FOF4 House Rule?

Bee · 05-13-2003, 09:25 AM

After some recent discussions in the Group Think Dynasty, I started thinking about house rules...

There seems to be a consensus that one of the biggest flaws in the game is the fact that it's too easy to spot breakout/busts. What if we played a career similar to the "Misfit Toys" house rule, but instead limit our players to Free Agents with more than 6 years experience and take out the breakout/bust factor? While it's pretty easy to find a few players with experience to fill in, do you think it might be challenging to fill an entire roster with older players? It could lead to salary cap issues quicker as well (since young players are generally much cheaper).

Comments?

Edit:

To clarify a little.
Start off with an empty cupboard.
Add only players in free agency with more than 6 years experience.
Give away all draft picks every year and only sign new players with more than 6 years experience.

QuikSand · 05-13-2003, 09:41 AM

Sounds fairly promising.

albionmoonlight · 05-13-2003, 09:55 AM

The challenge is there, but at some point, you are not really playing football. Not getting to draft or find rookie gems cuts out a large part of the game. I think that IMT and all of its permutations take out too many aspects of the game that I like for my taste. (Though they do present an entirely new challenge).

When I get my new computer, I have an idea for a challenge that will allow me to play all aspects of the game. It is called the Dan Snyder challenge.

Before I play the game, I come up with some various pushy and implusive owner demands. (i.e. "We need to open up the passing game--get the best QB available in FA, no matter the cost." "This team believes in loyality--keep our starting running back on the roster until he retires" "I need to make some money with this team--Make sure we are pulling in a profit of $X million within 3 years" "I don't like what our QB just said about us in the media--cut him," etc."

Every three seasons or so, I will pull a new demand out of the hat and will have to do what it says. That would allow me to play the game normally, but have challenges thrown in to keep it real. My hope is that the challenges seem realistic enough in the game universe and not just like some weird demands that I wrote down on paper.

JAG · 05-13-2003, 10:00 AM

Quote:

Originally posted by albionmoonlight
The challenge is there, but at some point, you are not really playing football. Not getting to draft or find rookie gems cuts out a large part of the game. I think that IMT and all of its permutations take out too many aspects of the game that I like for my taste. (Though they do present an entirely new challenge).

When I get my new computer, I have an idea for a challenge that will allow me to play all aspects of the game. It is called the Dan Snyder challenge.

Before I play the game, I come up with some various pushy and implusive owner demands. (i.e. "We need to open up the passing game--get the best QB available in FA, no matter the cost." "This team believes in loyality--keep our starting running back on the roster until he retires" "I need to make some money with this team--Make sure we are pulling in a profit of $X million within 3 years" "I don't like what our QB just said about us in the media--cut him," etc."

Every three seasons or so, I will pull a new demand out of the hat and will have to do what it says. That would allow me to play the game normally, but have challenges thrown in to keep it real. My hope is that the challenges seem realistic enough in the game universe and not just like some weird demands that I wrote down on paper.

That sounds fun, thanks for a good suggestion. To be a real Dan Snyder challenge though, it would have to be a new demand every year. "Get every bleepin free agent you can!" "Fire Norv!" "Fire Marty!" "Sign every bleeping restricted free agent you can!"

Jamie

QuikSand · 05-13-2003, 10:47 AM

Quote:

Originally posted by albionmoonlight
The challenge is there, but at some point, you are not really playing football. Not getting to draft or find rookie gems cuts out a large part of the game. I think that IMT and all of its permutations take out too many aspects of the game that I like for my taste. (Though they do present an entirely new challenge).

I agree with you - I'd much prefer to play a complete, realistic, and challenging game. However, with the game as out of balance as it is, it's necessary to make sacrifices if you want challenge. Somtimes it's eliminating major features of the game itself (like the draft), sometimes it's gong outside the box and imposing your own subsidiary rules and goals beyond what the game requires. It's the struggle that we all face, if we:

(a) are interested in getting a challenge from this game, and
(b) have opposable thumbs.

I'm assuming that there are a fair number of customers that don't fit into the two sub-classes above, and therefore are perfectly fine with the game as it is. I hope, for our sake, that our subset gets something out of the next and final game patch.

Bee · 05-13-2003, 11:27 AM

I agree it would be better to play the "complete" game, but that's so easy at this point there's little point.

