's the team that started out as the Boston Braves, became the Boston Redskins, then the Washington Redskins, then the Washington Commanders. It's done this way so that the file knows a player stayed on the same team, even if the name changed -- I didn't want it to look like a player was on a different team just because his team changed from Redskins to Commanders.
Number of teams in your league -- If you're versed in those 4 file types I mentioned, you can have however many teams in your universe as you want. The game needs to have a set number of players per team when you start, and a set number of rookies per team coming in each year, so the file needs to know how many fake players to generate. Don't worry, they're all benchwarmers, so you won't see these guys taking the spotlight away from the real players.
Pop Up Progress Reports -- This file can take a while to run, especially if you're going back to early years, so if you like a message box to pop up every so often to let you know it's still working, put YES here.
Make sure there is at least one player for each OVERALLRATING at each position -- Players are rated from 0-9, with 9 being the highest. It's progressively rarer as you go up, and only the top players at each position get a 9. If you don't have a lot of players in your universe (because your start year and end year are not very far away), you might not have enough for anyone to even be in the top 0.3% at some positions. If you put YES here, it makes the best player a 9, the 2nd best player an 8, making sure you have at least one player with each rating for each position.

Once that's all filled in, click "assign teams" -- and the two tabs it created should be ready to save as xxxx_quarterbacks.csv and xxxx_players.csv

Save Files -- If you put the directory where your player files are located, this button will put the files there so you don't have to copy and paste.

Download Player Images -- If you put the info for your saved game here, this button will look on wikipedia for an image of each player. Probably all you need to do is change the user name from agage to your own. You can enter in the years you want to get images for. This process takes a while, to make sure you don't get in trouble for violating wikipedia's scraping policies, but the good news is that you can start playing, and each image will just start appearing once it's been downloaded -- you don't have to start a new universe to get them in, or even restart the game.

[NawlinsFan]

[Passacaglia]

Don't go too crazy, I already know I need to make a version 4.1 - it looks like the "look back" players are getting a rating similar to their careers, instead of getting a 0.

[Passacaglia]

Here comes 4.1 at last!

fof player generator 1920 2024 v4.1.xlsm - Google Sheets

Fixes:

1. The Brett Favre bug -- when you are bringing back old players out of retirement to fill your universe, they now come in as bad players like you'd expect for someone who hasn't been in the league for a while, as opposed to looking like they are at the peak of their careers.
2. The Roger Staubach bug -- when a player was drafted but didn't actually play in the league for a few years, they now come in on the team that drafted them, instead of being free agents.
3. Added draft info. Purely cosmetic, but it's nice to have these filled in on player cards. Note that players drafted in a supplemental draft are shown as being drafted 100 rounds later than the round they were supplementally drafted in.
4. Made salaries more precise. Player salaries are set in the following way: In the first year, salary=bonus, and salary+bonus=1/53 of salary cap, for all players. A player is on contract for as long as he stayed on that team in real life, max 5 years. So if you're loading a 2008 season, Peyton Manning will be on the Colts, and have a 3-year contract, since he spent 3 more years on the Colts in real life. Salaries go up 10% each year, but bonus remains flat.
5. Fixed the progress indicator. I got rid of the percentages which weren't working, and added more indications to show what the macro is doing instead.

[NawlinsFan]

Clarification please. I create a 40 team league starting in 1980. I create a player file for the same. The needed additional players required to fill out rosters will be added to the players the file creates for the base year. I believe the answer is yes, just checking.

[Passacaglia]

Originally Posted by NawlinsFan

Clarification please. I create a 40 team league starting in 1980. I create a player file for the same. The needed additional players required to fill out rosters will be added to the players the file creates for the base year. I believe the answer is yes, just checking.

Hey sorry, I must have missed this!

The answer is yes. When you click "assign teams" the file looks at how many teams you said were in your league (cell J7), and adds dummy players to the players it already has. It ALSO adds players to later base years, to make sure it has enough players to fill in the incoming rookies.

