Text Sims: Testing<----->Spoon-Feeding Continuum

[Ben E Lou]

Jim made a comment in the Twenty Questions Thread (http://www.operationsports.com/fofc/showthread.php?t=55675) that I thought might be a springboard for a good discussion involving not just FOF, but other text sims as well.

I don't talk about engine details a lot, because I want people to figure these questions out for their own teams and learn without my help.

This is an area that Wade and I have debated back and forth via IM for quite some time, and I think this comment provides an excellent centerpiece for a larger discussion. I think it is fair to say that Jim gives out less details about the game engine than pretty much many any other text sim developer out there. Further, I firmly believe that less detail plus the number of moving parts in a football sim is what led to what some termed the "Golden Age" of FOFC, a time that was characterized by a myriad of public tests, results, and analysis. In order to understand more about the game, we *had* to break it down bit-by-bit and piece-by-piece, and for a lot of us, it was quite a bit of fun to boot!

However, it seems that today the public well of information has dried up, at least compared to that of releases of FOF2, FOF3, and TCY. It is assumed that MP is the primary culprit: that people don't want to share their detailed gameplans as was done with Fritz's AAA, Morgado's 75 Offense, and countless others in the past. Heck, I used to post my basic gameplan for each season in many of my dynasty threads.

I don't necessarily have a burning question to pose to y'all about the entire matter, but more just wanted to get your thoughts. I'll pick on Wade for a minute, because I know he can take it. ;) On the one hand, he gets frustrated with people when they ask for more information on certain things regarding game engines, because he feels it shouldn't be revealed. On the other, he gets frustrated with people like me who do precisely what Jim is saying we should do: figure these questions out and learn without his help. My point back is that I'm not doin

I guess ultimately what I'm trying to ask is a series of questions. Take your pick from any/all of these. What are your thoughts about it all? Does Jim give out too little info? Do other developers (I think of HR and Arlie in particular, who answer engine questions with great detail) give out too much? Where do you draw the line? Which of these things, discoverable by extensive testing, would you say shouldn't be tested:

that mentors work for only the first three seasons
a point of diminishing returns for increasing time in the training room
how to gameplan to best use highly-rated players in certain categories
how often you can throw deep, run right, etc. without getting the dreaded "the defense looked very familiar with that play"
can 100% usage of certain pass coverage schemes be effective
how combine numbers translate into skillsThoughts?

[QuikSand]

Which of these things, discoverable by extensive testing, would you say shouldn't be tested:

I think I'm missing the thrust of the argument if this is your central piece of it. Is there an argument (from wade, perhaps? that section got truncated, I think) that doing a detailed test to try to discover how stuff works is tantamount to cheating? I don't see a real debate there - seems to me testing in a SP environment on your own is completely fair game.


To me, the most interesting question is the more generic... if there's a continuum between complete disclosure to complete mystery... then where is the ideal balance to be found? (And implictly, in what direction might FOF be "off" from the ideal?)

[Dutch]

I can't recall how many times I thought, "Damnit, why doesn't the help file explain this?" On one hand, I want Jim to spell it all out for me. But on the other hand, it would make it boring in the long run if he did. So I'm glad he does it the way he does.

The key, IMHO, is trusting that there really is logic to the details. If that trust isn't there, I don't think the FOF mystique would last as long as it has. And the proof of that logic existing is in the almanac of stats.

FWIW, this is how I think I get all my info for FOF.

1. Reading everything I could @ FOFC/REFERENCE THREAD

2. Playing in both SP and MP mode (not everybody does this and both are very helpful)

3. Brute Force Testing

4. The Help File

As you can see, two are "figure it out yourself" and two are "somebody spell it out for me".

[wade moore]

I think I'm missing the thrust of the argument if this is your central piece of it. Is there an argument (from wade, perhaps? that section got truncated, I think) that doing a detailed test to try to discover how stuff works is tantamount to cheating? I don't see a real debate there - seems to me testing in a SP environment on your own is completely fair game.


To me, the most interesting question is the more generic... if there's a continuum between complete disclosure to complete mystery... then where is the ideal balance to be found? (And implictly, in what direction might FOF be "off" from the ideal?)

Just to clarify - Yeah, I've taken some issue with the extent of testing done in MP, but I don't think it's cheating - I just personally feel like it takes some of the "fun factor" out of the game.

But, I don't believe that's what SD was referring to. SD and I have had many conversations that revolve around the subject of your last paragraph. I believe strongly that there is a segment of this community that wants to really turn this game into a spreadsheet formula where Solecismic tells them exactly what cell to change in order to make A, B, C, D, X, Y, Z happen. IMO this is a football simulation game. If you start spelling out too much, then you are losing the emphasis on game.

Do I think there are some things Solecismic should probably clarify? Yes (and he did some of them in 20 Questions). However, I think that much of what people beg to be explained (i.e. the exact impact of mentors, exact makeup of pass coverages, what specifically does a red flag player do, etc) would take the mystery and "game" out of this game.

[-Mojo Jojo-]

[FONT=Arial][SIZE=2]In order to understand more about the game, we *had* to break it down bit-by-bit and piece-by-piece, and for a lot of us, it was quite a bit of fun to boot!


I've only resorted to brute force testing once or twice in FOF, and frankly I didn't find it to be quite a bit of fun. I found it to be a tremendous drag. Most of what I've learned about the game came from FOFC. Those details definitely enhanced my FOF experience. I don't understand why that information couldn't have been disclosed up front. What is gained by forcing a bunch of people to waste a lot of time on this? And now, as mentioned, the community-provided aid is drying up...

I think the problem, though, is really not about disclosing details about the game engine so much as design issues. I don't know how the FM engine works either (in fact I'd say I know quite a bit less about how it works than FOF). But it doesn't matter because the game has so much more feedback built in. It is much easier to tell how actions in the game are supposed to impact things, and generally it's easier to tell whether a particular action had its intended effect. When the game lets you do something, it links that to some means for you to get feedback. FOF is incredibly opaque. It has a ton of moving parts, many of them are poorly documented, and it's basically impossible to sort out the impact of tweaking any given one without massive statistical research. If you play the game as its meant to be played, even for a very long time, you're still not going to learn these things. That, IMHO, is poor design.

[st.cronin]

I've only resorted to brute force testing once or twice in FOF, and frankly I didn't find it to be quite a bit of fun. I found it to be a tremendous drag. Most of what I've learned about the game came from FOFC. Those details definitely enhanced my FOF experience. I don't understand why that information couldn't have been disclosed up front. What is gained by forcing a bunch of people to waste a lot of time on this? And now, as mentioned, the community-provided aid is drying up...

I think the problem, though, is really not about disclosing details about the game engine so much as design issues. I don't know how the FM engine works either (in fact I'd say I know quite a bit less about how it works than FOF). But it doesn't matter because the game has so much more feedback built in. It is much easier to tell how actions in the game are supposed to impact things, and generally it's easier to tell whether a particular action had its intended effect. When the game lets you do something, it links that to some means for you to get feedback. FOF is incredibly opaque. It has a ton of moving parts, many of them are poorly documented, and it's basically impossible to sort out the impact of tweaking any given one without massive statistical research. If you play the game as its meant to be played, even for a very long time, you're still not going to learn these things. That, IMHO, is poor design.

+1

[Ben E Lou]

To me, the most interesting question is the more generic... if there's a continuum between complete disclosure to complete mystery... then where is the ideal balance to be found? (And implictly, in what direction might FOF be "off" from the ideal?)Heh. That's what I was trying to say in my thread title, I think, but I got lost somewhere along the way. That's a good summation of the core issue that I'm wanting this thread to explore, and not just for FOF.

[kcchief19]

I never took part in the in-depth testing that a lot of people have done, but I have dabbled from time to time. Most if not all of the testing I have done in the past related to "bug" testing, checking to see if certain features worked they way they appeared to be intended. I read a lot of the testing and studying threads in the past and eventually found that I think they did drain a bit of the fun from the game in the sense that I found some "secrets" and was able to make decisions based on mathematical formulas and not based on "instinct." I continue to keep up on what I'd call "bug" testing because I know Jim is always looking at ways to improve the game, and there are times when I think people view things as bugs that aren't bugs.
However, it seems that today the public well of information has dried up, at least compared to that of releases of FOF2, FOF3, and TCY. It is assumed that MP is the primary culprit: that people don't want to share their detailed gameplans as was done with Fritz's AAA, Morgado's 75 Offense, and countless others in the past. Heck, I used to post my basic gameplan for each season in many of my dynasty threads.
That said, I think MP has had a major impact on the discussion of FOF. In someways that's good because more people are interacting and enjoying the game. In some ways that's a negative because there is less sharing of ideas. The ability to export gameplans has emerged as almost a useless feature since no one appears willing to share gameplans any more for fear of exposing their best stuff to MP. If there is anything that appears to be an exploit, people keep that to themselves for use in MP. But I think the discussion of exploits and the testing that took place in the past helped Jim make a better game.

[flere-imsaho]

I don't see a real debate there - seems to me testing in a SP environment on your own is completely fair game.