I actually tried something similar to what you are looking at albion. I hope you can come up with ideas that are more challenging than mine. Here's a few that I used:

Replace the (position group - i.e. secondary, receivers, linebackers, etc)

Youth movement (cut or trade everyone over 9 years experience, bring in all rookies).

Defense wins Championships (only draft and sign defensive players).

Maybe those will give you some good ideas for your challenge.

primelord · 05-13-2003, 12:18 PM

Quote:

Originally posted by Bee
Defense wins Championships (only draft and sign defensive players).

I say we adopt that one in the GT project.

Marmel · 05-13-2003, 12:24 PM

Quote:

Originally posted by QuikSand
I agree with you - I'd much prefer to play a complete, realistic, and challenging game. However, with the game as out of balance as it is, it's necessary to make sacrifices if you want challenge. Somtimes it's eliminating major features of the game itself (like the draft), sometimes it's gong outside the box and imposing your own subsidiary rules and goals beyond what the game requires. It's the struggle that we all face, if we:

(a) are interested in getting a challenge from this game, and
(b) have opposable thumbs.

I'm assuming that there are a fair number of customers that don't fit into the two sub-classes above, and therefore are perfectly fine with the game as it is. I hope, for our sake, that our subset gets something out of the next and final game patch.

I lost a thumb in a freak sandbox incident as a child.

Franklinnoble · 05-13-2003, 12:26 PM

Even under "misfit toys" rules, it's not too hard to field a winning team.

I tried a modified version of these rules (read about it here ) where I drafted only QB's, RB's, and WR's - the logic being that in FOF4, all that really matters is these three positions. I won 4 titles in 11 years - and quit after I had won 3 in a row.

I suspect you could do as well with just a top-notch QB and a solid RB.

Anrhydeddu · 05-14-2003, 08:43 AM

Here are my thoughts.

I think what we are seeing here is the different between the baseball and the football economic models, and thus, the difference between OOTP and FOF. The single thing in baseball/OOTP that can make the AI competitive is disparities between payrolls. You have anywhere from a very small market/payroll team to a large market/payroll team, with enough interesting variations to where a team can spend more than what they believe they can make. Therefore, the acquisition of players through free agency, primarily, as well as through trades can greatly vary from team to team. What this results in for OOTP is the simple rule of having you run a small/medium market team (like I am doing with the 1960s Cardinals) having to compete against the big boys of the Braves, Tigers, Yankees et al. Give a team with good revenues and fan interest and they will go after the best free agents (esp. SP), preventing you from doing so either because of the bidding war or because you just don’t have any money to spend.

In football, you have none of that. All teams spend exactly the same. In other words, there are no small, medium or large markets and therefore financials become irrelevant at that level. What this means for FOF is that the only way for the AI to be competitive is to be able to judge each and every player accurately since we (us and the AI) are competing equally from the same pool. But as we have seen over and over, the AI does a poor job in doing that in FOF4. Therefore to make the game some semblance of competitiveness, we have to resort to arcane rules that are unrealistic, like giving us a smaller cap or not drafting or no free agents. In baseball/OOTP, we can do that because that is realistic, in football we should not but in FOF4, we are forced to.

I think it is becoming clear, at least among those that are considered the “best” or hardcore customers is that significant effort must be paid to how the AI judges, acquires and manages their players over a career. The reason is, imo, that is the heart of the competitiveness. Nothing else really matters.

My frustration with FOF4 comes from the fact that the game truly performs well at the micro level but is lousy at the macro level. Take a look at the dynasties, esp. the GroupThink. Do you see anyone closely examining the logs seeing if the OLB blitzes 37.5% of the time or do you seeing folks simming 4-8-16 games at a time? The details and accuracy of the game always have been good, Jim needs to believe that and we need to have faith in that. But for FOF4, instead of looking more inwards, the direction needed to be more outwards. That has been my criticism of FOF4 and why I felt it took a big step backwards from previous versions. Reading the ideas on house rules and frustrations of some of the best FOF4 customers in playing the game, I think that is becoming obvious.

QuikSand · 05-14-2003, 09:01 AM

Quote:

Originally posted by Anrhydeddu
My frustration with FOF4 comes from the fact that the game truly performs well at the micro level but is lousy at the macro level.