If you don't want that and are actively looking for extra random good players to fill in your extra teams, I can put in a switch to prevent the file from creating extra players. Or, you can just put a really low number in cell J7 -- if you tell it you only have like 8 teams, the file will not create any dummy players at all, and nothing else will be affected.

[Passacaglia]

And v5 is ready!

Here are the changes:

1. I've increased the number of "dummy" players that the game creates, to make absolutely sure you don't get any game-generated players. This is going to be especially important once the image generator is in (that will be v6).
2. I've added an asterisk to the front of both first and last name. It looks more weird, but I'm pretty sure I've seen the game alphabetize by first and last name at different times, so I wanted to let people be able to sort these guys out in both places.
3. I've added the ability to save the files into your directory for you. It's nice to not have to take the tabs this thing generates, copy it, open the file you want, save it, for both QBs and players, so this button does it for you. Just be careful you don't overwrite anything you want to keep.
4. I tried to make it look a little more clear what the steps are. I've also added a "run all" button that does everything at once. The buttons are organized into steps because once you've generated all the players, the files pastes into Columns D and G all the teams it found that have players, and you need to tell it what ID to assign each team based on your default_teams file, in Columns E and H. But if you already know what you're going to get in D and G, and have the IDs already placed in E and H, you can just run it all. Again, be careful that you don't overwrite files you want to keep, since this will save the files, too.
5. I fixed an issue where a player who played on two teams was ending up on the team that came first alphabetically, and now he's on the team where he played more games.
6. Some attributes are in! Scramble, kick return, and punt return. As I was working on scramble, I reworked how the file decided what position everyone plays. I think it's better now.
7. I added a "salary cap suggestions" tab, that gives you a couple options for what you should set your salary to based on what year you're playing. It also generates the minimum salaries for the different experience levels, which you can paste into your league_info and xxxx_info.

There's some image stuff in the lower right, but that's not functional yet.

fof player generator 1920 2024 v5.xlsm - Google Drive

[NawlinsFan]

Salary CAP is an interesting try. I have attempted a number of times at lowering the overall CAP for earlier season starts. I found there is a fine line between the overall salary CAP and the experience minimums. To high and they game does some interesting roster moves. Even so, my experience has been that the game will still apply contracts that can all but ruin a franchise. I have seen some contracts, while using reduced CAPs, that absorb 40+% of the overall CAP.

[Passacaglia]

Originally Posted by NawlinsFan

Salary CAP is an interesting try. I have attempted a number of times at lowering the overall CAP for earlier season starts. I found there is a fine line between the overall salary CAP and the experience minimums. To high and they game does some interesting roster moves. Even so, my experience has been that the game will still apply contracts that can all but ruin a franchise. I have seen some contracts, while using reduced CAPs, that absorb 40+% of the overall CAP.

The game's documentation recommends keeping the experience minimums the same proportionally to the salary cap -- I just added those numbers to the salary cap tab, so it stays the same. But yeah, I'm not sure how the game will process these differences.

[NawlinsFan]

Originally Posted by Passacaglia

The game's documentation recommends keeping the experience minimums the same proportionally to the salary cap -- I just added those numbers to the salary cap tab, so it stays the same. But yeah, I'm not sure how the game will process these differences.

Yeah, I think what I will try to do, which I haven't and should have based on that i, is to just create a simple spreadsheet reflecting the proportionss for various seasons as a base. Even so though I have still seen some inflated contracts based upon the overall CAP. I haven't found anything that notes what the game uses to determine those contract ceilings.

[Passacaglia]

Originally Posted by NawlinsFan

Yeah, I think what I will try to do, which I haven't and should have based on that i, is to just create a simple spreadsheet reflecting the proportionss for various seasons as a base. Even so though I have still seen some inflated contracts based upon the overall CAP. I haven't found anything that notes what the game uses to determine those contract ceilings.

You can actually use the "salary cap suggestions" tab of this file for that! If you put you cap number in cell H2, the numbers to the right will use the proportions that are used in the original league_info file. If you just want to have the proportions themselves, they're in cells Q1:W1.

