BBCF to FOF draft file export utility ver 0.21 released.

[flounder]

I've been working on a utility to export BBCF players into an FOF draft file, and it's finally reached the point where I could use some user input. This is the first time I've written software for other people to use, so this could get pretty rough. I'm confident, though, that if we keep our wits about us we can get through this with minimal loss of life.

The utility gets the BBCF stats from the HTML files generated for the league website which means that the process is a bit convoluted. I've tried to make it as streamlined as possible, but it's still complex. Hopefully, it's not too much trouble.

You can find the utility at my website at http://www.drflounder.com/bbcf . Please send feedback to the email given at that site or just PM me here. I would welcome all comments about the program, especially about the program stability and the player distribution. Source code is included in the distribution so if any veteran coders want to poke through it and give me tips they would be greatly appreciated.

I would like to thank Cap Ologist, twothree, and nilodor for their help with the FOF draft file rating distribution and Richards for creating his Fictional Player factory that I used to compare my draft classes to. I'd also like to thank 21C for the SQL query that generates the draft class and Greg Stelmack for his Extractor utility I used to get ratings data back out of FOF. Most of all, I'd like to thank Jim and Arlie for writing two great games that hopefully will taste great together.

Installation

Just unzip the file into the directory of your choice. When prompted to in the instructions, double click on the BBCFDraftExport.exe file.

In order for the program to work, you will need Icy's Real College Pack (at least the pstats.dat file) available at
http://www.greydogsoftware.com/forum...ead.php?t=9935 . You will also need to make sure that the "Overall Rating Format" and the "Skills Rating Format" in BBCF are set to 1-100. You can change them in the Game Options screen.

Instructions

1. At some point during the regular season in BBCF (after the preseason, but before you click the "Advance Stage" button after the bowl games) you need to create a league website. Go under the options menu on the left in BBCF. Under "WEBSITE" click "Update team reports", then "Create website", and finally "Update player cards". Updating the player cards can take a while (~10 minutes). MAKE SURE THAT UNDER THE "GAME OPTIONS" SCREEN THAT BOTH OVERALL RATING FORMAT AND SKILLS RATING FORMAT ARE SET TO BE 1-100 BEFORE CREATING THE WEBSITE.

2. Proceed through the offseason until the underclassmen that are declaring for the draft are announced (Week 12 during recruiting). At that point, save your game.

3. Run the BBCFDraftExport.exe executable.

4. Click the button labelled "BBCF Save Directory". A dialog box showing the directories in your /GDS/Bowl Bound College Football/saves directory will be shown. They will be named according to the save name given in the BBCF Game options screen. Select the directory which corresponds to the save game you want to export.

5. Enter the year in the box below "Draft Year". This will be the year that just finished. So to export players whose senior season was 2005, enter 2005 in this box.

6. Click the "Export FOF Draft File" box. The program will ask you where you want to save the .csv file. The default is the FOF universe directory.

7. Watch the red bar move across the screen. When the "Creating Draft File" box disappears, the program is finished. Close it. On my computer, which is moderately out of date, this takes about 2-3 minutes.

8. Open the FOF Draft File Generator program that came with FOF2k4. (All of the default directory choices for this version are in the Front Office Football 2004 directory. Make sure you're choosing the right file and saving the result into the FOF2k7 directory.) Click "Generate Draft File". The program will ask you where you saved the .csv file. Navigate to it and select it. Save the .faf file generated.

9. In the FOF postseason, before free agency begins, there will be a button at the top of the screen labelled "Import Draft". Click it and find the .faf file generated in step 8.

10. Enjoy the juicy BBCF/FOF player goodness (hopefully).

Known Issues/Limitations

1. Player stats at this point are based entirely on the BBCF current rating. Getting the FOF stats to mirror the BBCF stats in a way which gives a good player distribution has proven to be a big challenge. This is what I'm devoting almost all of my development time to.

2. The ratings from BBCF are filtered through your scout's perception. Not having access to the real ratings in the save file, there's nothing I can do about that.

3. Right now there are no stats included in the draft file. I'll work on that once I get the player distribution ironed out.

4. pstats.dat files which are significantly different from Icy's Real Player File will not work. If minor modifications have been made (one team replaced by another) it may still work. You will need to modify the colleges.txt file included with BBCFExport in the following way. Open the FOF draft file generator program and click on the "College List" button. A list of colleges and their FOF ID will be given. Find the ID for your new college, or if it isn't in there, pick another college's ID. Open the colleges.txt file and add the following line to the end:

College ID , College Name exactly as it appears in the pstats.dat file

This should work, but I haven't tested it out yet. Consider custom pstats.dat file support to be highly experimental.