Completely agree. In fact, I think we all do this not only in other games, but also in life. If possible, doesn't everyone try to "test something out" before "trying it for real"? It's the reason engineers run simulations before building structures, for example. A lot of good games, I think, reproduce these tasks in less-critical, and more amusing environments, which is, I think, part of what makes a good game.

To me, the most interesting question is the more generic... if there's a continuum between complete disclosure to complete mystery... then where is the ideal balance to be found? (And implictly, in what direction might FOF be "off" from the ideal?)

I think people want to be able to feel sure that they have the tools with which to make educated guesses and reasonable decisions. If they don't have confidence in those tools, they feel frustrated.

So, a "tool" for someone might be knowing, for sure, that Blocking Strength is a base value that is used in a calculation against Pass Rush Strength. If one felt certain about that, one could then make an educated guess about how a Guard might fare against a particular pass rush.

I can't recall how many times I thought, "Damnit, why doesn't the help file explain this?" On one hand, I want Jim to spell it all out for me. But on the other hand, it would make it boring in the long run if he did. So I'm glad he does it the way he does.

I agree, but where I still see room for improvement is in a description of the attributes. To me, there still seems to be a fair amount of confusion over the actual nature of some attributes. When I play FM, for instance, I feel I know well what the attributes mean, and how they translate on the field. I know what Acceleration is, but I'm also comfortable with the fact that in some game situations Player A, with a higher Acceleration, may actually get up to speed slower than Player B.

I don't have quite that level of comfort with FOF. For instance, if a player has a high rating for Long Passes, what does that mean, exactly? His ability to get long passes on target? Well, maybe, but then what is "Accuracy" for? And I know what the Help File says on these, but I still don't feel completely comfortable with my knowledge here.

Since those attributes are the bedrock upon which I try to plan, I often feel as if I'm planning in the dark. It is as if someone was to say "I'd like you to estimate the sum of X + Y, where you know that X has a value between 1 and 30, and Y has a value between 20 and 70." I can do that, sure, but I don't feel particularly sure about it.

Of course, take that with a grain of salt. A lot of that might be my own frustrating in an inability to gameplan. :)

Just to clarify - Yeah, I've taken some issue with the extent of testing done in MP, but I don't think it's cheating - I just personally feel like it takes some of the "fun factor" out of the game.

I know this is a side point, but I do agree. However, I think it's a fact of life. In any MP league there are going to be those with the time and inclination to break the engine down using SP and will thus be gameplanning with more accurate values (to use the analogy from above) for X and Y than those who don't. The fact that most people, regardless of this, can still remain competitive, however, is probably a sign of the strengh of the game, though.

I think the problem, though, is really not about disclosing details about the game engine so much as design issues. I don't know how the FM engine works either (in fact I'd say I know quite a bit less about how it works than FOF). But it doesn't matter because the game has so much more feedback built in. It is much easier to tell how actions in the game are supposed to impact things, and generally it's easier to tell whether a particular action had its intended effect. When the game lets you do something, it links that to some means for you to get feedback. FOF is incredibly opaque. It has a ton of moving parts, many of them are poorly documented, and it's basically impossible to sort out the impact of tweaking any given one without massive statistical research. If you play the game as its meant to be played, even for a very long time, you're still not going to learn these things. That, IMHO, is poor design.

I was going to write more to compare with FM, but this sums it up better than I could. Except that I don't necessarily attribute it to poor design for FOF, but rather the fact that FM is just considerably further down the road. Remember, the CM/FM series is what, almost 15 years old now? And unlike FOF, FM has had a constant team of developers working full-time on it for at least 7-8 years now.

So I think Jim can design a good game. In fact, I think he has designed a good game. He just hasn't had the time/manpower to put in as much feedback and do as much refining as the FM team have, over the years. That's probably why I find FM more immersive and less frustrating than FOF.


To drag myself back to the original question, though, I find it interesting that while I think people know considerably less about how the FM engine works than the FOF engine, from a technological standpoint, it could be argued that in some ways we know exactly how the FM engine works.

Not to be too much of an FM fanboy, but the last few iterations have so accurately reflected how soccer is played, that if you understand the game of soccer you, by extension, understand the FM engine. I think we've gotten to the point with FM that everyone just trusts that it's simulating the actual game of soccer almost perfectly, which is a bit scary. To try and support this, I'll only say that for the past few versions I believe it is correct that no one has come up with a way to "game" FM, which is obviously still possible with a wide range of text sims.

I don't know quite where I was going with that, but hope it's an interesting observation, anyway. :D

[Tim Tellean]

Good question SkyDog. Let me preface first that I do own FoF it's just not my cup of tea so anything I say is on the many games I have tested including TPF, All GDS games, all Wolverine, and Medley Games. I have always looked at long-term issues whether its polls in a game or stats I haven't tried to figure out how something works but rather that it works correctly.
I agree with Mojo in disclosing details but wonder if this has to do with graphics? In FM you see the ball and players move, with BBCF many people question the game because you don't see a moving display so compensate Arlie has beefed up PbP is the same true with FoF? I don't know as I said I don't play the game I do know that FoF dumps a bunch of stats but is it enough?
With MP I've noticed many designers put it in late and don't test it very much so is that an issue?
One thing to add in comparing FM to FoF is that soccer has fewer trackable stats and thus soccer is always moving but football is a play by play setup.
Also I think developers respond too quickly to small sample data for example Arlie with BBCF and polls if a person posts a 1 week issue then he may tweak to fix but throws off another piece.

[Ben E Lou]

If there is anything that appears to be an exploit, people keep that to themselves for use in MP. But I think the discussion of exploits and the testing that took place in the past helped Jim make a better game.It is absolutely CRUCIAL to the long-term health of any text sim that users give the developer feedback regarding exploits. I don't have a problem with people not reporting exploits on a message board. I DO, however, have a problem with people not even reporting exploit to the developer. I've personally run across three potential exploits in my time with FOF since MP was added to the mix. None were explicitly mentioned in public, although information was given to Jim about all three. One turned out to be user error on the part of a league commish, the other two were addressed in some form or fashion.

That being said, in the current environment, I don't anticipate a new wealth of information regarding FOF to come forth from the community any time soon. I would also point to gameplanning, as kcchief did. There was a thread shortly after the release of FOF2K7 where people pledged to share their gameplans publicly. The last post in that thread was over two months ago, and FOFC is still awaiting its first public FOF2K7 gameplan. I can't find the FOF2K4 gameplan sharing thread, but I think that in roughly three years, there were less than 10.

I found it to be a tremendous drag. Most of what I've learned about the game came from FOFC. Those details definitely enhanced my FOF experience. I don't understand why that information couldn't have been disclosed up front.What specific information that you learned from FOFC do you think should have been disclosed up front?

[SteveMax58]

...if you understand the game of soccer you, by extension, understand the FM engine...

I'm a newbie to the FOFC board, but I've been playing sports text sims for years, and I have to say this statement right here is almost verbatim how I evaluate a sports text sim, or any sports game for that matter.

As stated already, I think most people, with regards to attribute definition, want to make sure the integrity of the sim engine is there, so they don't feel like they are spinning their wheels, and can make informed decisions (not to be confused with "precise" decisions). Football is not as easy to translate stats-to-ratings, as opposed to a sport like baseball or basketball, as it is truly a "team" game. So I think a little more definition to the attributes would be helpful, without giving away the "actual" formulas used by the engine.

[Senator]

I feel more than at ease with the FOF engine, and quite enjoy the mysteries of the engine. Red flag, for instance.

[-Mojo Jojo-]

What specific information that you learned from FOFC do you think should have been disclosed up front?

Player ratings and what they mean and how they interact with the game. Position changes versus playing out of position. Also how formations, coverages, and gameplanning options are supposed to work. More detail on things like mentors and red flags. Basically all the things that have to be exhaustively tested by FOFC'ers...

[Dutch]

Player ratings and what they mean and how they interact with the game. Position changes versus playing out of position. Also how formations, coverages, and gameplanning options are supposed to work. More detail on things like mentors and red flags. Basically all the things that have to be exhaustively tested by FOFC'ers...

I've gone through this same dilemna in the past.

With the example of Position Change vs. Playing Out of Position. There are two ways to go about it.

Lets try and be logical and see if we can dissect the differences ourselves.


Imaginary Help File

Position Change vs Playing Out of Position

Changing a players position will cause an experience hit. But over time, the player will become better at this position.

Playing a player out of position does not increase the players skill at the new position, it hinders his growth at his current position (since he isn't playing their to gain experience there) so should really only be used as a stop gap, not as a way to properly develop a player.

Veteran players on the other hand, do not require further developing, so can be played out of position less adversely. However, the individual rating hit for an older player playing out of position is greater.

Changing a veteran player to a new position, will scrub off lost of valuable experience and possible will take a future potential hit. It is not advisable to change a veteran players position.


I would love to see information like that, but I'm not sure Jim wants to break the game engine down that far. BTW: I just made up all that stuff and have no idea if any of that stuff is true.