[Passacaglia]

I feel pretty good about how the image downloader is working, and I wanted to get out a version that fixed an error in v5, so here's v6 with images! I still need to test it a bit, but so far so good.

Just a heads up -- the image downloader takes a LOOONG time. There's a lot of delays built in to make sure you're not violating wikipedia's scraping policies. It took me 24 hours to do all of 1966, which had 24 teams, so I guess an hour per team for the first year of your file, but then future years should take less time since it's just the rookies coming in. It also ties up your excel -- if that's annoying, let me know, and I can give you instructions to run it from the cmd prompt -- I just had it run from excel so it's all-in-one.

https://drive.google.com/file/d/1Quo...ew?usp=sharing

[NawlinsFan]

Originally Posted by Passacaglia

You can actually use the "salary cap suggestions" tab of this file for that! If you put you cap number in cell H2, the numbers to the right will use the proportions that are used in the original league_info file. If you just want to have the proportions themselves, they're in cells Q1:W1.

Sweet

[korme]

Here is a screenshot of my league in 1978 with 40 teams. Nearly every QB is very good.

Spoiler

[Passacaglia]

Thanks Shorty! I ran a 1978 file with 40 teams, and a 1966 file, and I'm seeing the same things, too.

A couple things:

1. Some of this is due to the fact that I have a lot of dummy players added to the file, because I *really* don't want the game generating players. I'd already attempted to account for that, and create a distribution based on how many there "should" be.
2. I'm using some numbers that KingZal gave me to set the distribution, but they may be too high.

0 53.40% Replacement level
1 23.90% Backup
2 12.20% Starting
3 4.00% Better starter
4 2.60% All-Star
5 1.50% All-Star
6 1.10% All-Star
7 0.60% All-Star
8 0.40% Best at his position
9 0.30% Generational Talent

I think the answer to both of these will come from looking at the player files that come with the game -- 2023, 2024, and 2025.

1. I'm combined these files to get a hopefully better feel for what the denominator should be for that distribution. In the past, I'd looked at various league sizes, looked at how many players the game seemed to need at each position (in the start year and future years), then added 25% as a cushion. When calculating the ratings (i.e. taking 0.3% of all players and making them a 9), I took out the bottom 25% and just used the number of players the game seems to think it should have. But in the files that came with the game, there are less players, so I'm going to try using that instead.

Here are the numbers I have for this. The first column is position, the next two columns are how many players per team the file generates for that position (first column is start year, 2nd column is future years). The next two columns are average height and weight for normalization (you can ignore them for this). The next column is how many players per team there are for that position, and the column after that is Column F / Column B * Column C -- just a scaled-down version of Column C to estimate how many per team of each position we should expect to be included in ratings.




2. I looked at the same files to try to match the distribution in them, and here's how it compares to what KingZal gave me:



I've looked at this before, and knew that the 9s and 8s from the game files would come in lower, but I didn't realize that 2-5 would be higher. This may actually work against us, so I'm going to stick with KingZal's numbers for now.

[Passacaglia]

Man, so I wish I could say that I came up with some data-driven way to make this balance the way it should. But after a lot of time spent playing with using different denominators for rating than the one used for roster sizing, trying to figure out what a percentage of "the population" means -- is it all the players in the game, the number of players in a file that comes with the game, the number of players required to fill out rosters for each team -- along with trying different distribution spreads -- numbers from the files that come with the game, KingZal's numbers, interpretations of what each rating number means -- I eventually just gave up. Instead I took what I already had, and just scaled down my "rating population" (columns F and G) until when I ran it, a 32-team league looked like what you get when you load the 2025 roster. So while I can't really show my "work", I can show my results.

Here's what you get when you start in 2025 using the player file that comes with the game:



And here's what I get loading a 2015 run of my file. This has 32 teams, so it should be similar:



Now for Shorty's league. I expect ratings to be a little better here, since there are 40 teams:



And as a bonus, here's my 1966 run. There were 24 teams then, so I expect ratings to be a little worse:



EDIT: One more. This is what you get when you run the game for a 32-team league without using the game's player file:

[Passacaglia]

And here's the new version, with that in mind.

