FOF7 July 30th Developer Chat Log

[Pyser]

i think quik nailed the draft discussion. the only way late round picks will be interesting is if:

1) every pick from the 3rd or 4th round on is minsal (realism be damned), and
2) a SIGNIFICANTLY bigger chance of late round picks booming

the automatic signing of rookies is a double edged sword. it takes away a seriously tedious part of the game, but as quik smartly points out, if the rookies arent the cheapest option to fill out your roster, no one will bother

also, ben's completely right that no player should sign a 3 year minsal contract. eliminating that option would be a gigantic game changer in mp.

[aston217]

The problem as I see it is that free agency is far too reliable. The draft is the lifeblood of a NFL team, but in FOF there is always an available pool of quality players you can draw from in free agency. ALWAYS. In SP, even more so.

What's the point of relying on your draft when you can sign all these moderately older players in FA and do just fine?

Rather than increase the chance for late round picks booming (how big can that get before it gets to be a problem?), I think there needs to be a much greater washout rate. Sure, there might be a number of guys that all teams know about and pursue in FA, but by the time the regular season rolls around, what's left is fodder, not stacks and stacks of capable leftovers.

If you get hit with a midseason injury, you're probably calling up an undrafted rookie, not signing some high 30s rated 5th year player who just happens to be sitting around along with two dozen of his like-rated buddies.

(None of this is to say that teams that sign FAs win; they don't. But teams can sign FAs to replace the bottom half of their draft, and fill the bottom half of their roster. Skimp on picks and you should land in serious, serious, trouble -- so nobody would be sane enough to do it.)

[sjshaw]

Originally Posted by Solecismic

I like PT in theory, but I think the best overall design, given what I've studied, is to remove it. Again, the question is whether I'll get enough in for FOF7 to justify that removal.

I agree, I just hope you give us a way to have starters pulled from blowouts (and not return). I've had numerous players injured in situations where IRL they would be out of shoulder pads and helmets.

[QuikSand]

As we talk about free agency providing too many players, and limiting the value of rookies... There's a departure from realism at work in FOF here, maybe part of a solution.

Lots of marginal players (or worse) sit out a whole season unrostered, but then show up in the FA pool basically none the worse for wear. Way too many of these guys, compared to the NFL, I'm sure.

Having a ton of cheap 24/24 guys who played a few games a couple of seasons ago, have pretty good positional exp, and can shake off all their rust in a couple weeks of practice is part of the reason I don't really mess around with too many of these 19/42 fools gold rookies in the draft.

[aston217]

Great point, Quik. The attrition rate of these guys, I think, should be far greater. Perhaps Jim, being the numbers guy that he is, knows the answer to this question: what % of players who have entered the league but don't spent time in an NFL training camp in a certain year, are still 'around' in any capacity the next camp cycle?

[MartinD]

Originally Posted by QuikSand

Lots of marginal players (or worse) sit out a whole season unrostered, but then show up in the FA pool basically none the worse for wear. Way too many of these guys, compared to the NFL, I'm sure.

This sparked a thought that's effectively the opposite of this point - in RL, you get the occasional star player who retires relatively early, then decides to try to come back a year or two later (Deion Sanders, for example - came back after a couple of years out of the game to play for the Ravens in 2004 and 2005)...

I agree with the point that Quik is making, though - it does seem like there are a lot of veterans on the free agent list who aren't good enough to make a roster but still have a bit of ability. (It's pretty common for me to see at least a couple of guys who were on my team a season or two before but were cut due to dropping off in TC or being beaten out by younger/better players.)

[QuikSand]

It's possible that the best way to insert something in-game along these lines would be for the training camp routines to involve a meaningful deterioration in skills for any player who isn't taken in to camp by any team.

Even if that marginal 4th year 24/24 guy is going to stick around, he at least ought to suffer the equivalent of gaining 20 pounds by being out of the game. Whether that turns him into 18/18 or 14/24 would be a developer decision, but I think that might be a good direction.

[Raiders Army]

Originally Posted by Solecismic Website

While the timeline of this new product is still uncertain, especially since we don't know the method of publication, enough has been done to say it's possible this product will be released in November.