5. At this point it doesn't matter at what point during the season you create the league website as long as it's not during the offseason. If stats ever get included, you will have to create it right after the last week of bowls so all games' stats are included.

6. Florida Atlantic and Florida International don't seem to be in the FOF college list, so all graduates of those schools get an honorary transfer to Florida A&M.

7. Hometowns are gleaned from the team roster files. However, some players hometown is only listed as their state (NY instead of New York, NY). Right now, those players are assigned a hometown of 'Outside the US'. This will be fixed eventually.

8. Volatility is calculated as follows. There are three factors.

The first is based on Work Ethic/Discipline. Players in the top half of the draft (or so) with high ratings in Work Ethic and Discipline have a lower volatility. Players in the lower half of the draft with high ratings in Work Ethic and Discipline have a higher volatility. This is intended to represent slacker superstars in college having a greater chance of busting and hard working, but less talented players having a greater chance of blossoming.

The second is based on school prestige. Players from big schools have lower volatility than players from small schools. Since I don't have access to the current prestige values in the save game file, this is based on the school's starting prestige.

The third is a random factor. Underclassmen have a higher range of random values than seniors do.

Right now I think this system gives top players too high a volatility rating so I would welcome feedback on this.

9. Top players' Solecismic test scores are definitely too high. This is based on the BBCF intelligence rating and most of those ratings are high for top players. Other than disregarding the BBCF ratings, I don't know what to do about this.

10. This program can run while BBCF is running in the background, but I wouldn't try to actually use BBCF while this program is running. Database conficts could occur.

11. I haven't run into any problems in testing, but still don't use this program on any BBCF or FOF careers that you care about unless everything is backed up.

12. If the player cards were to update after the season was finished, but before you run the export utility, that would be very bad. I'm not sure what exactly triggers a player card update, so try to do the export as soon after the underclassmen declare as possible.

[Franklinnoble]

Wow.

I mean, wow...

I am so all over this.

[dubb93]

wait a minute...why the hell is flounder posting in this thread ,

I hope atleast someone gets that.

[Groundhog]

So, have people given this a shot yet? I'm very interested in this but don't (yet) own BBCF.

[flounder]

I've put an updated version up on the website to address an issue Tigerfan1 ran across. I've also made a small change that should allow people with custom pstats.dat files to use the exporter successfully.

0.11 Tigerfan1 identified an issue where if you created a league website for the first time after a career had been running for several years, some player cards were missing and the exporter crashed. Now, if a player card is missing, it is just skipped. Also, if a player's college is missing from the colleges.txt file, if you're using a custom pstats.dat file for example, a college id is generated at random from the list of small colleges. This should allow people with custom pstats.dat files to use the exporter successfully.

[Franklinnoble]

I've been meaning to, just haven't had time. I'm probably going to start a fresh career in both games for testing purposes.

[kingfc22]

Wow. This looks to be a winner

[Emiliano]

Great job flounder. I gotta try this one.

[Groundhog]

bump

I finally bought BBCF, and now this link is dead...

Anyone got it?

[Emiliano]

Originally Posted by Groundhog

bump

I finally bought BBCF, and now this link is dead...

Anyone got it?

Upping it right now for you...

RapidShare link

[Groundhog]

Originally Posted by Emiliano

Upping it right now for you...

RapidShare link

Cheers!

[flounder]

Yeah, sorry about the link. My webhosting went under and I'm in the process of switching over. If people are still interested, I'll have the link back up soon (hopefully).

To be honest, I didn't think anyone was using it. I thought it was just too complicated a process.

[MizzouRah]

Originally Posted by flounder

Yeah, sorry about the link. My webhosting went under and I'm in the process of switching over. If people are still interested, I'll have the link back up soon (hopefully).

To be honest, I didn't think anyone was using it. I thought it was just too complicated a process.

As soon as Arlie gets the final update out, I'm all over this as well.

Thanks for your time!!!

[cougarfreak]

Is there going to be a new version of BBCF this year?

[FBPro]

Originally Posted by cougarfreak

Is there going to be a new version of BBCF this year?

Arlie is updating it, NOT making a complete new version. I'm guessing it will be in the form of a patch so to speak.

[cougarfreak]

Originally Posted by FBPro

Arlie is updating it, NOT making a complete new version. I'm guessing it will be in the form of a patch so to speak.

I may have to check it out........I haven't bought yet. Is there a cost for the patch?