I don't necessarily disagree with you, but here is another possible conclusion from your arguments:

In FOF, the fact that the AI environment stinks is patently obvious, because there's no catch-all way to just add to the overall difficulty. In a game like, say, OOTP, there's always the "go with a smaller market" concept, that can pretty easily help you gloss over the fact that your computer-controled rivals aren't making great decisions.

The fact that football, by its nature, is a much more nuanced game than baseball just adds to this problem. There are many, many more ways to tell when the computer AI is making poor decisions - where in baseball, there are fewer things to screw up (and forming an algorithm to cover them up is certainly easier).

Not looking to take sides - I think a lot of what you'rte suggesting is probably accurate. But I think your rush to judge that all Jim cares about is creating a realistic one-year sim is flawed. I do think that Jim cares about the career elements - I just don't think they have been executed as well as they could/should be. I suspect it's more a matter of difficulty than of focus.

albionmoonlight · 05-14-2003, 09:30 AM

Another question (and this is from someone who has never played CM--so maybe the "holy grail" answers this question) is whether such a robust AI is possible.

In order for the game to be fun, there has to be a way to determine with some degree of accuracy what players will be good and what players will be bad--otherwise, the game is just random, and that is no fun (to me).

Most of us agree that red and green bars that are almost never wrong (a' la FOF2) is too easy. However, whatever "hook" the designer uses--no matter how subtle--will be discovered by the gamer (or, more accurately, the gaming community.). Word will get out that "Yards per carry in college is the best way to judge running backs or [QS's analysis] is the way to find breakouts and busts." If there is no such hook, then I, personally, would not like the game because it would be too random. (and even then the gamer could exploit the fact that you cannot predict talent by doing things like trading first round picks for multiple late round picks (the opposite of the FOF2 tendancy)).

So, a game should have some way to determine who will be good and who will be bad. If such a way exists, the human will figure it out. The only way to counteract that knowledge is to put the AI on the same level. Then, of course, you are in a position no different that errorless red/green bars--just with a couple more levels of analysis.

So talent evaluation is one area in which I think that it would be very hard to make a robust AI. Roster management is the other main aspect of the game, and I think that the AI could do a better job there. Still, however, by virtue of being a finite, unchanging program, the AI roster management algorithim will be mastered by the human. If it is random enough to be unpredictable, then it will do dumb things. If it is sensible enough to manage a roster well, then it will be too predictable (i.e. you will know that a team with a starting QB rated over X will never draft a QB and you can plan accordingly).

I enjoy the FOF series a lot, in part because I don't need the challenge that others do. For one who does need the challenge, however, I wonder if multiplayer is the only way that you will ever be able to acheive it. I just think that humans are too smart and too adaptable not to master an AI.

Anrhydeddu · 05-14-2003, 09:34 AM

Not if there was a fog-of-war option. In other words, the AI would have all of the information necessary to judge a player accurately (or at least be permitted to use that information in an algorithm), but the humans would not. Is this realistic? I don't know, because I am agreeing with QS and albion on whether a football sim can truly be made competitive.

Bee · 05-14-2003, 09:51 AM

I think the problem is the AI is so bad that it's not even competitive with massive and extensive house rules. I don't think anyone can expect a perfect AI that plays like a human. But I don't think it's too much to ask for one that at least provides some competition when you limit yourself as much as we do playing FOF.

Anrhydeddu · 05-14-2003, 09:56 AM

Bee, I think it can be much better - if it is a priority. It wasn't in the original design of FOF4 (which I still perceive that it went more in the direction of a one-year on-field simulation) but with Jim not sharing and talking with us on the design for the update, what are we to expect or even hope for?

Bee · 05-14-2003, 10:08 AM

I got the feeling FOF4 was more of a base for future development. From my understanding Jim basically rewrote the entire program. He felt he had become a better programmer over the years and it was time to tear it down and start fresh (similar to what Marc and the guys did with CM4). I personally find this a good approach every so often, the problem with doing this with FOF is that Jim has already said he's not making FOF5 next year. By the time he gets around to FOF5, the code he did in FOF4 may have become obsolete again.