FWIW, I'd be glad if it's released next year. I'm happy for any news whatsoever. I have the utmost of confidence and trust that you will make a great product. Thanks for doing the chat and working on FOF.

[aston217]

I like the idea of a smaller free agent pool overall. There's a lot of opening for clever GMs to sift through a very large and hard to sort through pool of FA players for the diamonds in the rough that are nonetheless 24-rated, while less experienced GMs are, in comparison, completely fumbling in the dark. This divide is too significant in MP, and of course as long as it exists would render the player too powerful in SP.

[Ben E Lou]

I'm not sure we even need "smaller." Create a new game and look at the FA pool. It's much, much, MUCH less talented than what we see in SP and MP after 5-10 seasons. In a "mature" league, there are FAR more guys rated in the 25/25 to 40/40 range than there are at the beginning of a career. I'm starting to wonder if it's more an issue of player creation and/or development than anything else.

[aston217]

That's an interesting thought. Maybe the game does have it all worked out to begin with!...

But I don't think it's a creation/development issue, because the reasons for this should be simpler. In SP, it only takes 2 years or so and it's an AI cap handling issue. In MP, owners don't pay enough attention and sign to fill up their teams for the offseason/preseason.

I think AI/GM not valuing players correctly is a big part of it. Hanging on to 19/40 rookies who just went -5 in camp, or 35/35 players who went -5 in their rookie camps and are headed even further down. This mis-valuation also results in useful players being left in the FA pool.

I think the FA pool being so large is also a problem for this reason: there exist non-obvious useful players because that's how the game works, but they're inordinately difficult to see. The FA pool is littered with 40/40 veterans who are massive droppers. Completely hidden is the 15/32 guy who actually went +2 in camp, and who is creeping. The GMs who pay enough attention to sign guys like this usually do so completely without competition. And of course, the "24/24 guy with all the right bars who isn't dropping", hiding behind six or seven worthless turds.

I don't know if even the reduction of veteran FA pool size will help, though, because even among just rookies you run into this issue. And it's really a consequence of a player's entire career track being knowable as the result of one camp (or if in doubt, after a cut test, in most cases). If all we had were an extremely limited pool of veterans, and a large pool of rookies who might (somewhat randomly, somewhat meaningfully) trend in either direction, I think that would be better.

Might also help substantially and be a simpler implementation if, regardless of veteran rating, there existed a random chance of "working out" with a team or not, and if the guy doesn't, he will be play at a huge ratings penalty - but you won't know it except by looking at his stats - and he'll almost never 'work out' with the same team again, even if they resign him later.

[Ben E Lou]

Originally Posted by aston217

But I don't think it's a creation/development issue, because the reasons for this should be simpler. In SP, it only takes 2 years or so and it's an AI cap handling issue. In MP, owners don't pay enough attention and sign to fill up their teams for the offseason/preseason.

If you think this, then you need to actually take a look at the player pools that I'm talking about, 'cause you're dead wrong here. It happens that I have a year 1 league stored on my local MySQL database.

select * from playerratings where cur>29

Result??? The league has 860 players rated 30 or higher. 26.9 per team.

Run that same query on a few mature leagues, all using game-generated draft classes, and what does it look like?

IHOF: 1378
USFL: 1408
CCFL: 1381
SP League that is 8 years old: 1325
GML (only 5 years old): 1227

The lower numbers in the GML and my SP league raise even stronger suspicion that this is something that ramps up over time. It's not an issue of the AI not handling rosters or people not signing players--though both things happen. It's quite simple:

Leagues in FOF2K7 start out with ~27 players per team rated 30 or higher. Over time, there are ~44 players per team rated 30 or higher.

That's a significant enough increase in marginal players to create exactly the environment that Quik and I have noted and used to our great advantage.

I'll create another new league or two and run the numbers there to get something concrete before sending this one on to Jim, but with 860 vs. ~1400, it's pretty obvious that something is going on here.

[Ben E Lou]

Newly-created default NFL league with X-Factor turned off: 882.

So yeah, I suspect that this is a huge part of it.

[aston217]

Oh, my bad. I thought you were talking only about the FA pools, for which there seemed to exist a simple explanation. Overall league talent, I see, and you certainly have the numbers to back that up. And that is a huge difference, I agree.