[astrosfan64]

Originally Posted by cougarfreak

I may have to check it out........I haven't bought yet. Is there a cost for the patch?

He is going to give away the patch for free. Look at it as a free expansion pack to the game. He is adding 3 major new features.

[MizzouRah]

Yep, then it's off to an NFL sim.

[Emiliano]

Originally Posted by flounder

Yeah, sorry about the link. My webhosting went under and I'm in the process of switching over. If people are still interested, I'll have the link back up soon (hopefully).

To be honest, I didn't think anyone was using it. I thought it was just too complicated a process.

To tell you the truth, I still have to try this thing, but if you have time please continue to improve it. It's a great idea (well, at least until Jim doesn't make a new college game or Arlie doesn't make a new pro game...).

Again, great job!

[Groundhog]

I just started up a BBCF game last nite and I definately intend on using it as soon as I'm done with the season. I'll post my results in this thread.

[Groundhog]

Well, I finally got the point of the game where I can try this out, and no luck.

I output the team and player pages at around week 9, then progressed through to the offseason in the same round that underclassmen declare, then saved the game, and ran the utility. When I enter the correct year (the year that shows up on the player cards for the season just played) and hit 'Export FOF Draft File' the 'Export Progress' box shows up, but doesn't do anything.

In the error log it says:

Exception in Tkinter callback
Traceback (most recent call last):
File "Tkinter.pyc", line 1345, in __call__
File "BBCFDraftExport.py", line 72, in draftexport
File "BBCFDraftExport.py", line 106, in getplayers
File "Player.pyc", line 85, in __init__
ValueError: invalid literal for int(): 2.5

[flounder]

Are your overall and skills rating formats in BBCF both set to 1-100?

[Groundhog]

Originally Posted by flounder

Are your overall and skills rating formats in BBCF both set to 1-100?

*slaps head*

[flounder]

If anyone is interested, I tested the utility out with the new public beta and QBs are way overrated. I'll need to do some rebalancing of the skill distributions before its usable. However, rather than take the chance on my effort being wasted I'll probably wait until the official release of the new version.

[Comey]

I am quite interested. I will likely start a new dynasty between BBCF/FOF when this is released. So please, let me know...my email is [email protected].

Thanks,
John

[JeffNights]

Originally Posted by Groundhog

*slaps head*

ha, nice reaction. made me acatully laugh.

[Groundhog]

Originally Posted by flounder

If anyone is interested, I tested the utility out with the new public beta and QBs are way overrated. I'll need to do some rebalancing of the skill distributions before its usable. However, rather than take the chance on my effort being wasted I'll probably wait until the official release of the new version.

Count me in as "interested".

[KJDelaney]

Any thoughts on an "upgrade" to be compatible with FOF 2007?

[flounder]

Assuming the draft import mechanism is the same as FOF2k4, it shouldn't be any problem to make it compatible. I'll probably still wait until the final release of the BBCF patch though.

[flounder]

I've been getting some PMs asking about the status of this. I still plan to do an update but I won't be able to get to it until after Thanksgiving. It shouldn't take more than about a week or so to do it once I get time. So look for it sometime in early December.

[flounder]

I'm not sure how much time I'm going to be able to devote to this over the holidays so I'm going to go ahead and release what I've done so far. I'm still tweaking the distribution, but it's getting pretty close.

The biggest problem I'm having is that in the BBCF careers I've run it looks like the talent level declines over the years. I'm not sure if it's just small sample size, but it makes it difficult to have a consistent talent level in the draft. Right now I think the first few draft classes have too much talent and after about 10 years they have too little talent.

Anyway, version 0.21 is compatible with FOF2k7 6.0d and BBCF version 1.51. You can find it here. Once again, if you use it please email or pm me and let me know if it's working and what you think.

Updates

0.21 Tweaked the FOF player distribution to match game generated draft files.

0.2 Compatible with FOF2k7 and BBCF version 1.51.

[Franklinnoble]

Any chance this still works for 2k4?

[flounder]

It will work with 2k4, but I don't know if the talent distribution will be ok. If you give it a try let me know how it works.

[MacroGuru]

Slight bump....has anyone tried this with a MP league at all? Or only homebrew?

[st.cronin]

Originally Posted by MacroGuru

Slight bump....has anyone tried this with a MP league at all? Or only homebrew?

I am planning on doing some testing later today, with the idea of using this in the bb-bbcf. I will be testing mainly for functionality.

[MacroGuru]

Ok cool...I just started running an Online MP league and we want to use TCY or BBCF for our draft classes....

Just wanted to know.