A great example, as Senator brought up, is the Red Flag. Now, I've joked that it actually does nothing. It's just a mirage that makes you think you want to get rid of the player.

What if that's true? Seeing that information in the game manual would kind of suck. Now, what if it's not true? What should the help file say beyond what it does say "Might mess up team cohesion" or something-rather? What would be better? What does Terrell Owens do to the Dallas Cowboys? Specifically? Does he put a ratings hit on Romo? Does he make the team win more games? What? Who knows.

The same can really be said about the Weak-Side and Strong-Side formation. Does lining the tight end up on the weak side of the formation really make running the ball over there better? What about the extra defender in man coverage that's over there now?

What about a Run Aggressive and Pass Aggressive. If the help file says that Run Aggressive makes you stop the run better, what happens when somebody runs for 15 yards against your run aggressive defense? "Hey! There's a bug! The help file says this, but the game engine did this!" There are so many other factors involved, that it's not even fair for the help file to suggest that.

That's why I've really changed my mind and enjoy the mystery of the whole thing. It's a leap of faith. But one that I trust is true. How do I know? The stats at the end of the game, season, or even a player's career tell me so.

I guess. :)

[Raiders Army]

The key, IMHO, is trusting that there really is logic to the details. If that trust isn't there, I don't think the FOF mystique would last as long as it has. And the proof of that logic existing is in the almanac of stats.
I agree completely that there has to be trust in order for us to blindly believe Jim that the game behaves much like in a simulation of professional football. Note that I say much like, not exactly like.

After some mulling over this topic in my mind during the great early games today (Raiders coming out on top, #1 in April! :rolleyes:), I'm still not sure if that trust is there for me. There are questionable AI decisions both during games and during the course of the season (releasing of stars with huge contracts) that doesn't make it much like a simulation of professional football. It's hard to have that trust that the game mechanics work in some logical fashion when you have no tangible results to prove that they do. I'm not looking for a "if you do x, then y happens" but a better explanation on some of the game mechanics. There's also a difference between knowing how something works and knowing that what you're doing affects the game. For instance, you know that BnR coverage skills are more important to a CB than a Safety. You find a CB in the draft with poor BnR skills and you convert him to a Safety. You do x, then do you still know for sure y will happen? Nope. There are a lot of factors that don't make it a for sure result of a successful conversion to Safety. Your scout's opinion may be wrong, he may not develop as quickly due to experience, etc.

The other thing that doesn't affect me at all is the fact that Jim competes in multi-player leagues. Obviously you guys don't have a problem with this since it's been a few years now in the IHOF, but I might have a problem with it since the creator of the game knows exactly how everything works and you don't. Now I know he's said in the past that he doesn't take advantage of it, but I've also read where you've deduced information about team chemistry from his teams, specifically the fact that he doesn't have any conflicts on his team, so everyone should strive to do the same. I tried to find the exact post that said it, but I can't at this time.

[Tim Tellean]

I'm not sure if trust is an issue since all these games are really just generated random numbers so how you get to a result only helps so much.
Dutch I agree with your point but the imaginary help file I think goes too far in spoon feeding the player. Also I think people want too much real world explained to them in a games help file. The help file should explain how something works, press A and goto B but shouldn't say if you press A then on screen B you will have a chance to alter X and then Y will happen.

[Ben E Lou]

The other thing that doesn't affect me at all is the fact that Jim competes in multi-player leagues. Obviously you guys don't have a problem with this since it's been a few years now in the IHOF, but I might have a problem with it since the creator of the game knows exactly how everything works and you don't. Now I know he's said in the past that he doesn't take advantage of it, but I've also read where you've deduced information about team chemistry from his teams, specifically the fact that he doesn't have any conflicts on his team, so everyone should strive to do the same. I tried to find the exact post that said it, but I can't at this time.I doubt that Jim uses every bit of knowledge he has, but when he agreed to join IHOF, it was with the understanding that he's competitive and would go all-out at times, even using knowledge that we didn't have. We picked up that there was no preseason rust from him sitting his best starters during the first preseason, and we were further convinced about the importance of chemistry because of the way he put his roster together.

[Raiders Army]

I doubt that Jim uses every bit of knowledge he has, but when he agreed to join IHOF, it was with the understanding that he's competitive and would go all-out at times, even using knowledge that we didn't have. We picked up that there was no preseason rust from him sitting his best starters during the first preseason, and we were further convinced about the importance of chemistry because of the way he put his roster together.

IMHO, if you guys don't have a problem with it, I don't have a problem with it. In fact, it probably benefits the community because we can derive answers from his game style. However, this is also in the vein of people keeping information about FOF to themselves instead of sharing it because of MP...so that's why I mentioned it.

[Mac Howard]

I think the problem, though, is really not about disclosing details about the game engine so much as design issues. I don't know how the FM engine works either (in fact I'd say I know quite a bit less about how it works than FOF). But it doesn't matter because the game has so much more feedback built in. It is much easier to tell how actions in the game are supposed to impact things, and generally it's easier to tell whether a particular action had its intended effect. When the game lets you do something, it links that to some means for you to get feedback.

A crucial element in producing a game that can be trusted by users is to balance the information feedback on the consequences of user input with the range of challenges presented both in the form of input options and influencial criteria. Otherwise you leave the gamer with the equivalent of solving the equation x + y = 15. You must include x - y = 9 or you're cheating him. If that balance is built in to the game then there should be no need of explanations of of the mechanics of the match engine.

On the point of difference between FOF and the likes of FM/CM/SaaP it's not just that soccer is not a game that leads to many numerical outcomes but, whereas FOF is essentially about the analysis and manipulation of statistical data, the soccer management game has always historically been a role playing game causing the designer to think more in terms of the qualitative mechanisms of the real-world game than in mathematical representations.

[st.cronin]

One of the things that makes FM/EHM and even BBCF and FBCB very appealing to me is that there are BASIC ratings. In FM, players are rated for their ability to "jump," their "strength," their "decisions." FOF makes a nod in this direction with the combine and some things like "blocking strength" but for the most part all it has is technical ratings.

[Mac Howard]

One of the things that makes FM/EHM and even BBCF and FBCB very appealing to me is that there are BASIC ratings. In FM, players are rated for their ability to "jump," their "strength," their "decisions." FOF makes a nod in this direction with the combine and some things like "blocking strength" but for the most part all it has is technical ratings.

I'm clenching my teeth and resisting the temptation ;)

[st.cronin]

I'm clenching my teeth and resisting the temptation ;)

Um, go ahead.

[Ben E Lou]

One of the things that makes FM/EHM and even BBCF and FBCB very appealing to me is that there are BASIC ratings. In FM, players are rated for their ability to "jump," their "strength," their "decisions." FOF makes a nod in this direction with the combine and some things like "blocking strength" but for the most part all it has is technical ratings.I'd argue that for football, technical ratings make more sense in most cases. The difference between a 7-yard loss and a 50-yard TD pass can be only six inches--the left tackle getting one of his hands six inches too far inside or not. DeAngelo Hall is the league's fastest man, and looked like a slightly above-average corner at best this year, because his technique was lousy. If basic athletic ratings were the key in football, Niner fans everywhere would be talking about the great Renaldo Nehemiah, rather than the great Jerry Rice.

[Ben E Lou]

Um, go ahead.I'm guessing he's commenting that the dude who posted after him left out SaaP.

[Ben E Lou]

Dola:

Oops. You ARE the dude that posted. :p

[st.cronin]

Dude.

[sabotai]

I'm guessing he's commenting that the dude who posted after him left out SaaP.

Actually, I thought Mac was resisting the temptation to talk about leaving out attributes all together.

[Mac Howard]

Sabotai has it exactly right.

I'm guessing he's commenting that the dude who posted after him left out SaaP.

No, no, Mojo Jojo supports his point with whatever evidence he chooses.

My original comment was that a balance has to be achieved between the information provided and the unknown variables in the game. Provide too little information and the problems are insoluble as in the equation x + y = 15 with its infinite number of solutions. Provide too much information and there is little challenge to the game. Including the raw skill numbers is an example of the latter. As Jim once put it in announcing a new release of FOF: "I won't be revealing the raw numbers, where's the fun in that?": there would simply be no challenge to analysing the statistical data as far as the qualities of the players are concerned and essential challenge of the game removed.

I resisted the temptation to take up the point because it is an essential difference between FM/CM and SaaP and something of a hobby horse of mine. St cronin's statement was something I disagree with fundamentally but realised was not something most readers would wish to see pursued - again :)

[Arles]

I had to make a tough decision on the ratings idea when I went from TPF to BBCF. I decided to go with both athletic ratings and some football skill ratings because I found that it would be almost impossible to eventually do a 2-D engine without athletic ratings. Plus, there were other difficulties for the engine. Let's say (in TPF) you had a 100 pass rush speed DE chasing a 100 breakaway speed RB, who wins? Without absolute ratings, you just don't know. Whereas, in real life, someone like Dwight Freeney would get toasted in a sprint by Reggie Bush. Same things go for a LBs and WRs and so forth. Plus, I think it makes changing positions much harder as you really don't have any idea if your corner has the athletic skills and hands to play WR or agility to play RB.