Fixes:

1. Improved the image downloader so that it is less likely to get the wrong person. It also takes much less time, and no longer ties up your excel while it's running.
2. Reduced salaries so that teams are less likely to cut good players immediately.
3. Improved the position assigner -- biggest impact here is that I reduced the required rank to qualify as a RB from 2.5 to 1.5 times the number of teams in the league. This prevents guys that are clearly true FBs like Kyle Juzczyk and John Kuhn from getting labeled as RBs, while still letting players listed as FB like Jim Brown, Larry Csonka, and Franco Harris end up as RB, like they would be called if they played today.
4. Adjusted quality so that it more closely matches what you would see in a typical FOF game. I also added a Quality Balance Factor that you can play around with if you feel the need to adjust things.

fof player generator 1920 2024 v7.xlsm - Google Drive

[korme]

I've got the new version generating a player file right now, Pass. First two attempts gave me a runtime error when entering the draft, so I'm back to doing Steps 1-3 one-by-one, and I think maybe I had the salary cap too low. I'll report back if I go 0 for 3 here. But I just want to say thanks for doing all of this, it's really fucking cool.

edit: Started in 1950, and it the generator worked! I think when I was doing the All Steps button, I hadn't prepped the columns for team assignments and it was erroring out in game. But I did it manually on step 2 and we are in business.

And the best part is that the ratings overall look excellent. There are maybe 10 quality starting QBs in 1950. That feels good. Thanks dude! I'm going to go waste most of my Sunday now.

[Passacaglia]

Nice, I'm glad it's working out for you!

Yeah, those two columns are what I was talking about when I wrote "only do this if you know what you're doing" for the run all button. Step 1 looks at your start year, finds the players that were active that year and what team they were on that year. It uses that to generate Columns D and G, and you need to populate Columns E and H with the codes (or 0s if you don't care). Hitting "run all" is basically saying "I've done this before, and I know what I'm going to get in D and G, and I have E and H already set with what I want" and "I'm okay with Step 3 potentially overwriting any files I already have in that directory" -- I separated it out because some people might want to save it somewhere else in case they're just testing things. I should put a warning in Step 2 if any of E and H are blank, and offer an option to treat blanks as 0s, rather than letting it create a file that the game won't read. Next version.

Keep me posted on 1950, I've never run it for that era! What's your league structure like? Did you put players on their actual teams, or use a preference draft? Also, did the image downloader work for you?

[korme]

I took NawlinsFan's 1980 fictional merger 40 team files and just edited it down to be 1950. I used a preference draft, so I zeroed out step 2 in your tool.

Image downloader is running! I'm curious if it will catch me. It's currently processing in 1959, and I fastsimmed to get some league history but I've slowed down now. I'm up to 1968. It's fun to look back at the league history or some of the older guys and see their pictures. I have Unitas as my QB, and it's cool that he has a photo. I wish FOF made it more easy to transfer this work league to league.

I am thinking the default QB ratings are still a bit high now.
As of 1968 QB ratings,
2 80+
9 70-79
11 60-69
A ton of QBs in the 50s before I can't figure out how to flip to the next page on the player finder screen. I know now I can adjust that for future leagues, but maybe worth running a few tests yourself.

[korme]

I've messed around with the quality balance factor and the ratings still seem to be way too high across the board. I'm wondering if having 40 teams with such a small pool of real players is increasing the challenge of more appropriate ratings distribution.

[Passacaglia]

I think it is -- when I run this for a 40-team league, it looks to me like that ratings are appropriate. But I haven't actually played it.

Some things I'm curious your thoughts on:

1. Where specifically are you seeing the ratings that are too high? The number of 70s, 60s, or 50s? Or all of them?
2. If you run this for a 32-team league, would you still think the ratings are too high?

[korme]

I’m at work, but offhand, there are just too many quality starters. Realistically I shouldn’t have to scroll through any position where the entire first page is 50+ rated players. There just shouldn’t be starting quality players at each position for every team with a league tha big in that time. I’ll try to get some screenshots up for you tonight.