[QuikSand]

Just be forewarned... using custom drafts of any sort can lead to *serious* issues (problems) with a league's long-term balance of player talents. The consensus, I think, is that while the idea of linking draft classes from another game is really exciting in concept, it has yet to be productive in practice. You might find it to eventually be functional, but using this sort of file can have some pretty nasty long term consequences and can really undermine league integrity.

I have no personal stake in whether you do this for your league, or whether these various developers do or do not gain business. I'm just hoping you take this into consideration when making a major decision for your MP league.

[MacroGuru]

Originally Posted by QuikSand

Just be forewarned... using custom drafts of any sort can lead to *serious* issues (problems) with a league's long-term balance of player talents. The consensus, I think, is that while the idea of linking draft classes from another game is really exciting in concept, it has yet to be productive in practice. You might find it to eventually be functional, but using this sort of file can have some pretty nasty long term consequences and can really undermine league integrity.

I have no personal stake in whether you do this for your league, or whether these various developers do or do not gain business. I'm just hoping you take this into consideration when making a major decision for your MP league.

It is why I asked this question...and why I am going to do a massive test run with both ways, looking at every season statistically to see if it has a major blow up, negative impact.

[st.cronin]

It may be possible, if somewhat time-consuming, to manually edit the draft classes if the problem is simply one of balance.

[QuikSand]

Originally Posted by MacroGuru

It is why I asked this question...and why I am going to do a massive test run with both ways, looking at every season statistically to see if it has a major blow up, negative impact.

I think the issue is not so much that you'll see one oddity in a particular season, but rather that there could be a game-wide imbalance, as st. cronin intimates above For example, I think that using draft classes generated from TCY tends to yield way too much high quality talent. If you use that for ten drafts, you could end up with a league where there are so many quality veterans to go around at every position, there's not much value to, say, a guy rated 60/60. But all his in-game programming is telling him that 60/60 is a solid starter, and that he ought to be demanding big money for his services. Multiply this times 20 or 50 or 250 players across a league and you can end up with a game that starts to become one about things that aren't very interesting, rather than the cap challenges and young/old trade offs that make contemporary football roster management enjoyable.

In any event... surely do your own testing, and make your own call. Just putting it out there so you can do so wit your eyes open, as best I can. Best of luck.

[Ben E Lou]

Originally Posted by BBCF-->FOF Web Site

13. With the new version of BBCF, quarterback ratings are higher and there are more quarterbacks in the 90-95 range. This means that scout error of just a few points can move a QB from the good starter to superstar category. This could result in an overabundance of superstar QBs in FOF. IMO this is the most important issue to nail down moving forward.

This is a significant issue, and all signs point to it being not just superstar QBs, but G/VG one as well, which can really change the entire strategy, as Quik pointed out. I just generated a draft class with this utility. 14 QBs were adjusted-rated 6.0 or better, and there were three guys with all red combines except for 40-yard dash, and corresponding bars --three guaranteed superstars in other words. I then generated 6 consecutive classes with FOF: only one of them had as many as four QBs adjusted-rated 6.0 or better. If this sort of thing goes on for just a few seasons, the value of good QBs will go down significantly, since everyone will have access to one. This is the sort of game-changing (not in a good way) stuff that Quik is warning about.

FYI, I'd have some other major concerns about some other things as well, just at few glances:

• No QB ran faster than a 4.79. That's probably systemic as well, and will result in no QBs with mobility. I guess Buc would be happy with that. *shurg*
• Not *one* QB has a single green rating in strength, Sol, Agi, or Position drill. There will be a buttload of solid QBs running around the league.
• Every FB in the draft class was "green" in 40 time. Best time was 4.90. By comparison, in a game-generated draft class only six FBs were at 4.90 or worse, while 28 were better. I'm not 100% certain what skill the 40 times correspond to for FBs, but clearly that skill is going to be severely lacking.
• Other than FBs and QBs,every single player in the draft class had scores in the 40 and agility drills that were black, red, or blue--and the black scores were near the middle of the black range. For example, no DT had a slower agility drill than 7.80. That's on the fast side of the black range in game-generated classes.

Definitely imbalanced, and by a long shot.

[flounder]

Holy cow! People are still using this thing?

I think SkyDog's right, it's definitely imbalanced. It takes so long to generate one draft class that it was impossible for me to create a large enough data set to develop any real balance. I was hoping a lot of people would use it and would give enough feedback that I could fine tune it. However, no one that I know of has used it on a long term basis.