I also think the idea of giving information is tough. I tend error on the information side as I don't want people having logical misconceptions about the game and getting frustrated/angry with it. A great example is Quiksand's very good questions on prestige change in BBCF. Now, it is very logical to think that by simply exceeding board expectations and winning your conference you would get a prestige bump. However, the game was designed to have prestige be more of a national rating that impacts recruiting than a value reflecting the relative performance of your team. The logic here is that just because an Ohio or Troy wins their conference doesn't mean that they will suddenly rise to the level of a midlevel major conference team like Arizona State or Minnesota after a few years. They need to beat some major conference teams, win their bowl game and continue to dominate their other opponents. So, BBCF sets up prestige change to be more of a factor of national ranking, CPU ranking (SOS) and recruit class ranking. However, if this was not explained, many well thought out individuals may just think the system was "broken" if they didn't make gains after winning 9-10 games against patsies as a 40-45 level prestige team. I think you can give information on what impacts different aspects of the game without giving away "secrets" on how the game actually works. And I think most developers have figured out this happy medium (with SI probably being the best).

[st.cronin]

Sabotai has it exactly right.



No. - Mojo Jojo - is right to use whatever evidence he sees that he feels supports his point.

My original comment was that a balance has to be achieved between the information provided and the unknown variables in the game. Provide too little information and the problems are insoluble as in the equation x + y = 15 with its infinite number of solutions. Provide too much information and there is little challenge to the game. Including the raw skill numbers is an example of the latter. As Jim once put it in announcing a new release of FOF: "I won't be revealing the raw numbers, where's the fun in that?": there would simply be no challenge to analysing the statistical data as far as the qualities of the players are concerned and essential challenge of the game removed.

I resisted the temptation to take up the point because it is an essential difference between FM/CM and SaaP and something of a hobby horse of mine. St cronin's statement was something I disagree with fundamentally but realised was not something most readers would wish to see pursued :)


Funny, I think FM is actually more mysterious than FOF. The only thing that makes FOF mysterious is the scout error. If you knew the ratings of the players, you would pretty much know exactly how they would perform. Not so in FOF/EHM.

[Mac Howard]

Funny, I think FM is actually more mysterious than FOF. The only thing that makes FOF mysterious is the scout error. If you knew the ratings of the players, you would pretty much know exactly how they would perform. Not so in FOF/EHM.

I take it you mean FM, st cronin.

The reason is that the FM designers eventually recognised the point I'm making and introduced the "hidden numbers" which render the skill numbers inaccurate.

[st.cronin]

I take it you mean FM, st cronin.

The reason is that the FM designers eventually recognised the point I'm making and introduced the "hidden numbers" which render the skill numbers inaccurate.

Ok, I don't know about any hidden numbers, but what about FBCB? I have the same experience playing that game, and to me it's a superior experience to what FOF provides.

[gstelmack]

I had to make a tough decision on the ratings idea when I went from TPF to BBCF. I decided to go with both athletic ratings and some football skill ratings because I found that it would be almost impossible to eventually do a 2-D engine without athletic ratings. Plus, there were other difficulties for the engine. Let's say (in TPF) you had a 100 pass rush speed DE chasing a 100 breakaway speed RB, who wins?

I would think that, for a pro sim at least, having some athletic ratings would make aging easier. Most players gain more and more football skills as they play, but reach a peak and then decline in physical skills.

On the topic of FOF and information, I think the key was stated by Mac Howard: it's all in the feedback. Why give us a rating without a decent explanation of what it means? From watching guys, I should know how well a "Mentor" helps other players develop. I've watched him work with those players for crying out loud!

Take the case of the old dilemma in FOF: is your star CB not getting picks because he's not really as good as you think he is, or because the QB won't throw his way? But the QB may choose not to throw his way because of double-coverage, or because the CBs are lining up by side and there's a lousy WR over there, or maybe cohesion sucks in your defensive backfield so people are blowing coverages, or any of a number of other factors. Or something even more simple: did the WR make a good play on the ball on a catch, did the DB blow the coverage, was the QB perfect with the throw? There are a lot of factors hidden from you in the FOF stats. You get some of them back if you watch the games in a viewer, but even then you are missing some bits (you don't know who had the coverage on a pass completion, for example; all you know is after the fact, CB X was thrown at Y times, but no clue WHICH plays he was thrown at on). I think BBCF has a start on this with its offensive / defensive performance e-mails, where you know who committed penalties and other sundries that are missing from your FOF game film reviews. But we're still nowhere near the feedback a real NFL coach would have on a player's performance and why that performance happened.

I'm going to be curious to see what comes out of the new log formats as we break them down. There is a lot more information stored in these logs, and I *think* more than is shown in Solevision. It looks like we may know all 22 players on the field for each play, for example. It's going to be interesting looking back in a few months and seeing what we've been able to glean from all this data and if it fills in any of these holes.

[Mac Howard]

Ok, I don't know about any hidden numbers, but what about FBCB? I have the same experience playing that game, and to me it's a superior experience to what FOF provides.

I know nothing about FBCB, st cronin, so can't comment.

An essential part of the challenge to most text sims is discovering the qualities of the players and therefore how you can expect them to perform. If the basic numbers are revealed then there is little challenge to that, as you point out with reference to FOF. In FM the link between basic numbers and performance is modified by a number of factors that are not revealed and therefore your predictions will be inaccurate and you have to use performance data itself to work things out. That's the way it works in FM as far as I understand.

[gstelmack]

An essential part of the challenge to most text sims is discovering the qualities of the players and therefore how you can expect them to perform.

I disagree 100%. The challenge of a text sim to me is landing the right mix of players and finding the correct system to play them under. Things that I would expect to find out through experience might be how well two players get along together, or how a player handles pressure, but I sure as heck know well a guy can throw the ball.

I could see some ratings (my pressure example above) starting off as ranges that narrow as a player plays (and I get more observation of him), and I think ratings growth can be hidden from the player (a strong case could be made for hiding a player potential, or going with the FBB single potential rating), but the ratings and stats are all I have to go on for players in most of these sims. Those need to replace the real-world information that a true coach or GM has. All "hidden" ratings do is add an arbitrary challenge and add artifical difficulty for the player.

[Joker]

I like the mystery.
There are 2 types of players.

2 new guys start playing fof mp

#1 guy quits because he gets tired of being beat over and over by more experienced players.

#2 guy gets upset that his face is in the dirt...and reads and digs for answers and soon...becomes a winner.

If your reading this your probably player #2

[Mac Howard]

The challenge of a text sim to me is landing the right mix of players and finding the correct system to play them under.

Those are certainly part of the challenge but in a game that purports to simulate a coach's experience then surely judging the skills of players is another. Judging the players, getting the right mix, devising effective tactics based around these. Coming up with the right answers from this mix of judgement is the challenge.

Things that I would expect to find out through experience might be how well two players get along together, or how a player handles pressure, but I sure as heck know well a guy can throw the ball.

But you know this from experience of his performances - you've seen him throw it or seen stats that refer to it. I doubt you know it because someone told you he had 15 or of 20 points for "throwing", right? Similarly the game should inform you with some indication of performance.

Ratings can only come from performances.

I could see some ratings (my pressure example above) starting off as ranges that narrow as a player plays (and I get more observation of him),

That's a viable but artificial way of removing the accuracy.

and I think ratings growth can be hidden from the player (a strong case could be made for hiding a player potential, or going with the FBB single potential rating), but the ratings and stats are all I have to go on for players in most of these sims.

That certainly is a flaw in many sims. Or, to more accurate, there are often too many variables and too little feedback.

All "hidden" ratings do is add an arbitrary challenge and add artifical difficulty for the player.

Again I'm tempted, but you are right :)

[JonInMiddleGA]

I disagree 100%.

I'll agree with your disagreement just about the same 100%

(In other words, WYS).

[Marc Vaughan]

I tend to guage things according to the feedback I get from the community - if everyone appears confused about a feature and there's no community support springing up explaining things then I tend to try and explain things - if some people 'get it' then I tend to let the community help themselves, just dropping in to clarify specific points.

Where I do give advice (ie. hints and tips files etc.) I tend to try and do it via. 'real world' situations and examples rather than give out geeky game mechanics as much as possible ... this leaves the engines mystique intact and also imho leaves people thinking of it as a 'game world' rather than a computer program.

(at least this is my take on things - not everyone at SI takes my approach, for instance PaulC often gives out a bit more geeky info on some areas than I would personally)

[Marc Vaughan]

I know nothing about FBCB, st cronin, so can't comment.

An essential part of the challenge to most text sims is discovering the qualities of the players and therefore how you can expect them to perform. If the basic numbers are revealed then there is little challenge to that, as you point out with reference to FOF. In FM the link between basic numbers and performance is modified by a number of factors that are not revealed and therefore your predictions will be inaccurate and you have to use performance data itself to work things out. That's the way it works in FM as far as I understand.