Given the timespan, it's almost as if the talent of a typical draft class doesn't match up (i.e, it's superior) to starting conditions, which is the creation issue you say. I'm curious as to what the BFL numbers show -- and if you do more testing, perhaps use your draft class generator to turn down the talent even further, and see if over enough time matches the IHOF numbers. To see if there's a development issue here as well.

These numbers are truly surprising Ben, thanks for sharing. It makes me wonder about something else: cohesion. I remember my first introduction to cohesion was the argument that leagues do not tend to start out with powerful passing attacks, but over time, as cohesion takes effect, the power of the strong pass attack starts to show itself. Given that WRs tend to outmatch CBs of a similar ability -- and correct me if I'm wrong in this assumption -- what if this tremendous increase in talent explains it, and cohesion isn't as meaningful as thought?

[Ben E Lou]

Well, I haven't checked the high end of the talent pool lately, but I have in the past, and it does tend to match up rather well with the way leagues are created originally. I suspect that this is just isolated to talent that's mediocre or worse.

The "cohesion" answer, fwiw, originally came from a Q&A with Jim, not from me, so I'd tend to believe it's true.

[scorp]

One question I have or thought is if we don't have interviews in the new version. a possibility is the scouting going to be a little more realistic vs hole rec on a RB or route running on a WR?
If a WR can run a route no scout will have them as a very good route runner, he may feel the play could develop but he won't say he good at it.

Same for a RB and hole rec, it's pretty obvious if he has that ability or not, you shouldn't need to know the player intelligence to know the scouting was way off in those 2 cases.

There should be scouting error, and we should get fooled at times, but some things scouts know how good the rookies are at present, just how good they might become is the grey area in several areas.

maybe show the Route running as low but has a high ceiling, that ceiling could be way off but the rookies current route running should be close to starting skill level it just may not grow.

[Sef0r]

Originally Posted by scorp

Same for a RB and hole rec, it's pretty obvious if he has that ability or not, you shouldn't need to know the player intelligence to know the scouting was way off in those 2 cases.

There has to be some type of association, I agree that I would prefer not having SOL meaning they will have a high HolRec bar. He either associates it with a combine result or makes it random.

[A-Husker-4-Life]

Serious injuries to qb's when they hand the ball off, I've had it happen numerous times and it could be something to look at.

[Ben E Lou]

Originally Posted by A-Husker-4-Life

Serious injuries to qb's when they hand the ball off, I've had it happen numerous times and it could be something it look at.

I strongly suspect that this is a "bigger-picture" thing in that injuries are just random events that don't have anything to do with the specific play that's occurring. As best as I can tell, there's an x% chance that an injury will occur to player y on any given play. It's the sort of implementation that causes the big picture (overall injury frequency and duration) to work out correctly, but creates some situation-specific anomalies.

[Passacaglia]

Whenever that happens, I just pretend the QB really got hurt by banging his helmet against the wall of the stadium.

[corbes]

How would a football sim respond to the new packaged-play concepts? I've been trying to imagine how a game designer would respond to that and incorporate it within a play-calling scheme.

[conception]

I don't see how having talent in the FA pool doesn't mimic reality. Truth be told every year there are lots of potentially effective guys watching at home on sunday because teams take chances on draft picks more often than bringing in one year rentals.

Are Willis McGahee and Michael Turner incapable of helping a team this year? Half the guys who make NFL rosters will perform worse in game action than these two guys could, but NFL teams would rather take a chance on future potential behind their rather than an average veteran. Why have a guy on the roster for a year when another guy may perform as well this year but continue performing the next 3-4 years or even longer.

[redfox000]

Originally Posted by conception

I don't see how having talent in the FA pool doesn't mimic reality. Truth be told every year there are lots of potentially effective guys watching at home on sunday because teams take chances on draft picks more often than bringing in one year rentals.

Are Willis McGahee and Michael Turner incapable of helping a team this year? Half the guys who make NFL rosters will perform worse in game action than these two guys could, but NFL teams would rather take a chance on future potential behind their rather than an average veteran. Why have a guy on the roster for a year when another guy may perform as well this year but continue performing the next 3-4 years or even longer.

It probably also has to do with salary as well as any extra attention/drama the veterans may add?