To be honest, I had pretty much given up development of this utility. If people are really interested in helping me hammer out the balance I would be happy to incorporate their feedback. Unfortunately I think this was a good idea in theory, but just ended up being too cumbersome to use in practice.

[Ben E Lou]

Originally Posted by flounder

It takes so long to generate one draft class that it was impossible for me to create a large enough data set to develop any real balance.

Yeah, going through the process led me to think that time is the major limitation of what you're trying to do. BBCF sims slowly when compared to FOF/OOTP/TCY/FBCB. From looking at time stamps, it took me between 30 and 45 minutes to do everything required to generate just one draft class, and the vast majority of that time was the process of creating the player pages and simming BBCF forward to the appropriate stage. That kind of time definitely prohibits any sort of long-term testing.

[Ben E Lou]

Oh, and to be clear, I applaud you for your effort, flounder. Getting it this far along with the testing limitations is quite impresssive, and I absolutely understand why you haven't been able to take it any farther.

[flounder]

Originally Posted by SkyDog

Oh, and to be clear, I applaud you for your effort, flounder. Getting it this far along with the testing limitations is quite impresssive, and I absolutely understand why you haven't been able to take it any farther.

I appreciate that. I've learned a lot about this game from your posts and those of other people, and I've gotten a lot of use out of the utilities other people have written. I started this as an attempt to give back to this community. It's a shame it looks like it's going to fall short, but if Jim has a hard time balancing incoming draft classes from TCY when he has access to the internals of both games, I think it was probably unlikely this would work from the start.

[st.cronin]

Originally Posted by flounder

Holy cow! People are still using this thing?

I think SkyDog's right, it's definitely imbalanced. It takes so long to generate one draft class that it was impossible for me to create a large enough data set to develop any real balance. I was hoping a lot of people would use it and would give enough feedback that I could fine tune it. However, no one that I know of has used it on a long term basis.

To be honest, I had pretty much given up development of this utility. If people are really interested in helping me hammer out the balance I would be happy to incorporate their feedback. Unfortunately I think this was a good idea in theory, but just ended up being too cumbersome to use in practice.

If you are interested, I have a couple of suggestions. I will send you a pm later tonight.

[st.cronin]

Here are the suggestions I sent dr. flounder, in case anybody else has input or is working on something similar.

Caveat: I don't even play FOF, just TCY. I do peruse the FOF threads, though, and have some ideas of how the game works. Also, I am no programmer - I have no idea how your program works, or what it does. But this is my macro view of how it should work:

Your program should have a generic or standard concept of what the skill distribution should look like for the average draft. Using QB as an example, the program should expect to have x number of qbs at skill level n, y number of qbs at skill level n-1, z number of qbs at skill level n-2, etc.

Now, instead of marrying specific bbcf ratings to specific fof ratings, the program could use the OVERALL rating of the qb, rate all qbs in the class, and match them to the curve. So if the program expects to have 1 qb at skill level n, the top rated qb would get skill level n. The shape of the fof ratings could then be tweaked to reflect the bbcf ratings - for example, if he's a qb with high arm strength and mobility, bump up his fof ratings in those categories, and downgrade somewhere else. Now, to determine "is this a strong or weak draft for qbs," I think the simplest method is simply to count the total number of starts for the qbs in the class. If its a high number, say, more than 10 per qb (not sure if that's a high number or not), then its a strong draft for qbs, and more qbs could float to the top of the fof skill ratings. If its a low number, vice versa.

In addition to overall rating, the program should also take into account the number of games started (possibly weighting more heavily the number of games started as a senior), as well as how close to "ideal" size the player is. For example, a RB 5-7, 175 lbs who is very highly rated in bbcf would probably not be highly rated in fof - he would be like that Wisconsin back from a couple seasons ago, Brian Calhoun, who was a borderline Heisman candidate in college, and a 3rd round draft pick who has done nothing in the pros.

I don't know if this is helpful or not - I think your utility is a wonderful idea.

[lou4gehrig]

I am interested in using this export utility. However, everytime I try to run it I get the "stats.dat not found or corrupt error". Anybody have any ideas why?

I'm using v1.52 of BBCF and have tried it on Vista and XP. I also used the filepack thing the utility tells me to use.

[brandocamandoSM]

I am going to bring this baby back to life and have a question .. I would like to use this for a league I am starting on FoF but the problem I am coming up with is not having FOF Draft File Generator program that came with FOF2k4. I never had FOF2k4 so I dont have that program, is there any way to do this without the program?

[tripps04]

Just D/L FOF 2k4 and it's there you don't have to buy it comes with the D/L