Interpreting Coverage Ratings

[flair1234]

Here is a for instance. Say a safety is rated 100 in zone and 20 in bump and Run.

When using say a cover 2 bump and run coverage, which rating is being applied for the safety? Is it the 100 since his roll in the cover2 bump and run is to play the deep zone, or is it the 20 for Bump and Run.

Classic noob question I am sure, but I have been wondering.

[Ben E Lou]

I would think that if the safety is playing deep zone, then his zone coverage rating is being used.

[PSUColonel]

Here is a for instance. Say a safety is rated 100 in zone and 20 in bump and Run.

When using say a cover 2 bump and run coverage, which rating is being applied for the safety? Is it the 100 since his roll in the cover2 bump and run is to play the deep zone, or is it the 20 for Bump and Run.

Classic noob question I am sure, but I have been wondering.

Not really a bad question actually. I myself am not sure of this either.

[PSUColonel]

I would venture to guess that man to man and bump and run are more important for CB's, while zone and punishing hitter along with run stop is more important to S. I would also guess that both are important for LB's.

[Vinatieri for Prez]

I would think that if the safety is playing deep zone, then his zone coverage rating is being used.

I believe that is the answer, but there is no confirmation from that in game or from Jim.

[JeffW]

If it's not true, the game manual is very misleading.

[CU Tiger]

While I would agree with SkyDog...

Lets play Devil's advocate.

If a team lines up in a 3 wide set

Who steps up to cover the "X"?
In most cases (in RL) this would be the SS...but are we suggesting that the LBs cover WR in FOF...hrmmm.....

if NOT..would we have 2 CBs playing BnR and 1 Safety at the line playing a loose zone

[JeffW]

If a team sends 3 WR in, wouldn't the defense send a nickel package in and cover the slot with a CB? The offense can only have 11 men in the huddle.

Even with no-huddle offenses, there is a rule in effect that the defense must get a chance to respond to personnel changes.

[Daimyo]

If it is the case that the safety's zone coverage is what counts that would make the BnR and M2M ratings pretty useless for safeties and that would be a decided step backwards for the franchise.

[PSUColonel]

If it is the case that the safety's zone coverage is what counts that would make the BnR and M2M ratings pretty useless for safeties and that would be a decided step backwards for the franchise.

I'm sure it counts to some extent, but probably not near the amount zone coverage does. Rarely do you see a safety line up man to man. Ony if the wrong personnel group was on the field, or on an odd blitz package, a cover 0 or a cover 1, these are run packages.

[Vinatieri for Prez]

We need a definite answer on this one since it's pretty important to know. Jim, any help here?

[Narcizo]

If a team sends 3 WR in, wouldn't the defense send a nickel package in and cover the slot with a CB? The offense can only have 11 men in the huddle.

Even with no-huddle offenses, there is a rule in effect that the defense must get a chance to respond to personnel changes.

In real life but not in the game. Your nickle and dime coverage is decided by the percentage you set, which I've always thought was a bit odd and, I think, one of the reasons that overloaded passing schemes were so successful in 2004.

[mrsimperless]

I'm confused about this as well. I was assuming that if I ran a bump and run D that the safeties' bump and run would be used. But here is some compelling evidence: When you're in a "full zone" coverage the game calls this in the PBP a cover-7 zone. (4 linemen, + all LBs and DBs in zone coverage) When you run a bump and run or man to man it says a cover-2 zone. This would lead me to believe that safeties' BnR and man skills are never used unless you're using a cover-1 scheme then the SS would be playing man / BnR.

[mrsimperless]

Hmmmm, the game planning screens don't support my previous post. The options are 1-deep BnR/man, 2-deep BnR/man, 3-deep zone, 4-deep zone and the mixed coverages. I'm not sure what is happening with "Cover-7".

Still confused!

[CU Tiger]

Hmmmm, the game planning screens don't support my previous post. The options are 1-deep BnR/man, 2-deep BnR/man, 3-deep zone, 4-deep zone and the mixed coverages. I'm not sure what is happening with "Cover-7".

Still confused!

