FOF Historic Files - rev0 released

[Passacaglia]

I spot-checked a few of these info files, and it looks like they all have a salary cap of $100M and a minimum salary of $40K. If this helps any, here's my research on what the salary cap is in each year from this site

2023 2248
2022 2082
2021 1825
2020 1982
2019 1882
2018 1772
2017 1670
2016 1553
2015 1433
2014 1330
2013 1236
2012 1206
2011 1204
2010 0
2009 1230
2008 1160
2007 1090
2006 1020
2005 855
2004 806
2003 750
2002 711
2001 674
2000 622
1999 573
1998 524
1997 415
1996 408
1995 371
1994 346
1993 322
1992 300
1991 280
1990 261
1989 243
1988 226
1987 211
1986 196
1985 183
1984 170
1983 159
1982 148
1981 138
1980 128
1979 120
1978 111
1977 104
1976 97
1975 90
1974 84
1973 78
1972 73
1971 68
1970 63

I changed it so that it's in 100K increments, as the readme for the info file says it needs, instead of in millions. For years before 1994, I used a 7.34% average trend to go backward. The site says 2010 was uncapped, which is weird, but I'd imagine if you put anything in the 1200s you'd be fine. I think the league minimums should be changed, too.

Am I the only one noticing this issue here?

[Passacaglia]

Originally Posted by korme

I haven't taken a look thoroughly at the csv's yet - but does this mean we have the capability to make a historic coaches file as well?

Originally Posted by NawlinsFan

Yes sir.

Originally Posted by korme

That's almost equally exciting.

I'm trying my hand at creating a 1976_coaches file. So far, so...okay. I've been using ChatGPT to get a lot of this stuff, and 1976 was one year before I was born, so I'm relying on it to be correct with only minimal checking from Wikipedia, since I can't rely on my own memory at all. I've got all 28 head coaches, assigned to the teams they should be, with their birth cities and colleges. I got ChatGPT to give me position groups for them all, plus their offensive and defensive styles -- huge grains of salt here that GPT knows what it's talking about. There were lots of coaches who were balanced, and none with Erhardt-Perkins or spread, so I asked it to convert some of the balanced to those, but no idea if it put any thought into that. I also asked GPT to give me PAYSCALE numbers that I hoped would affect the coach's quality. I got good numbers -- high salaries for coaches I've heard of (Tom Landry, Don Shula, Chuck Noll) and low salaries for coaches I hadn't (Marion Campbell and Lou Holtz, who I didn't know coached in the NFL, so it works). But quality seems random, and from the text file

Coaches are randomly rated in Front Office Football Nine.

So I guess I shouldn't have bothered. This actually might have an adverse effect for all I know, where highly-paid coaches get fired and aren't in the league after a couple seasons. Something to keep an eye on.

Anyway, all the head coaches are in there, correct birth city and college, a reasonable effort to replicate coaching styles, but random quality. Drop me a PM if you want me to email it to you. I still plan to add coordinators, then analysts who will become head coaches, and maybe some unemployed future head coaches, before I look into figuring out how to host it myself.

[Passacaglia]

Originally Posted by Passacaglia

This actually might have an adverse effect for all I know, where highly-paid coaches get fired and aren't in the league after a couple seasons. Something to keep an eye on.

I did one sim, and the coaches I've seen fired were Lou Holtz, John Ralston, Ted Marchibroda, Jack Patera, Bart Starr, Bum Phillips, and Bud Grant. It took until 1979 for that to happen, which I like, since it kept things consistent for a while. They all found gigs as HC or coordinator somewhere else, except Patera, who ended up as an assistant. I feel fine with all that, but if anyone thinks the pay numbers are throwing things off, let me know.

[Passacaglia]

I made a change to the 28_6_16_schedule.csv file if anyone is interested. That file sets the league up to rotate 3 different schedules. However, one of those 3 schedules has all 16 games played over 18 weeks, instead of 17. That in itself is a little weird, with every team getting two bye weeks, but the 18th week was getting counted as a playoff game for record-keeping in the coach's profiles, so when I realized I could fix it, it had to go. I just took out the bad schedule, so now there's only two instead of three, but I don't think that will matter -- the schedules are based on division ranking from the previous year, so I don't think it will feel repetitive.