[Ben E Lou]

Originally Posted by conception

I don't see how having talent in the FA pool doesn't mimic reality. Truth be told every year there are lots of potentially effective guys watching at home on sunday because teams take chances on draft picks more often than bringing in one year rentals.

Are Willis McGahee and Michael Turner incapable of helping a team this year? Half the guys who make NFL rosters will perform worse in game action than these two guys could, but NFL teams would rather take a chance on future potential behind their rather than an average veteran. Why have a guy on the roster for a year when another guy may perform as well this year but continue performing the next 3-4 years or even longer.

That's all fine, but I suspect that this is where you run into an issue of "what makes a good computer game" conflicting with perfect realism. And the truth of the matter is that based on how the initial talent pools are created--both with the default (NFL) player file and a game-generated fictional file--FOF is designed to work best when there are 850-900 (I checked a few more newly-created leagues just to be sure) players rated around 30 or better in the league file. Some of them may be Free Agents. That's fine. But it should be obvious that if a game file has 850-900 players rated in that range to start with, but in every single league checked, that number jumps to close to 1400 after 10ish seasons, it's going to significantly devalue later-round draft choices. (Why bother with a fifth round pick who *might* make it to 30 or 35 in three seasons when there are already plenty of guys rated 30 to 35 to fill your roster?)

[QuikSand]

Originally Posted by conception

Are Willis McGahee and Michael Turner incapable of helping a team this year?

I don't think anyone objects to the notion that a successful NFL player remaining in the free agent pool in the year following a starting gig is a stretch. That's not where I or others find fault with the FA system.

Let's play it forward, though. If Turner can't get a contract this year or next, should he still be sitting as a seemingly productive option in September of 2015? In FOF terms, maybe he used to be a 65/65 guy in 2010. In 2012 he slipped to 45/45. Now maybe he's 36/36. In two years, should be still be 32/32 and looking like a semi-viable option as a fill-in?

Flip it around. A couple of seasons ago, guys like Derrick Ward and Steve Slaton were marginal NFL running backs. They couldn't get work in 2012, and now (after missing a full season and two full camps) they simply don't exist in the relevant NFL universe - they are high school assistant coaches or A&P stock clerks or bank vice presidents or whatever. In FOF, these guys would still be puttering around free agency, rated 22/22, and only a week or two on the roster away from being productive. That's a stretch.

[BowTieSports]

Originally Posted by QuikSand

A couple of seasons ago, guys like Derrick Ward and Steve Slaton were marginal NFL running backs. They couldn't get work in 2012, and now (after missing a full season and two full camps) they simply don't exist in the relevant NFL universe - they are high school assistant coaches or A&P stock clerks or bank vice presidents or whatever. In FOF, these guys would still be puttering around free agency, rated 22/22, and only a week or two on the roster away from being productive. That's a stretch.

Forgot my old log-in info, so first time posting after a long layoff. Anyway.... I agree with Quick.

Just a thought, but to some degree can this be solved by adding a "second" retirement period into the game. As is, players only officially retire on the switch-over between seasons. Add in a second period of retirements after Week 2 or Week 3 of the preseason -- make it an "automatic rule" based on player age ... any player over 28 (or 30 or 32) who was not on a roster the season before and was not on a training camp roster for the current season is automatically retired.

It wouldn't totally address the increase in players rated 30+, but it would cut down the pool to some extent and it would reduce the veteran free agents available for in-season injury replacements.

[Caratacus]

GOG might be a good alternative to Steam and might be worth a look. GOG.com

[A-Husker-4-Life]

Originally Posted by A-Husker-4-Life

Serious injuries to qb's when they hand the ball off, I've had it happen numerous times and it could be something to look at.

Just had this happen again in my MP League, that's twice this season.. Kinda crazy, hope Jim will look at it.

[Dutch]

Originally Posted by Caratacus

GOG might be a good alternative to Steam and might be worth a look. GOG.com

Good call.

[redfox000]

Originally Posted by A-Husker-4-Life

Just had this happen again in my MP League, that's twice this season.. Kinda crazy, hope Jim will look at it.