TOTAL SPECULATION

1-DeepBnR - CB- BnR/man FS - Robber (1deep) SS- run support or TE coverage

2-deep BnR/man CB- BnR/man FS & SS Robber OLB plays TE or 3rd WR

3- Deep zone Both CBs and FS play deep zones and SS plays run support
LBs play R/R

4-deep CBs and SS and FS play Deep Zone cover LB play read/react

Cover 7 Both CB,FS,SS and 3LB (not SILB in 3-4) play zone.

I havent done enough testing to confirm this, but it is what seems to be happening based on pass defense stats and players involved in plays

[TimL]

While I would agree with SkyDog...

Lets play Devil's advocate.

If a team lines up in a 3 wide set

Who steps up to cover the "X"?
In most cases (in RL) this would be the SS...but are we suggesting that the LBs cover WR in FOF...hrmmm.....

if NOT..would we have 2 CBs playing BnR and 1 Safety at the line playing a loose zone

In real life, if the defense didn't adjust their personnel nor their coverage, a cover 2 man defense would still have the 2 safeties deep and then break out a linebacker to cover the slot receiver man 2 man. Normally, this linebacker would be covering the TE or the RB.

It's not the worst thing in the world. He still has 2 safeties to help out deep.

[QuikSand]

ping jkat ... this discussion could really use some more insight. You got anything here?

[cmp66]

Is the safety's man coverage used when doubling a receiver?

[jkat]

As Far as Coverage
I don't have anything concrete on how coverages work. My understanding from the few dropped hints about how coverages operate is that in 07 there is significantly more crossover between the ratings and coverages. A 2 Deep BnR won't just use zone for the safeties, instead they'll use their BnR to recognize and quickly adjust to the wide out if released to them, zone to pick up anyone entering their area, and MtM to stick close to whichever player once they've picked them up.

Unfortunately, I doubt there are too many calculations internally but instead simply a certain amount of weight to each rating when the formula is ran. In a coverage where a safety might of only used zone in 04 it might be .7 * zone and .3 * MtM in 07, as an example.

Then there are the questions as to how much the game looks at individual ratings as opposed to global, what ratings are used by each position in each particular coverage, and if other ratings might only affect certain coverage ratings (like whether play diagnosis might have a disproportionate synergistic effect with zone as opposed to BnR and MtM).

If I had to come up with a theory I would say that rather than a particular coverage rating it is a blend wherein: MtM is the quality of coverage, Zone is the proper pickup/release of players going through the zone, and BnR is the general ability to use BnR plays (pressing for corners and understanding of who/where they are covering for deeper players). So while a player with very high zone might consistently cover the right player if he has very low MtM he wouldn't be very effective if the ball was thrown to his man. This is all questionable at best, however.

For Most People
Between:
The quickly diminishing returns of heavily adjusting the defensive coverages.
The (possible) weighting of coverage ratings
The "acceptable" results of using the recommend button to determine coverage percentages.I'd say having fairly balanced (as far as coverages) players, a little "skewing" of the coverage percentages towards what you want, and gameplaning around the player/s who are a serious liabilty is your best bet if you want to avoid taking some legitimate risks in using what you suspect "might" work.


Experimentally
You could find how much each rating matters to each coverage on a global, not individual level, with data similar to the "Passing Comparison" table from http://www.fof-research.com/passinfo.php. Though you would need to be have a way to extrapolate that information from logs/solevision.

Comparing that sort of data from a default roster 2006 universe with 1 in each coverage rating, as a control, to a 9 in one and 1s in the other two would give you an idea of the global weighting of each rating in a particular coverage. 9 in two and 1 in the other and all at 9 would complete that global data set. The differences in the tables should show how much each rating matters, globally, to a particular coverage assignment.

Individually you would need to be able to get more information, that I don't believe is there, concerning who was covering who from the logs/solevision. The above process with 1/9 in the ratings for one position at a time would be a possibility but you would need extremely large sample sizes because of the limited effect on the overall result from only 1 player from 7 being altered. Plus, you would have to be prepared for the effects of positional bias in assisted depth charts and the effect of personnel changes/double coverage in assisted game plans or go through all of it manually. Doing all of that would be, well, slightly insane in my opinion.