I'll host this with the 1976_coaches file when I'm done, but if you want to change it yourself, all you have to do is delete the rows with a "2" in the first column, then change the rows with a "3" in the first column so that they're a "2"

[nilodor]

Originally Posted by Passacaglia

I've been using these files and starting in 1976. A couple things I've noticed:

1. A lot of players seem like they are only on 1-year contracts. Maybe that's true to the time, but most players were out of contract by 1977. There were four teams that had 0 players under contract when the 1977 year began. Not necessarily a huge deal, since they have a chance to sign their out-of-contract players again before anyone else can. But I'm not sure the AI is aggressive enough to do that.

2. Most teams were way below the cap as 1977 started. Somewhat a function of the fact that so few players were under contract, but also, the amounts that players were asking for didn't seem to sync up with the salary cap. The number in the 1976_info file is 1000, and that gets the $100M cap the game has, but $100M seems like too much given what salary everyone asks for.

I'll have to take a look. The way I did the years was I looked at how long the player was on the team for. So if you're using a 1977 file and a player didn't play in 78 or were on a different team, they would have a 1 year contract. If they were on the same team until say 85 they would have a 5 year contract.

As far as the overall salaries, I had scaled it against talent. So the most talented team would have a total salary of ~100 million, but that's the most talented team ever. The salaries are then scaled by their individual rating, so a 9 overall would have a much higher salary than a 1 overall. Same with experience, a veteran would have a higher salary than a rookie. I'll have a think on how to scale the salaries better, I would agree that it's a bit low. It wasn't an issue in previous versions, but seems to be more of a deal here.

[Passacaglia]

Originally Posted by nilodor

I'll have to take a look. The way I did the years was I looked at how long the player was on the team for. So if you're using a 1977 file and a player didn't play in 78 or were on a different team, they would have a 1 year contract. If they were on the same team until say 85 they would have a 5 year contract.

That seems like a pretty good way to handle it. I don't think it's working right, though. I think I said before that four teams have everyone hit FA at the end of the year. I don't remember them all, but Houston was one, and if you take your 1976_players file and filter on team 5, you'll see that the salaryears field is 1 for all of them. The most notable single exception I see is Joe Theismann, who hits FA at the end of the year. The one time I've simmed to 1985, it looks like they franchised him, and the next year he signed with Minnesota. I'm not sure if that all is a huge deal, since you have to expect players to be on different teams from reality, but having the four teams with no players under contract seems like it will put them in trouble. I think some easy solutions are to either a) add a year to everyone's contract, or b) put a 2-year minimum on every contract. I've created a file where I added a year to everyone's contract, and it seems like that's working well.

As far as the overall salaries, I had scaled it against talent. So the most talented team would have a total salary of ~100 million, but that's the most talented team ever. The salaries are then scaled by their individual rating, so a 9 overall would have a much higher salary than a 1 overall. Same with experience, a veteran would have a higher salary than a rookie. I'll have a think on how to scale the salaries better, I would agree that it's a bit low. It wasn't an issue in previous versions, but seems to be more of a deal here.

I think that makes sense, too. I changed salaries in my file, but really all I did was make it so the salary cap was 9.7M instead of 100M, and I scaled everyone's first year salary down so the Steelers were right at the new cap, so it's effectively the same as yours, in terms of what percent of the cap each player is taking up. Then I increased the salary aspect of future years of each contract by 10% but kept bonus the same. There's still an issue that the Giants are only spending like 50% of the cap, compared to 100% for the Steelers, but that's not far removed from what the game is doing on its own. I've finished FA in 1978 and just about to start the season, and there were plenty of good FAs available, but 6 teams still have used less than 50% of the cap, and almost everyone is under 80% of the cap. The Steelers are over, and have lost 3rd round picks due to violations in 1978 and 1979, just signed 20 players at the league minimum to fill their roster the day before training camp, and are STILL over, and the Rams are also over it, too, after losing a 3rd pick for 1979 and signing 15 players at the league minimum, so the cap is affecting some teams at least. I think it's working fine -- the Steelers have a great team, and are having to deal with the trouble of keeping it together, and the Giants don't, but have a lot of room to play with and build it up, and it will all even out over time.