Mac's correct in that FM gives you an overview of a players abilities - however it isn't clear cut because of the interaction between the various attributes a players has (ie. if you have two players with 19 finishing one might be better under pressure and so much more successful than the other) and also because of the high level scores displayed to the user (ie. the scale displayed to the user differs from that in game, for instance if an attribute in game has a scale of 1-100 then its factored down by 5 for user display - this means there is a difference between two players with the same visible attribute level at times), on top of this there are also some 'hidden' attributes which can have an affect upon a players performances etc. ....

I agree with Mac that learning about players is part of the 'fun' of sports sims - however imho I think you need to give users enough information as a starting point to be realistic (ie. what you'd know as a manager from watching a team in training or asking coaches working with them etc.).

[Vinatieri for Prez]

I've only resorted to brute force testing once or twice in FOF, and frankly I didn't find it to be quite a bit of fun. I found it to be a tremendous drag. Most of what I've learned about the game came from FOFC. Those details definitely enhanced my FOF experience. I don't understand why that information couldn't have been disclosed up front. What is gained by forcing a bunch of people to waste a lot of time on this? And now, as mentioned, the community-provided aid is drying up...

I think the problem, though, is really not about disclosing details about the game engine so much as design issues. I don't know how the FM engine works either (in fact I'd say I know quite a bit less about how it works than FOF). But it doesn't matter because the game has so much more feedback built in. It is much easier to tell how actions in the game are supposed to impact things, and generally it's easier to tell whether a particular action had its intended effect. When the game lets you do something, it links that to some means for you to get feedback. FOF is incredibly opaque. It has a ton of moving parts, many of them are poorly documented, and it's basically impossible to sort out the impact of tweaking any given one without massive statistical research. If you play the game as its meant to be played, even for a very long time, you're still not going to learn these things. That, IMHO, is poor design.


Yep.

A designer should give enough feedback so as to aid a player in making decisions about certain game features. Without sufficient feedback, the usefulness of the feature is diminished. Do that enough times and the game begins to suffer. FOF does not give enough feedback and the game suffers as a result.

I'm not saying the feedback needs to be great or completely overt, but there needs to be enough to guide decision making. FOF2K7 is better, but not there yet. Prior iterations were dreadful. We went how many years without ever really getting secondary coverages explained in even minor detail.

If you want to grow a greater base of fans, you have to do that. Many buyers do not test or even know about FOFC. You can see how the game is not as much fun as it could be for them.

[gstelmack]

Those are certainly part of the challenge but in a game that purports to simulate a coach's experience then surely judging the skills of players is another. Judging the players, getting the right mix, devising effective tactics based around these. Coming up with the right answers from this mix of judgement is the challenge.

But you know this from experience of his performances - you've seen him throw it or seen stats that refer to it. I doubt you know it because someone told you he had 15 or of 20 points for "throwing", right? Similarly the game should inform you with some indication of performance.

Ratings can only come from performances.

(first, let me make it clear I'm debating text sims in general and am NOT talking about errors in any specific game)

The way I see it, the sim's job is to present to me the information that I and my staff would otherwise glean from watching players practice and/or play for 40+ hours during the week. They can do it through text or numerical ratings or however they choose, but I should be able to read a player report and know how that player is practicing and playing. It's not my job to sift through reams of numbers strewn across multiple locations and figure out how to interpret them. If you want to make "evaluation" a task, then fine, make it a game task (e.g. the draft interviews in FOF2k7). It's up to me to play the player, determine who I watch closely in practice, whatever. But converting practice and play performance into a report I can read is the game's job.

Now, if the game wants to build in error for players who aren't playing (i.e. there is not enough information for me to build an accurate picture), that's fine, but I ought to know that error as well. The game should tell me I haven't seen enough of the guy to really know how well he plays. That gives me the feedback I need to decide if I should play the guy more to get a better picture of his skills.

To get back to the original topic of the thread (and come back to a specific game), no I don't want to have to quicksim 10+ seasons and run a bunch of statistical reports to figure out what all the numbers in the game mean. That's essentially just busy work and a group project to fill in holes in the game documentation.

For a concrete example, why did we have to run a bunch of tests to figure out that a mentor only helps if the mentor is active, the player is active, and the player is in his first through third seasons? That's actually counter-intuitive: I'd expect a mentor to help by being available at practice and sitting on the sidelines talking to players as they come off the field, and I'd expect him to help anyone who hasn't developed much. And note early on in 2k4 people would bring in a mentor to ride the pine, not realising it wasn't doing any good at all (because you really don't know why the guy is or is not progressing). I'm okay with the game implementation vs. my expectations, but we had to do some work to dig up that info that should have been available in the game's reference. Or, if you want to hide the details behind the scenes, that's fine, but then I need feedback that TELLS me "QB X learned a new tip from mentor Y this week in practice" or somesuch so I can see what impact the mentor is having. Or better yet, hide the mentor part entirely, and just drop things like "QB X learned a new tip from QB Y" and track how many times QB Y has helped someone (I don't want to have to keep a notebook beside me when I play, that's why I own a computer) so I can figure out that having Y around is helping other QBs develop.

[Dutch]

Here is some useful information that Skydog was mentioning earlier. It doesn't answer all of your questions about mechanics, but it answers a lot of questions I've seen thrown around FOFC from time to time.

For instance, earlier this week somebody asked, "Why is there a QB rating for (Long) Passes and then Accuracy. Don't they basically mean the same thing?" According to this, they mean two very different things. Pretty cool information and thanks to Skydog for reminding us of this info.

I think there is a lot of information available to us, but it's up to us to recognize it, put it all together and pick the right guys to put on our squad that work within the framework of our gameplan. I won't even pretend to say I'm very good at this, but I know a few people around here that are pretty good at knowing what to look for and what they want.


Column Y: Completing Screen Passes.
Column Z: Completing Short Passes.
Column AA: Completing Medium Passes.
Column AB: Completing Long Passes.
Column AC: Completing Very Long Passes.
Column AD: Completing Third-Down Passes - ability to convert those pressure-packed third- and fourth-down throws.
Column AE: Scramble Frequency - desire to abandon the pass and scramble downfield. Unlike most categories, this rating is not assigned on a logarithmic curve.
Column AF: Avoid Interceptions - ability to avoid throwing interceptions.
Column AG: Passing Accuracy - ability to connect with receivers in full stride, maximizing their ability to gain yardage after the catch.
Column AH: Passing Timing - ability to connect with receivers who deviate from the planned play.
Column AI: Sensing the Pass Rush - ablility to avoid sacks.
Column AJ: Reading the Defense - ability to adjust to effective defensive pass coverages.
Column AK: Two-Minute Offense - ability to pass when the team is running a hurry-up offense.
Column AL: Preferred Play Style - the set of plays a quarterback is most comfortable running. You can choose a roll-out offense (1), which is best for quarterbacks who like to run a lot, or a short-pass offense (2), ideal for what's often called the "West Coast Offense", or a long-pass offense (3), which is a more traditional approach to passing. Use 0 for non-quarterbacks.
Column AM: Breakaway Speed - ability to break free for a long gain.
Column AN: Running Power Inside - ability to break through the line for yardage inside.
Column AO: Third-Down Running - ability to gain yardage in third- and fourth-down and short situations.
Column AP: Hole Recognition - ability to make the most of running paths forged by the offensive line.
Column AQ: Running Elusiveness - the ability to avoid getting caught behind the line of scrimmage.
Column AR: Running Speed to the Outside - ability to "turn the corner" to gain yardage on outside runs.
Column AS: Running Back's Blitz Pickup - the ability to handle a blitzing linebacker.
Column AT: Receiving Hands - ability to catch a thrown ball.
Column AU: Getting Downfield - ability to gain yardage after the catch.
Column AV: Route-Running - ability to shoulder the load as a receiver. Receivers with high ratings in this category can be the target of more frequent pass attempts.
Column AW: Third-Down Receiving - ability to make catches in clutch third- and fourth-down situations.
Column AX: Big-Play Receiving - ability to catch long and very-long passes.
Column AY: Courage over the Middle - ability to hang in there on routes planned across the middle of the field.
Column AZ: Adjusting to Bad Throws - ability to catch poorly thrown balls.
Column BA: Punt Returning - ability to return punts.
Column BB: Kickoff Returning - ability to return kickoffs.
Column BC: Avoid Fumbles - ability to avoid fumbling the football.
Column BD: Run Block Technique - ability to block for running plays.
Column BE: Pass Block Technique - ability to block for passing plays.
Column BF: Blocking Strength - ability to handle physical defenders.
Column BG: Punt Distance.
Column BH: Punt Hang Time.
Column BI: Directional Punting - ability to kick the ball inside the opponents' 20-yard line.
Column BJ: Kickoff Distance.
Column BK: Kickoff Hang Time.
Column BL: Field Goal Accuracy.
Column BM: Field Goal Distance.
Column BN: Run defense - ability to stop the run.
Column BO: Pass Rush Technique - ability to rush the passer.
Column BP: Man-to-Man Pass Defense.
Column BQ: Zone Pass Defense.
Column BR: Bump-and-Run Pass Defense.
Column BS: Pass Rush Strength - ability to handle physical linemen.
Column BT: Play Diagnosis - ability to recognize the flow of an offensive play.
Column BU: Hard Hitter - ability to punish the ball carrier.
Column BV: Pass Intercepting
Column BW: Endurance.
Column BX: Special Teams Play.
Column BY: Long Snapping.
Column BZ: Holding for Placekicks.