As far as the update, I don't really expect any major changes. Maybe a tweak here or there, but Jim seems to be pretty satisfied with the game "as is" which is pretty disappointing to me. I agree more feedback and discussion would probably help Jim flesh out his ideas for the game, but that doesn't seem to be the direction he wants to take.

kid_dynamite · 05-14-2003, 10:31 AM

Quote:

Originally posted by albionmoonlight
If it is sensible enough to manage a roster well, then it will be too predictable (i.e. you will know that a team with a starting QB rated over X will never draft a QB and you can plan accordingly).

but cant you basically do this in real life as well?

albionmoonlight · 05-14-2003, 11:06 AM

The difference is that in real life you may never know what a team is going to do--you can guess, but it won't always be right because people are very complex (i.e. Vikings drafting Culpepper in '98). A program has limits, and any program that will run on our desktop CPUs will have limits that we will probably be able to figure out. The AI teams have to be run by rules, and we are smart enough to divine those rules and undercut them accordingly. Adding complexity may mean that it takes a matter of months instead of weeks to "learn" the system, but it will always be learnable.

This is not a slam on programmers, but a realization that, despite the Matrix:Reloaded hype, AI still has a long way to go to match the creativity of the human brain.

Looked at another way--a game that is suited to computers (chess) still takes a super computer dedicated to the task to match against the best humans. I don't think that our Pentium 4s running a program that can fit on one CD have a chance to match our wits in a game like football. Maybe in 2050, but not now.

cthomer5000 · 05-14-2003, 11:15 AM

Quote:

Originally posted by albionmoonlight
The difference is that in real life you may never know what a team is going to do--you can guess, but it won't always be right because people are very complex (i.e. Vikings drafting Culpepper in '98). A program has limits, and any program that will run on our desktop CPUs will have limits that we will probably be able to figure out. The AI teams have to be run by rules, and we are smart enough to divine those rules and undercut them accordingly. Adding complexity may mean that it takes a matter of months instead of weeks to "learn" the system, but it will always be learnable.

This is not a slam on programmers, but a realization that, despite the Matrix:Reloaded hype, AI still has a long way to go to match the creativity of the human brain.

Looked at another way--a game that is suited to computers (chess) still takes a super computer dedicated to the task to match against the best humans. I don't think that our Pentium 4s running a program that can fit on one CD have a chance to match our wits in a game like football. Maybe in 2050, but not now.

but the computer in FOF4 does not always make the same decisions. Run the first 20 picks of a draft 10 times and you'll get a wide variety of results. Sometimes a player will get drafted in the 1st round, sometimes the 6th. So i don' t think the predictibility you think would be there would necessary be so obvious.

Bee · 05-14-2003, 11:31 AM

I'd also rather have a computer AI that is predictable to one that just continually makes bad decisions like we have now.

Honolulu Blue · 05-14-2003, 12:23 PM

We discussed this in this thread a few months ago.

To make a long story short, I think Jim is at or near the limits of what he can do as a programmer to imitate pro football GMs, and in order to test the brightest of us, he'll either need some help or the computer needs to cheat.

05-13-2003, 09:41 AM	#2
QuikSand lolzcat Join Date: Oct 2000 Location: Annapolis, Md	Sounds fairly promising.

05-13-2003, 09:55 AM	#3
albionmoonlight Head Coach Join Date: Oct 2000 Location: North Carolina	The challenge is there, but at some point, you are not really playing football. Not getting to draft or find rookie gems cuts out a large part of the game. I think that IMT and all of its permutations take out too many aspects of the game that I like for my taste. (Though they do present an entirely new challenge). When I get my new computer, I have an idea for a challenge that will allow me to play all aspects of the game. It is called the Dan Snyder challenge. Before I play the game, I come up with some various pushy and implusive owner demands. (i.e. "We need to open up the passing game--get the best QB available in FA, no matter the cost." "This team believes in loyality--keep our starting running back on the roster until he retires" "I need to make some money with this team--Make sure we are pulling in a profit of $X million within 3 years" "I don't like what our QB just said about us in the media--cut him," etc." Every three seasons or so, I will pull a new demand out of the hat and will have to do what it says. That would allow me to play the game normally, but have challenges thrown in to keep it real. My hope is that the challenges seem realistic enough in the game universe and not just like some weird demands that I wrote down on paper.