[Passacaglia]

Originally Posted by Passacaglia

I made a change to the 28_6_16_schedule.csv file if anyone is interested. That file sets the league up to rotate 3 different schedules. However, one of those 3 schedules has all 16 games played over 18 weeks, instead of 17. That in itself is a little weird, with every team getting two bye weeks, but the 18th week was getting counted as a playoff game for record-keeping in the coach's profiles, so when I realized I could fix it, it had to go. I just took out the bad schedule, so now there's only two instead of three, but I don't think that will matter -- the schedules are based on division ranking from the previous year, so I don't think it will feel repetitive.

I'll host this with the 1976_coaches file when I'm done, but if you want to change it yourself, all you have to do is delete the rows with a "2" in the first column, then change the rows with a "3" in the first column so that they're a "2"

Never mind on this -- once I got to the 3rd season, there were no games. I'd also like to eliminate bye weeks, so I'm going to see if I can redo the whole file. Not sure why the third season is not working, from everything I read, if I just have two schedules, it should rotate between those two.

[nilodor]

Originally Posted by Passacaglia

That seems like a pretty good way to handle it. I don't think it's working right, though. I think I said before that four teams have everyone hit FA at the end of the year. I don't remember them all, but Houston was one, and if you take your 1976_players file and filter on team 5, you'll see that the salaryears field is 1 for all of them. The most notable single exception I see is Joe Theismann, who hits FA at the end of the year. The one time I've simmed to 1985, it looks like they franchised him, and the next year he signed with Minnesota. I'm not sure if that all is a huge deal, since you have to expect players to be on different teams from reality, but having the four teams with no players under contract seems like it will put them in trouble. I think some easy solutions are to either a) add a year to everyone's contract, or b) put a 2-year minimum on every contract. I've created a file where I added a year to everyone's contract, and it seems like that's working well.

Hmm, it seems to be the teams that changed their team number in the league structure file. I have a master database where each player has one line. There are a column for each season they played, with their team numbers in them. Anyways, I'll need to fix that.

Originally Posted by Passacaglia

I think that makes sense, too. I changed salaries in my file, but really all I did was make it so the salary cap was 9.7M instead of 100M, and I scaled everyone's first year salary down so the Steelers were right at the new cap, so it's effectively the same as yours, in terms of what percent of the cap each player is taking up. Then I increased the salary aspect of future years of each contract by 10% but kept bonus the same. There's still an issue that the Giants are only spending like 50% of the cap, compared to 100% for the Steelers, but that's not far removed from what the game is doing on its own. I've finished FA in 1978 and just about to start the season, and there were plenty of good FAs available, but 6 teams still have used less than 50% of the cap, and almost everyone is under 80% of the cap. The Steelers are over, and have lost 3rd round picks due to violations in 1978 and 1979, just signed 20 players at the league minimum to fill their roster the day before training camp, and are STILL over, and the Rams are also over it, too, after losing a 3rd pick for 1979 and signing 15 players at the league minimum, so the cap is affecting some teams at least. I think it's working fine -- the Steelers have a great team, and are having to deal with the trouble of keeping it together, and the Giants don't, but have a lot of room to play with and build it up, and it will all even out over time.

I might scale things a bit differently. I think it's one of the 90's 49ers teams that had the most talent ever. Does the salary cap work well using realistic numbers for the year? It didn't in FOF8 because the increases were by dollar amount so the first increase basically doubled the salary cap, if FOF9 by percentage?