[Ben E Lou]

Here is some useful information that Skydog was mentioning earlier. It doesn't answer all of your questions about mechanics, but it answers a lot of questions I've seen thrown around FOFC from time to time.

For instance, earlier this week somebody asked, "Why is there a QB rating for (Long) Passes and then Accuracy. Don't they basically mean the same thing?" According to this, they mean two very different things. Pretty cool information and thanks to Skydog for reminding us of this info.

I think there is a lot of information available to us, but it's up to us to recognize it, put it all together and pick the right guys to put on our squad that work within the framework of our gameplan. I won't even pretend to say I'm very good at this, but I know a few people around here that are pretty good at knowing what to look for and what they want.
That's funny you mention that. I'm writing an article for Coaches' Corner right now based on the oft-missed information from csv.txt.

[gstelmack]

One of these days, when I'm done working on the utility suite stuff (which means never), I'm going to pull together an updated FOF2k7 help file with two key changes:

1) Add information from other sources. For example, I'd never think to read csv.txt unless I was generating a player data file, so why does it have critical info not in the help file?

2) Convert the file from .hlp to something like .chm, but more importantly into a format where I can just read it front to back. The .hlp file makes you go back-and-forth to the table of contents and is very difficult to print out in any kind of good order. I'd love to just be able to print the freakin' thing as a manual I can read at my leisure. I wonder if the fact that people will have to download a .hlp reader for Vista will force a change in the format?

[SteveMax58]

As a noob to FOF, I think I speak from experience...er...from lack of experience...er, whatever, when I say the .csv is actually the best way I have been able to learn the game, aside from actually playing it. Of course browsing through the FOFC boards has helped as well, but mainly because I've needed some additional info due to the fact that I have also been building a roster set, and need some deeper understandings of how some vitals are used, and with what type of weight. Doing the roster set & testing it out has also helped me learn much more than I would otherwise, but I realize most people do not have the patience or time to do this.

I think all in all, if you follow & know football extremely well, you can pick up FOF very quickly. If you have a casual understanding of football, it might require some reading (like the boards or the .csv). But if you have no clue about football, mainly what positions are for doing what, and why certain vitals are important to those postions...then you may have a tough time "getting it" from just the help file & .csv.

Just my 2 cents, FWIW.

[Honolulu Blue]

I think FOF, in all its forms, has done well at the basic job of a text sim; that is:

Good ratings -> Good players -> Good stats -> Winning teams

I've always been far more interested in beating the game of football than I have in beating the idiosyncracies of FOF. The things I know and have learned about football have been sufficient for me to do well at the game. Therefore, the lack of specific details about Jim's sim engine does not bother me.

[Dutch]

One of these days, when I'm done working on the utility suite stuff (which means never), I'm going to pull together an updated FOF2k7 help file with two key changes:

1) Add information from other sources. For example, I'd never think to read csv.txt unless I was generating a player data file, so why does it have critical info not in the help file?

2) Convert the file from .hlp to something like .chm, but more importantly into a format where I can just read it front to back. The .hlp file makes you go back-and-forth to the table of contents and is very difficult to print out in any kind of good order. I'd love to just be able to print the freakin' thing as a manual I can read at my leisure. I wonder if the fact that people will have to download a .hlp reader for Vista will force a change in the format?

I started to actually try cutting and pasting the help file into a Word document for that exact reason (off computer reading) back in the day. But I found it an impossible task to do. I'll owe you a beer one day if you can figure out how to do it automatically. :)

[stevew]

I really don't much care for doing the copious amount of testing that others do. I really appreciate the insight, although to some degree I think that it makes the game too mechanical. The engine to FOF is pretty solid, so even when people test the hell out of it, and develop trends, it still seems to have interesting outcomes. I used to be in a basketball league, with a far inferior product to FOF, and the way people would end up testing and testing that game made playing in the league unplayable. Text sims shouldn't be where you have a 90% certainty of what your exact shot distribution should be. Even if it was annoying when RB2 got too many carries, it was a still nice bit of variation, so that a team couldn't "reherse" what they were going to do.

In short, I like not knowing everything, merely cause it keeps the results random enough so that a bunch of different outcomes may occur. Even if the help file is a little obtuse.

[Mac Howard]

The way I see it, the sim's job is to present to me the information that I and my staff would otherwise glean from watching players practice and/or play for 40+ hours during the week. They can do it through text or numerical ratings or however they choose, but I should be able to read a player report and know how that player is practicing and playing. It's not my job to sift through reams of numbers strewn across multiple locations and figure out how to interpret them. If you want to make "evaluation" a task, then fine, make it a game task (e.g. the draft interviews in FOF2k7). It's up to me to play the player, determine who I watch closely in practice, whatever. But converting practice and play performance into a report I can read is the game's job.

I couldn't agree more. My own choice is to present to the manager-gamer the assistant coach's assessment of player performance on the training ground at team selection time. Alongside the team-sheet is a written report detailing the week's training regimes (set by the gamer-manager at the beginning of the week) and the coach's assessment of each player's performance in those training sessions. By selecting the various regimes you can highlight any particular skills you might need to monitor as well as developing those skills long term and preparing for the coming match. There's also an option to see the coach's assessment of the current form of the player taking account of the player's fitness, confidence and commitment as well as his inherent skills.

Albeit, this information should be subject to any weaknesses in your assistant's judgement thus introducing another individual and his particular problems into the game :)

It's not only important to provide the gamer-manager with the information he needs but at precisely the time he needs it.

I'd expect a mentor to help by being available at practice and sitting on the sidelines talking to players as they come off the field

I think there are two aspects to this - the players skills development and the development of an effective mental approach to the game.

I argued above that the gamer-manager is responsible for developing his own opinions on players' abilities and I would add to that that he's also responsible for the motivation and development of players (though this may be limited to soccer).

For the first I think to talk to a player during the match and training sessions to praise him, criticise him etc depending on his performance and thus influence his motivation and confidence. The particular influence should operates via the players' temperament - it's important to understand what comments work with different temperaments.

The second is effected by defining individual training regimes for each player and monitoring progress. That monitoring can come from the assistant coaches detailed to observe changes in the player performance both in training and in matches.

but then I need feedback that TELLS me "QB X learned a new tip from mentor Y this week in practice" or somesuch so I can see what impact the mentor is having. Or better yet, hide the mentor part entirely, and just drop things like "QB X learned a new tip from QB Y"

I very much like the idea of conversation - with players, with coaches, with media even with fans. By using conversation with a player the effect on him can be displayed by his verbal response as well as his performance. Conversations between the gamer-manager and players, coaches etc can not only provide useful information about the player's abilities and state of mind but also increases the feeling of realism in the game.

Again, this is all said in the context of a role-playing text sim rather than a stats based one. I think we sometimes exaggerate the similarities between the text sims like FOF and the soccer management games.

[AlexB]

I'm in agreement that the basic information (i.e. a desciption of what each skill means as above, a description of what each slider in FM actually means, etc) should be clearly presented.

The above information is fantastic for me, and IMHO should be far more obviously available in FOF: after seeing this I can now see connections between QB skills and RB/WR skills that would be complimentary if in tandem, would be compensatory if in opposition, and disastrous if present in neither.

IMHO we should be given the information to make decisions and deductions, which I feel Jim & SI (being the developers of the two games I play more than any others) keep slightly too close to their chests at the moment

[AlexB]

Dola,

Mac, I've been impressed with almost everything you've posted here on text sims in general over the last couple of years, but found myself unable to get into SaaP on the couple of occasions I dl'd the demo.

Can't explain it - from your posts here it should work for me http://www.talkingballs.co.uk/images/smilies/102.gif

[Mac Howard]

I've just released SaaP 2007, Jari, so try the demo download again. One day we'll make the connection ;)

The problem is that often these ideas are difficult to put into practice and are sometimes not as clear in the application as the description. It does require much work and, as a loan designer trying to compete with teams of multiple programmers funded by multi-million dollar publishers, there is only so much time for me to do that and also produce what is essentially a more difficult simulation. Innovation takes time and often the first implementation is lacking but it gets better as time proceeds and I find more users are beginning to see the ideas emerge as the game evolves.

So, just check it out every so often. I'm told by users who have had similar problems that it suddenly hits you smack in face and you wonder from then on why you missed it. Not what I intended but I'll take it :)

[gstelmack]

I think FOF, in all its forms, has done well at the basic job of a text sim; that is:

Good ratings -> Good players -> Good stats -> Winning teams

Amidst my critiques, I'll highlight this as something I agree with wholeheartedly. The criticisms are in the details and dealing with certain features, not with the fundamental nature of the game and engine (as I think everyone brought out in the poll thread).