05-13-2003, 11:27 AM	#6
Bee Pro Starter Join Date: Oct 2000 Location: Fairfax, VA	I agree it would be better to play the "complete" game, but that's so easy at this point there's little point. I actually tried something similar to what you are looking at albion. I hope you can come up with ideas that are more challenging than mine. Here's a few that I used: Replace the (position group - i.e. secondary, receivers, linebackers, etc) Youth movement (cut or trade everyone over 9 years experience, bring in all rookies). Defense wins Championships (only draft and sign defensive players). Maybe those will give you some good ideas for your challenge.

05-13-2003, 12:26 PM	#9
Franklinnoble Banned Join Date: Jul 2002 Location: Placerville, CA	Even under "misfit toys" rules, it's not too hard to field a winning team. I tried a modified version of these rules (read about it here ) where I drafted only QB's, RB's, and WR's - the logic being that in FOF4, all that really matters is these three positions. I won 4 titles in 11 years - and quit after I had won 3 in a row. I suspect you could do as well with just a top-notch QB and a solid RB.

05-14-2003, 08:43 AM	#10
Anrhydeddu Resident Curmudgeon Join Date: Oct 2002	Here are my thoughts. I think what we are seeing here is the different between the baseball and the football economic models, and thus, the difference between OOTP and FOF. The single thing in baseball/OOTP that can make the AI competitive is disparities between payrolls. You have anywhere from a very small market/payroll team to a large market/payroll team, with enough interesting variations to where a team can spend more than what they believe they can make. Therefore, the acquisition of players through free agency, primarily, as well as through trades can greatly vary from team to team. What this results in for OOTP is the simple rule of having you run a small/medium market team (like I am doing with the 1960s Cardinals) having to compete against the big boys of the Braves, Tigers, Yankees et al. Give a team with good revenues and fan interest and they will go after the best free agents (esp. SP), preventing you from doing so either because of the bidding war or because you just don’t have any money to spend. In football, you have none of that. All teams spend exactly the same. In other words, there are no small, medium or large markets and therefore financials become irrelevant at that level. What this means for FOF is that the only way for the AI to be competitive is to be able to judge each and every player accurately since we (us and the AI) are competing equally from the same pool. But as we have seen over and over, the AI does a poor job in doing that in FOF4. Therefore to make the game some semblance of competitiveness, we have to resort to arcane rules that are unrealistic, like giving us a smaller cap or not drafting or no free agents. In baseball/OOTP, we can do that because that is realistic, in football we should not but in FOF4, we are forced to. I think it is becoming clear, at least among those that are considered the “best” or hardcore customers is that significant effort must be paid to how the AI judges, acquires and manages their players over a career. The reason is, imo, that is the heart of the competitiveness. Nothing else really matters. My frustration with FOF4 comes from the fact that the game truly performs well at the micro level but is lousy at the macro level. Take a look at the dynasties, esp. the GroupThink. Do you see anyone closely examining the logs seeing if the OLB blitzes 37.5% of the time or do you seeing folks simming 4-8-16 games at a time? The details and accuracy of the game always have been good, Jim needs to believe that and we need to have faith in that. But for FOF4, instead of looking more inwards, the direction needed to be more outwards. That has been my criticism of FOF4 and why I felt it took a big step backwards from previous versions. Reading the ideas on house rules and frustrations of some of the best FOF4 customers in playing the game, I think that is becoming obvious.

05-14-2003, 09:30 AM	#12
albionmoonlight Head Coach Join Date: Oct 2000 Location: North Carolina	Another question (and this is from someone who has never played CM--so maybe the "holy grail" answers this question) is whether such a robust AI is possible. In order for the game to be fun, there has to be a way to determine with some degree of accuracy what players will be good and what players will be bad--otherwise, the game is just random, and that is no fun (to me). Most of us agree that red and green bars that are almost never wrong (a' la FOF2) is too easy. However, whatever "hook" the designer uses--no matter how subtle--will be discovered by the gamer (or, more accurately, the gaming community.). Word will get out that "Yards per carry in college is the best way to judge running backs or [QS's analysis] is the way to find breakouts and busts." If there is no such hook, then I, personally, would not like the game because it would be too random. (and even then the gamer could exploit the fact that you cannot predict talent by doing things like trading first round picks for multiple late round picks (the opposite of the FOF2 tendancy)). So, a game should have some way to determine who will be good and who will be bad. If such a way exists, the human will figure it out. The only way to counteract that knowledge is to put the AI on the same level. Then, of course, you are in a position no different that errorless red/green bars--just with a couple more levels of analysis. So talent evaluation is one area in which I think that it would be very hard to make a robust AI. Roster management is the other main aspect of the game, and I think that the AI could do a better job there. Still, however, by virtue of being a finite, unchanging program, the AI roster management algorithim will be mastered by the human. If it is random enough to be unpredictable, then it will do dumb things. If it is sensible enough to manage a roster well, then it will be too predictable (i.e. you will know that a team with a starting QB rated over X will never draft a QB and you can plan accordingly). I enjoy the FOF series a lot, in part because I don't need the challenge that others do. For one who does need the challenge, however, I wonder if multiplayer is the only way that you will ever be able to acheive it. I just think that humans are too smart and too adaptable not to master an AI.