[Passacaglia]

Originally Posted by nilodor

Hmm, it seems to be the teams that changed their team number in the league structure file. I have a master database where each player has one line. There are a column for each season they played, with their team numbers in them. Anyways, I'll need to fix that.

That makes sense. I think I saw the team numbers are different in one of the league files, depending on the league format. Hopefully not too difficult of a fix.

I might scale things a bit differently. I think it's one of the 90's 49ers teams that had the most talent ever. Does the salary cap work well using realistic numbers for the year? It didn't in FOF8 because the increases were by dollar amount so the first increase basically doubled the salary cap, if FOF9 by percentage?

FOF9 is still based on dollar amount, but you can change it in the settings, and it appears you can change it as you progress through the career. I set the min at 0 and the max at 1, so I get an increase from 0 to $1M the first year, and my plan is to increase those numbers over time, keeping them at around 5-10% of the cap. That seems to work fine, but the drawback is you have to remember to change it every year. I think you'd have that issue with a $100M cap after a few years into the future anyway, because if you don't change it, your cap increase is too small.

This seems like an area where LESS customization would be preferred in the game. Just set the minimum at 5% and the maximum at 10%, and call it a day. But if the game does want to allow customization, it should let the user customize the percentages rather than the dollars.

I see what you're saying about the 90s 49ers teams, but I think given how some teams are so far below the cap, it's probably best to scale it so the team with the most talent THAT YEAR is at the cap, rather than only the most talented team ever. Maybe even a little over, so they have to get rid of some players. I mean, if the 76 Steelers are 90% as good as the 94 49ers, and the 76 Steelers are at 90% of the cap because of that, it just brings the 76 Giants down from 50% of the cap to 45% of the cap. I dunno. It's probably very subjective depending on how people want to play, and maybe it's more fitting to a pre-salary-cap NFL for a team to have a ton of talent and still not have to worry about the cap.

[Passacaglia]

Here's my 1976 coaches file.

Dropbox - 1976_coaches.csv - Simplify your life

This should have pretty much all the correct head coaches and coordinators. The only exception I could think of is that the Lions DC, Jim Carr, didn't have a wikipedia page, and since I couldn't find his birth date, city, or college immediately, I put Jerry Glanville there instead. That gave me room to put Joe Bugel in as an assistant coach. It did NOT leave me room for Bill Belichick, who was a special teams coach. I included him in the game, but as unemployed. He's by far the youngest coach in the file, at 24 years old.

Some other "exceptions" where that a lot of teams didn't have coordinators, just position coaches. In that case, I chose the coach who would soonest become a head coach or a "real" coordinator, and put him there.

If there were more assistants than I could fit that would become head coaches in the future, I took the one that would become a head coach furthest off, and put him in the game as unemployed. For example, Dallas had Tom Landry as HC, Jim Myers as OC, Ernie Stautner as DC, and Dan Reeves and Mike Ditka as assistant coaches. After adding in the HC, OC, and DC, I only had room for one assistant, so I put Reeves in that position, since he would become a head coach before Mike Ditka would. But Ditka is in as unemployed. I *think* I added an occasional assistant in the game as unemployed if he was destined to become a coordinator, but not sure.

Next, I added any coaches who would become a HC by 1985 -- most of these guys were coaching in college in 1976.

That only gets 20-25 unemployed coaches, so the game is filling a lot of people on its own. Not sure what to do about that. I'm worried about adding coaches that will become a HC in the late 80s or 90s, because I'm not sure we want a situation where they become coaches like way earlier than they should. Maybe seeing what happens to Belichick will be a good test for that -- the youngest coach that the game generates is 32, and I don't know if the fact that he's 8 years younger will make him not as good yet, or will mean that when he's 32, he's had 8 years to develop. I could add coaches that will become an OC or DC by 1985, but I'm not sure if that's all that interesting to anyone.