[NawlinsFan]

Reading Kormes post I had a thought. Would there be any way to have the utility nerf the ratings based upon the start year? using Kormes setup as an example. If you start in the 50's the player ratings would nerfed to try and not have as many high ratings. Then nerf it less for the 60's, 70's, 80's, etc. Just thinking out loud.

I am happy Korme is making use of the various files. I found myself pulled away with other things and do need to try and get back here.

[Passacaglia]

Would you want that to happen just depending on what the start year is, or are you imagining a situation where in one player file the ratings "grow" over time? I'd argue that to some extent that's happening already -- you're definitely going to see QBs getting better as you get closer to present day. Maybe that's where some of the differences are coming from between what korme and I are seeing, though -- if I was testing from an earlier year, I might be seeing the ratings come in lower than he did. I'll look into that.

[NawlinsFan]

Based on Kormes post I have been trying to construct a 32 team schedule with the vision of having the current NFL teams start in 60 and move forward. Right now I am working through a 16 game/no bye week format. Then, as a package, I hope to have graphics for various decades, 60/70/80, etc., as well as player sets from Passacaglia's geenrator. Then you could start at the beginning of any decade. Trying to commit time to this but life kicks my ass at times. Still, working it.

Now Jim - give us 10! Oh, it would also be nice if we could add specific team fields to add to/create some more immersion.

[Passacaglia]

kormre and NawlinsFan, here's a question for you about your 40 team leagues. If you're trying to use real players in a year where there are 32 teams, what is it you are looking for teams 33-40 to do? In my head, I've been trying to make this so that when you expand your league size, the 40th best QB will be about as good as the 32nd best QB in a true league. But is the issue that that's not what you're going for, and you're looking to have teams deal with the fact that teams 33-40 are going to have a QB that's actually worse than what we consider a marginal starter in a 32-team league?

I've been playing with this starting in 1920. There's so little quality -- players rated in the 30s are stars. The first 4 seasons, everyone was an RFA, but now I'm in 1926, and those guys are asking for tons of money.

I thought this was too exploitable, because I was doing way too good. I drafted a QB in the 5th round in 1922 who is like 36/36 and just killing it. I went 8-5 that year, lost the NFL Championship, then went 10-2-1 and won the NFL Championship, then 13-0 and won it again. Then my QB held out after I spent all my money on free agents, but came back in after only missing 4 games -- we still went 10-3 but lost the Championship. Now his contract is up, my spending is close to the cap, and I have to decide if it's worth paying this guy or going with my backup and making improvements in other places. I do think I was taking advantage of the serious modding for a while, but that it's starting to settle into being normal.

[NawlinsFan]

I feel that having the additional players fall into a population that will represent a moderate quality. Maybe no stars, 2-3% elite, 20% good, 60-70% average and the remaining as roster fodder. With this at QB your on point. They need to be competent and fall in to the lower third of the historical class. Not upper tier but moderately capable of helping a roster that has other depth to offset it.

For your point on wages. I have consistently found that no matter how you adjust the CAP after a few seasons the contracts become almost unmanageable. I have had moderately good QB's (and not just QB's) have expectations that were equal to 25+% of the overall CAP for the first year of a multi-year contract. With that I also see other teams chasing the CAP by signing then releasing players due to the CAP constraints. It's almost as if the engine understands the CAP for that year but fails in managing them well in the long run. For me I have found I start to drift away from those elite players while trying to build a roster with better then average players with depth in a few areas. They tend to have more reasonable contract requests then do those superstars.

[NawlinsFan]

I am running v5 and after adding the team abbreviations in D and G, starting in 60, resetting the CAP and setting the teams to 32 I run it and get a runtime 13 error. After acknowledging it it goes to the roster sheet and indicates it is processing player 1 of 27971 players. Question, is the wrror something you have seen? Also, does the processing player count change as it is processing? I never looked at it before.

[NawlinsFan]

NVM - running v7 now.