It's probably similar to other card-football games. On each play there is a percent chance of injury and if it occurs then a percent chance for each position. Then when a player comes up it probably is a dice roll based on hidden player ratings as to whether or not the player actually gets injured or not. So, the point is, all of this is irrelevant of pass/run.

[aston217]

Here's a question related to both the MP & SP experience of the game.

Contracts in FOF2k7 are open-ended, an environment that naturally calls for either 'house' or 'league' rules, depending on the league. Will FOF SP players and commissioners be equipped with better tools to enforce artificial, custom contract rules as they like?

This would be an extremely sensible customization thing to put in the game (along with things like adjusting salary cap automatically) -- and it would greatly lower the design burden of trying to create a catch-all financial system that is balanced and unbreakable. The community will figure out the best rules eventually; they should be have the tools to enforce it.

[Kozure]

I'll add to this. There needs to be more than one stage for the franchise designation. Too many GMs forget about "tagging" players.

I'm not sure how Jim is going to handle the offseason stage order. But assuming that each offseason begins with staff hiring, allowing us to franchise the player all through staff hiring would go a long way to ensuring that players get tagged.

[Passacaglia]

Originally Posted by Ben E Lou

[19:49] SOLECISMIC: REQUIREMENTS/DISTRIBUTION/RELEASE/ETC
[19:49] SOLECISMIC: Q: What are the system requirements for FOF7?
[19:50] SOLECISMIC: Windows XP or later. A screen with a resolution of 1024x768 or above.

So this is probably a dumb question, but -- that includes Windows 8, right? Is there any reason it wouldn't?

[Dutch]

If you are still concerned about getting FOF7 on Steam because of lack of graphics...

Save 15% on Super Amazing Wagon Adventure on Steam

I think you should still pursue this.

[NawlinsFan]

Originally Posted by Mike D

If you are still concerned about getting FOF7 on Steam because of lack of graphics...

Save 15% on Super Amazing Wagon Adventure on Steam

I think you should still pursue this.

Agreed Mike. There are a number of products on Steam that are so not graphically appealing. At the end of the day I think it will be the content and playability that will make the difference.

[aston217]

Passc, yeah, that would include Windows 8. It'd be very silly to release something that supports Win7 but not Win8 at this point.

I had a thought yesterday, that while this game isn't online, what if we made it possible to submit exports entirely from online?

Jim wouldn't need to write the tools for this, only to make it possible. If it is, you could have people doing contract offers or changing gameplans and then uploading right from their phones, or without opening up the game.

[MIJB#19]

[19:44] SOLECISMIC: 23. End of Game Time Management. Some fine-tuning, if possible. It got a lot better in the last patch, but it's still limited by a lack of knowing what happened on the last play.

I noticed this element in the chat log, but it doesn't specify whether this also applies to the end of the first half. I think I see a lot of time mismanagement and horrible time out calling in the final minutes of the first half (nearly concluding that the teams are reading the scoreboard in reverse). In my experience it's a lot worse at the end of the first half, than at the end of the game.

[BowTieSports]

I know FOF7 is still new, but has anyone run tests on player ratings for more mature leagues in the new game? Wondering if the player development/player retirement tweaks have limited the explosion of 30+ rated players in FOF7.

Originally Posted by Ben E Lou

If you think this, then you need to actually take a look at the player pools that I'm talking about, 'cause you're dead wrong here. It happens that I have a year 1 league stored on my local MySQL database.

select * from playerratings where cur>29

Result??? The league has 860 players rated 30 or higher. 26.9 per team.

Run that same query on a few mature leagues, all using game-generated draft classes, and what does it look like?

IHOF: 1378
USFL: 1408
CCFL: 1381
SP League that is 8 years old: 1325
GML (only 5 years old): 1227

The lower numbers in the GML and my SP league raise even stronger suspicion that this is something that ramps up over time. It's not an issue of the AI not handling rosters or people not signing players--though both things happen. It's quite simple:

Leagues in FOF2K7 start out with ~27 players per team rated 30 or higher. Over time, there are ~44 players per team rated 30 or higher.

That's a significant enough increase in marginal players to create exactly the environment that Quik and I have noted and used to our great advantage.

I'll create another new league or two and run the numbers there to get something concrete before sending this one on to Jim, but with 860 vs. ~1400, it's pretty obvious that something is going on here.