If anyone sees issues with the file or ways to improve it, let me know. The playstyles were mostly the result of chatting with GPT, so they're very quick and dirty. Maybe it'd be better to have more Smashmouth coaches based on the era, and how it seems when looking at it from present day, but I don't think that's accounted for at this time. I do think there's only like 1-2 spread coaches, so that should reflect the era pretty well.

[KingZal]

I really am curious to see how Belichick develops under those conditions. I haven't simulated far enough to see how coaches fare over the long run, or how much variance there is in their development, but it would be cool to see him become a head coach in the early 90's. My guess, though, is that he would become a head coach much sooner or never reach that rank at all.

[Passacaglia]

Dropbox - 1976 new schedule.zip - Simplify your life

For those interested, I created a 28_6_14 schedule file. I didn't like how the 28_6_16 file had a bye week, and as I thought of ways to improve it, I decided I also didn't like how the NFL schedules anyway. Here's the highlights of my 28_6_14:

1. I only have one preseason game. I wanted to get rid of all of them, but whenever I had none, the dates got all messed up. It's all AFC vs. NFC based on standings from the previous year. AFC East #1 plays NFC East #1, AFC Central #1 plays NFC Central #1, AFC West #1 plays NFC West #1, and so on. AFC West #5 plays NFC Central #5. One year is all AFC teams hosting, next year is all NFC teams hosting.

2. Weeks 2-14 are a round-robin within each conference. This means that teams don't play their division rivals twice in a season like usual. Division games are still important for tiebreakers and essentially a double effect on the standings, plus they're all pushed to the end as much as possible, which should make them more interesting (and strong division rivalries are at the very end). It also means there's less chance of a rematch in the Super Bowl (unless it's from Week 1).

3. Week 1 is an AFC-NFC rivalry week. Some of these rivalries make sense and seem like a lot of fun, some don't and are just created to fill the schedule.
Baltimore Colts-Washington Redskins
Philadelphia Eagles-Pittsburgh Steelers
Cleveland Browns-Detroit Lions
New England Patriots-Green Bay Packers
Tampa Bay Buccaneers-Miami Dolphins
St. Louis Cardinals-Kansas City Chiefs
Chicago Bears-Cincinnati Bengals
Seattle Seahawks-New Orleans Saints
New York Jets-New York Giants
Oakland Raiders-San Francisco 49ers
Buffalo Bills-Atlanta Falcons
Dallas Cowboys-Houston Oilers
Los Angeles Rams-San Diego Chargers
Minnesota Vikings-Denver Broncos

4. The rotation is pretty simple, everyone plays the same team each week, flipping home and away. I may make it so that it gets mixed up every so often, but this was kind of a pain, so it might be a while. Also, Baltimore plays 4 away games in a row, Kansas City plays 3 home games in a row, and there may be others. I tried to make it so a team never played home or away more than twice in a row, but gave up. I might try fixing that later, too. It's something that looks annoying, but probably doesn't matter in terms of FOF gameplay.

5. Only 4 teams in each conference make the playoffs -- the three division champs and one wild card. If you want the 10, it's easy to change that.

There's also a default_teams and a league_info file included. They're both needed to get the 28_6_14 schedule to work. You'll also need to edit your xxxx_info file and add 28_6_14 to the schedule field. I called the zip file a 1976 schedule, but I think you can use it any year.

Just a note -- The game crashes a lot when I use these files whenever I try to sim through a whole season with one click. I would suggest saving often and simming one day at a time using the check box on the sidebar as much as possible.

[Passacaglia]

Dropbox - 1976_players_empty_expansion.zip - Simplify your life

One more set -- I also made 1976_players and 1976_quarterbacks files where the expansion teams (Seattle and Tampa Bay) have no players. Other teams will start picking up those players in FA before you get a chance to do anything if you start as one of these teams, including the expansion team you DON'T start as, so your team will be the only empty one. You'll be really bad in 1976. This file uses the cap changes that I mentioned earlier -- it's a $9.7M cap and player salaries are adjusted accordingly, plus contracts from nilodor's file were all extended by a year. There does end up being a pretty decent FA market though, since the AI releases players to stay under the cap.