[twothree]

A game should give me the same information and mechanics that the AI is using to make it's decisions.

Example: If the AI knows "that mentors work for only the first three seasons" then somewhere in the instruction and/or help file, I should be given that information.

Edit: To clarify. I am not talking about knowing why the AI makes a decision or what goes into it (the formula). However, any "absolute" rule that would include the words (always, only, never) that the AI is using to make it's decision, should be known to the player.

Basically, I want to be playing by the same set of rules as the AI.

[Marc Vaughan]

I'm in agreement that the basic information (i.e. a desciption of what each skill means as above, a description of what each slider in FM actually means, etc) should be clearly presented.
Normally in FM you can get the 'full text' for any abbreviations by hovering over a title until a tool-tip shows, for in-depth descriptions of screens please refer to the in-game tutorial which describes how most things work (but isn't perfect and this is an area we're actively trying to improve each version).

[AlexB]

Normally in FM you can get the 'full text' for any abbreviations by hovering over a title until a tool-tip shows, for in-depth descriptions of screens please refer to the in-game tutorial which describes how most things work (but isn't perfect and this is an area we're actively trying to improve each version).

This was a reference to the mass confusion over what mentality and closing down actually meant, as there were differing descriptions in the game and the manual. However, you have have indeed clarified a lot in FM07.

[cuervo72]

That being said, in the current environment, I don't anticipate a new wealth of information regarding FOF to come forth from the community any time soon. I would also point to gameplanning, as kcchief did. There was a thread shortly after the release of FOF2K7 where people pledged to share their gameplans publicly. The last post in that thread was over two months ago, and FOFC is still awaiting its first public FOF2K7 gameplan. I can't find the FOF2K4 gameplan sharing thread, but I think that in roughly three years, there were less than 10.

I usually try to be collegial about findings, I just don't find things very often. Oh, and I freely shared my gameplan once.

Haven't read this whole thread, but in FOxL chats PineTar has mentioned many a time that FOF is somewhat a collection of Jim Meta-Puzzles greatly increases the shelf-life of each version. Though yes, it can be frustrating at times, especially when there are some things that never seem to be figured out. It's almost as if the game has to be a puzzle that you can think you have insight on, but aren't really ever sure...and even if you unlock one part of the game, there are enough things going on that it doesn't ensure success.

Something like mentoring though - even though it's testable if you're into that sort of thing, guidelines on mentor limitations are nice I think. I don't think knowing that they are limited to R-3 really hurts, if anything it adds a little more decision-making, rather than just playing generically that "mentors are good" (akin to "playcalling is good", "good scouts are good", or "motivation is good").

[cuervo72]

2) Convert the file from .hlp to something like .chm, but more importantly into a format where I can just read it front to back. The .hlp file makes you go back-and-forth to the table of contents and is very difficult to print out in any kind of good order. I'd love to just be able to print the freakin' thing as a manual I can read at my leisure. I wonder if the fact that people will have to download a .hlp reader for Vista will force a change in the format?

Word.

[Warhammer]

Column AQ: Running Elusiveness - the ability to avoid getting caught behind the line of scrimmage.

I never realized that this is what elusiveness measured. No wonder backs with it tend to do better.

[Warhammer]

Here's my total .02...

I want to know how the engine looks at things.

For example, do the formations in our game plan set up matchups against the defense, or is there a bias for or against certain types of plays?

On defense, what exactly do the coverages mean?

In this regard, FOF2k7 is lightyears ahead of FOF2k4. But, there are still some gray areas. Do LBs actually engage in BnR coverage, or are they too far off the line to use it? Will safeties ever actually use BnR skills, or are they essentially just M2M and zone defenders?

The other problem I have with FOF is some extreme happenings in the game. If my defense has held the other QB to 8 of 24 passing, why does the guy look like Johnny U in the last two minutes of a game? Or, you hold the other team to a total of 150 yards for the game and then you look like the Keystone Kops over the last 2 minutes of the game. What is happening here? All I see is that I gave up 2 TDs over the final two minutes, but what went wrong, did they play better, was I in the wrong coverage, what?

[astrosfan64]

Personal opinion here.

But, I believe that any game that doesn't have a "2d" engine has zero proof that the actual game is being played behind the scences. We've argued this point here before, but there really are only 2 games that are accurately portrayed in our genre. That is EHM and FM.

Now TPB has a 2d engine, but you can tell by the player movement that the players have "fixed" places on the court and players only move in very basic patterns. Yes, it is a start to a 2d engine, but it isn't an accurate dipiction of a true game of basketball.

Total Pro Football, Front Office Football both to me had decent text based engines, but are a long way off from "true football". As much as everyone hates Madden here, in coach mode it is the closest thing to true football to date.

I believe that some developers are particulary vague about how certain things in their engine work, because either
A. they don't really work correctly or
B. they are gimics that don't really effect anything.

Most developers want you to believe that all the ratings and numbers mean something and maybe they do, but overall it is your belief that they do that is important.

Take this quote below as an example. This to me is an example of a developer fudging the truth a bit rather then admit a deficiency in his engine. (I'm not picking on Jim most of them have done it, this was just the most readily available quote to use)

10. Since cornerbacks and safeties don't have pass rush ratings, what ratings, if any, should we look at to determine whether a defensive back will be an effective or ineffective blitzer?
Look at the offensive line and the players being kept in to block. A defensive back doesn't stand much of a chance of getting to the quarterback if the blitz is recognized. I'd run less blitzes against teams that don't give up a lot of sacks, and teams that have low numbers of running back receptions. Defensive back blitzes are more a function of good defensive game planning in the NFL, not individual skills. Don't listen to the propaganda.

So the great Safeties of our time, who hit harder, move faster and read offensive plays better then their counterparts, will only get a sack if the offense doesn't pick it up?

Or the elite CB's who are extermely quick/agile and hard hitters won't fare any better then some scrub?

Yeah that makes absolutely zero sense....

Don't get me wrong, it isn't that big of a deal. But, don't attempt to give us a real world explanation why something in the game isn't accurately simmulated.

[astrosfan64]

Column AQ: Running Elusiveness - the ability to avoid getting caught behind the line of scrimmage.

I never realized that this is what elusiveness measured. No wonder backs with it tend to do better.

Does this mean that elusiveness stops being a factor once you get past the line of scrimmage? It shouldn't.

[Icy]

First at all, i would love to see one day a text sim developed by gstelmack, I fully agree with all that you post. You have expressed exactly how i feel about text sims in general and FOF in particular. What about a Raimbow Six text sim where you must choose, train and equip your team, deal with their personalities, conflicts, team work etc to acomplish the different missions? :)

Just an example of why we need more info from the developer or at least from the match engine (better the second).

Column AH: Passing Timing - ability to connect with receivers who deviate from the planned play.

Umm ok, i understand it in real l life... but what about FOF? How can you know or where is it said in the game pbp or which stat represent that a receiver deviated from the planned play??

In a 2D/3D engine and knowing the play that was called, you could see a receiver running a wrong pattern but in a text sim, we would need it written in the pbp like "Pass Incomplete, WR John Smith ran a wrong pattern and QB Peter Smith could not find him". that way we would know:

1- WR John Smith is not good running the right routes.
2- QB Peter Smith Passing timing ratting is not good enough to compensate the inability of WR John Smith.

Also which receiver rating represents the abillity to run the right route? I would say Route running is the one that makes the most sense but...

Column AV: Route-Running - ability to shoulder the load as a receiver. Receivers with high ratings in this category can be the target of more frequent pass attempts.

Does it mean that the receivers good at RR run better the planned routes or just that they will be thrown more passes?

That kind of questions have nothing to do with the experience of being a real NFL coach/GM that is what I look for when I play a game, an RPG experience where "I feel I'm there". If i need to do long term tests to know how the game works, then it becomes a tedious job or a puzzle that kills the RPG experience for me. I don't want to be a puzzle player but an NFL GM.

Another example and how to ruin or enhance the RPG experience:

I decide to look at the Attitude Advisory Screen and i see that player X is disgruntled. Umm ok, but why? Reading the manual i see that it's because lack of play time and that it could affect future player contract negotiations. Ok, i got the info now but... wouldn't it be way more immersing if instead i would receive an email from player X telling me that he is concerned about his playing time and that if things don't change he will consider becoming an FA instead of resigning a contract?

Both give you the same info, but the second option is way more immersing and you get immediate feedback into the game instead of having to look at the manual or read the boards etc.

That is where FM shines, on what I call the RPG experience and on feeling I'm a real coach interacting with players, media, coaches etc etc instead of just looking and analyzing statistical data shown in different screens and trying to understand what does it mean.

Note: I used FOF as example as it's what we discuss more in this board, but it applies to most of the text sims, don't take it wrong, I love FOF and how it's engine is really accurate to simulate real football but i really dig the FM experience and i would like it applied to every text sim, of course not just copied but the same idea developed differently by every games programmer.