05-14-2003, 09:34 AM	#13
Anrhydeddu Resident Curmudgeon Join Date: Oct 2002	Not if there was a fog-of-war option. In other words, the AI would have all of the information necessary to judge a player accurately (or at least be permitted to use that information in an algorithm), but the humans would not. Is this realistic? I don't know, because I am agreeing with QS and albion on whether a football sim can truly be made competitive.

05-14-2003, 09:51 AM	#14
Bee Pro Starter Join Date: Oct 2000 Location: Fairfax, VA	I think the problem is the AI is so bad that it's not even competitive with massive and extensive house rules. I don't think anyone can expect a perfect AI that plays like a human. But I don't think it's too much to ask for one that at least provides some competition when you limit yourself as much as we do playing FOF.

05-14-2003, 09:56 AM	#15
Anrhydeddu Resident Curmudgeon Join Date: Oct 2002	Bee, I think it can be much better - if it is a priority. It wasn't in the original design of FOF4 (which I still perceive that it went more in the direction of a one-year on-field simulation) but with Jim not sharing and talking with us on the design for the update, what are we to expect or even hope for?

05-14-2003, 10:08 AM	#16
Bee Pro Starter Join Date: Oct 2000 Location: Fairfax, VA	I got the feeling FOF4 was more of a base for future development. From my understanding Jim basically rewrote the entire program. He felt he had become a better programmer over the years and it was time to tear it down and start fresh (similar to what Marc and the guys did with CM4). I personally find this a good approach every so often, the problem with doing this with FOF is that Jim has already said he's not making FOF5 next year. By the time he gets around to FOF5, the code he did in FOF4 may have become obsolete again. As far as the update, I don't really expect any major changes. Maybe a tweak here or there, but Jim seems to be pretty satisfied with the game "as is" which is pretty disappointing to me. I agree more feedback and discussion would probably help Jim flesh out his ideas for the game, but that doesn't seem to be the direction he wants to take.

05-14-2003, 11:06 AM	#18
albionmoonlight Head Coach Join Date: Oct 2000 Location: North Carolina	The difference is that in real life you may never know what a team is going to do--you can guess, but it won't always be right because people are very complex (i.e. Vikings drafting Culpepper in '98). A program has limits, and any program that will run on our desktop CPUs will have limits that we will probably be able to figure out. The AI teams have to be run by rules, and we are smart enough to divine those rules and undercut them accordingly. Adding complexity may mean that it takes a matter of months instead of weeks to "learn" the system, but it will always be learnable. This is not a slam on programmers, but a realization that, despite the Matrix:Reloaded hype, AI still has a long way to go to match the creativity of the human brain. Looked at another way--a game that is suited to computers (chess) still takes a super computer dedicated to the task to match against the best humans. I don't think that our Pentium 4s running a program that can fit on one CD have a chance to match our wits in a game like football. Maybe in 2050, but not now.

05-14-2003, 11:31 AM	#20
Bee Pro Starter Join Date: Oct 2000 Location: Fairfax, VA	I'd also rather have a computer AI that is predictable to one that just continually makes bad decisions like we have now.

05-14-2003, 12:23 PM	#21
Honolulu Blue Dynasty Boy Join Date: Jan 2001 Location: Michigan	We discussed this in this thread a few months ago. To make a long story short, I think Jim is at or near the limits of what he can do as a programmer to imitate pro football GMs, and in order to test the brightest of us, he'll either need some help or the computer needs to cheat.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)