[Narcizo]

I agree with Icy - if we can't see what's happening then we need to be provided with more commentary so we have something other than stats. Can you imagine a coach basing his decisions on stats and the word of his staff? There pretty random nature of the game makes this even more necessary. Of course we can't follow training sessions and the like so you have to have numbers involved somewhere but at least we could be given the tools to assess what is actually happening in a game.

Another problem with Jim's method is that it delays the discovery and fixing of bugs. Take the running back endurance bug in 2004. Nobody was sure if that was a bug for ages and lots of effort was applied to understand if it had to do with the running back taking part in special teams, or passing plays or whatever. Finally it gradually emerged that running back with high endurance couldn't run the ball as often as one's with average endurance. In a more open system I'm sure this would have been discovered sooner.

[Warhammer]

Does this mean that elusiveness stops being a factor once you get past the line of scrimmage? It shouldn't.

Actually no, that is how Jim set it up, that is how it is supposed to work.

While I agree with you in theory, after all it is how I viewed it as well prior to yesterday, that is how it is set up in game, there is nothing wrong with it. The rating could be called "ability to get back to the line of scrimmage" and no one would have an issue with it.

So the great Safeties of our time, who hit harder, move faster and read offensive plays better then their counterparts, will only get a sack if the offense doesn't pick it up?

Or the elite CB's who are extermely quick/agile and hard hitters won't fare any better then some scrub?

I tend to agree with Jim on this. One of the reasons why corner blitzes are rare in the NFL is that it is an all or nothing gamble. Typically you either get the sack, or it goes the other way for a big play. All it takes is the RB to pick up the blitz and the offense is fine. The primary factor here is some one reading the blitz.

Column AH: Passing Timing - ability to connect with receivers who deviate from the planned play.

The other problem is that in the help file for FOF2k4, this is referred to as the ability to capitalize on defensive break downs. While I can reconcile the difference, they are not necessarily the same thing. So which is it, and I agree with Icy, what WR skill corresponds to this skill, or is it purely on the QB side?

[mrsimperless]

Actually no, that is how Jim set it up, that is how it is supposed to work.

While I agree with you in theory, after all it is how I viewed it as well prior to yesterday, that is how it is set up in game, there is nothing wrong with it. The rating could be called "ability to get back to the line of scrimmage" and no one would have an issue with it.

The argument you just used to defend this is actually the core of the problem. The issue is not that he decided to implement it that way. The issue is that this is not readily known to the player. If in fact he HAD called it "ability to get back to the line of scrimmage" there would have been no problem. But I've been playing FOF for what, 5 years now and I just learned this today? That isn't right.

[Raiders Army]

I agree with Icy - if we can't see what's happening then we need to be provided with more commentary so we have something other than stats. Can you imagine a coach basing his decisions on stats and the word of his staff? There pretty random nature of the game makes this even more necessary. Of course we can't follow training sessions and the like so you have to have numbers involved somewhere but at least we could be given the tools to assess what is actually happening in a game.

Another problem with Jim's method is that it delays the discovery and fixing of bugs. Take the running back endurance bug in 2004. Nobody was sure if that was a bug for ages and lots of effort was applied to understand if it had to do with the running back taking part in special teams, or passing plays or whatever. Finally it gradually emerged that running back with high endurance couldn't run the ball as often as one's with average endurance. In a more open system I'm sure this would have been discovered sooner.

RAmen.

[Gary Gorski]

Now TPB has a 2d engine, but you can tell by the player movement that the players have "fixed" places on the court and players only move in very basic patterns. Yes, it is a start to a 2d engine, but it isn't an accurate dipiction of a true game of basketball.

Since you brought up the 2D engine that is in TPB (as well as TCB and Total Pro Golf) I figured I would chime in here. Yes, players have "fixed" areas on the screen meaning that when a player is in the low post on the right side of the basket he's always displayed in the same location on the screen - however I don't think that takes it or the engine behind it away from being an "accurate dipiction of a true game of basketball".

There are more cosmetic things that can be done with it. The positioning of the icons could be moved around a general area as opposed to always being in the same location. If that's not the kind of thing you're referencing here I would love to hear what you think needs to be done to take the 2D engine in TPB and TCB to the level you place FM and EHM on. Certainly if you and others feel that my 2D engines are a step below then I want to do whatever I can to bring them up to that level.

[Marc Vaughan]

I believe that some developers are particulary vague about how certain things in their engine work, because either
A. they don't really work correctly or
B. they are gimics that don't really effect anything.
I'd opt for:

C. Realise that the workings are too complex to describe easily (for example I often get asked how the valuations in FM are calculated - there IS no easy way to describe this - the main function is around a thousand lines long and relies upon various sub-functions to assist it).

[Honolulu Blue]

Can you imagine a coach basing his decisions on stats and the word of his staff?

No, but I can imagine a GM making decisions that way. Most GMs in the NFL are neither current/former coaches or players, so they have to base their decisions on what they can see (the stats) and hear (the coaches).

I've always thought of FOF as a GM game with head coaching options grafted on. From that perspective, the ratings and the way they work are fine. If you think of FOF as a head coaching game with some GM options, then, yeah, there are some issues to work through.

[SteveMax58]

Column AV: Route-Running - ability to shoulder the load as a receiver. Receivers with high ratings in this category can be the target of more frequent pass attempts.

Does it mean that the receivers good at RR run better the planned routes or just that they will be thrown more passes?

FWIW...I took that to mean that recievers who run their routes correctly & precisely will get more balls thrown to them, as IRL...but QB's may complete passes to receivers who miss-run their routes more often than otherwise if they have a high rating in Pass Timing.

I do agree with many of your points, though. Just throwing my $.02 in the event its useful.

[astrosfan64]

I'd opt for:

C. Realise that the workings are too complex to describe easily (for example I often get asked how the valuations in FM are calculated - there IS no easy way to describe this - the main function is around a thousand lines long and relies upon various sub-functions to assist it).

Marc - in no way was my comment directed toward your games. You have written guides on how your game works and explain things to the user. I don't want actual formulas, I just want to know how the general ratings work and what the effects are related to the user.

If I think strength means how strong a player is, I assume it effects tacking, blocking, break tackles, etc...

But if it isn't a factor with any of these ratings, then I need to know this, or I could draft or build my whole team incorrectly.

I do realize option C. exists for many different games and I'm ok with that. My point was directed to certain games and features.

In FOF or TPF there is no way you are going to convince me that all 22 players on the field are modeled during a play. Each with their own AI that are making realtime decisions as a play develops.

You guys in the past have describe both your hockey and soccer engines in great detail. You've even explained how often the game updates the playing field in terms of .(x) amount of seconds. You've explained to us what different skills do and have given us real world examples.

[wade moore]

In FOF or TPF there is no way you are going to convince me that all 22 players on the field are modeled during a play. Each with their own AI that are making realtime decisions as a play develops.

I know this is a bit of a threadjack, but...

So?

[astrosfan64]

I know this is a bit of a threadjack, but...

So?

I don't get the point of your So?

If this isn't happening, then you aren't playing football. You are just playing a random number generator. If that is what you want then great, but I want a football simulation.

FM and EHM are what I consider simulations of their sports.

Madden of all things in a coach mode, is a simulation of football. All players are accounted for and react to things on the field.

I guess it is a matter of taste.

[Brian Swartz]

If this isn't happening, then you aren't playing football. You are just playing a random number generator. If that is what you want then great, but I want a football simulation.

FM and EHM are what I consider simulations of their sports.

This is just my opinion, but no matter what sim you are talking about there are 'random number generator' parts to it. For example, do the games you like simulate the precise effect of humidity, air density, etc. on everything on the field(the grass, the players, the ball/puck, the ice, etc.) for each nanosecond of every game. Of course not, it's ridiculous -- you'd need a supercomputer to do that. To this extent it could also be argued that these games are not a true 'simulation of their sport'.

Similarly, a football game which(and this is hypothetical, I have no idea how FOF/TPF do it) is based on probabilities could still reproduce the sport to a high degree of accuracy(though not absolute precision, which you will never get for any sport). This use of general probabilities will always come into play, it's just a matter of how much of it you use. The use of it does not mean that the game in question is not a true simulation; the test of that is in how well the results it generates mimic the real-world sport.

[astrosfan64]

This is just my opinion, but no matter what sim you are talking about there are 'random number generator' parts to it. For example, do the games you like simulate the precise effect of humidity, air density, etc. on everything on the field(the grass, the players, the ball/puck, the ice, etc.) for each nanosecond of every game. Of course not, it's ridiculous -- you'd need a supercomputer to do that. To this extent it could also be argued that these games are not a true 'simulation of their sport'.

Similarly, a football game which(and this is hypothetical, I have no idea how FOF/TPF do it) is based on probabilities could still reproduce the sport to a high degree of accuracy(though not absolute precision, which you will never get for any sport). This use of general probabilities will always come into play, it's just a matter of how much of it you use. The use of it does not mean that the game in question is not a true simulation; the test of that is in how well the results it generates mimic the real-world sport.

Good points, but to me it is different.

Especially when one of these games lets us start designing our own plays and watching the onfield action develop in front of our eyes.