[Daimyo]

*sigh*

Its such a shame Jim doesn't give more info on things like this and instead we're left to theorize and crunch numbers to try to figure it out. I don't think this bullcrap about keeping things mysterious enhances the game as much as he seems to think it does. Instead you end up with the few people with the time, inclination, and knowledge required to run tens or hundreds of trials and do the linear regression type stuff having a pretty significant advantage over those who lack any of the three.

Perhaps there is less to the actual engine than meets the eye and the lack of information is simply a defensive mechanism. Judging from what little we've been able to chip away about how it works and Jim's reluctance to share any more details, I guess that may be the conclusion we're left with unfortunately...

Anyway, thanks for sharing the insight you've been able to uncover jkat. You've obviously spent a good deal of time hacking away at it. You should post more often. :)

[yabanci]

The coverage system in FOF2k7 is completely asinine, the unfortunate consequence of thinking that unintelligible = complex and challenging.

[Subby]

jkat - have you ever emailed with solecismic? I would be interested to see if he validates any of your assumptions. I had some decent exchange with him a while back that at least seemed to get me going in the right direction.

[Icy]

jkat - have you ever emailed with solecismic? I would be interested to see if he validates any of your assumptions. I had some decent exchange with him a while back that at least seemed to get me going in the right direction.

(I'll reserve comment on how emailing the developer in order to learn how to play his game seems kind of goofy. Ok, comment made. ;))

Can you post a resume of your questions and answers? maybe it can tell us a bit more.

[Icy]

The coverage system in FOF2k7 is completely asinine, the unfortunate consequence of thinking that unintelligible = complex and challenging.

Agree, it's like saying that playing Madden with your TV turned off is better as it's more challenging.

[Synovia]

Agree, it's like saying that playing Madden with your TV turned off is better as it's more challenging.

Agree. I bought this game because I was told it had a great sim engine. Its pretty clear to me now that the sim engine is a lot simpler than people think, and isnt all that impressive. The fact that the developer wont even tell us what certain coverages do, or what ratings matter is ridiculous.

Its typical security by obscurity. Hes worried about us knowing what ratings are used because the game IS so simple, and if we knew what the ratings did, it would make the game extremely easy. Hes using obscurity to hide a poor engine.

[gstelmack]

As a detractor of late, I will step in and say I don't think you can consider the sim engine "poor". The sim engine does what it is supposed to do: generate realistic NFL-style results influenced by player skill and gameplanning choices.

The issues I think come about in the interaction with the sim engine. By itself it is fine, but does not work for how gameplanning is done and scout error is modeled.

It's like the debates about whether using table lookups or force models were best for WWII flight sims. Neither is inherently wrong, rather it's the wrapper you put around them. For example, I hate 6DOF force model flight sims when I don't have force-feedback sticks, because the model requires that kind of feedback to fly well. FOF is like a table lookup sim that you are flying with force-feedback sticks: you think you are getting lots of good feedback, but the reality is the sim is just making up reasonable numbers to hand back to the feedback system so it "feels" close, but the reality is the sim is taking care of auto-trimming and the like for you and you're much better off flying with a regular pair of stick and rudders so you aren't fooled about what's happening. And under the covers it is still a fine piece of flight simming.

[st.cronin]

Well said, g.

[Synovia]

under the covers it is still a fine piece of flight simming.

Theres no way to tell without a proper interface, and I honestly dont think its going to turn out that way if we do get a good interface.


The engine assigning stats after computing the results of a play is a very clear sign that the engine is on the simple end.

[wade moore]

Well said, g.

qft.

[bmerryman]

I think I'll defend the developer on this one. The game seems to operate fairly intuitively to me. Against the pass, a Free Safety with good zone coverage and/or good Play Diagnosis (which is basically what is most important most of the time for FS), generally does a good job. If he's also good in BnR coverage and you run BnR coverage then he'll continue to do a good job most of the time. If he's facing a five receiver set in any non-dime set then his m-2-m is going to weigh in more heavily. To me, the engine can still be simple (just like football is simple) but the gameplanning is not simple in either case.

[gstelmack]

The engine assigning stats after computing the results of a play is a very clear sign that the engine is on the simple end.

Simple != bad. This engine has been around for many years and for many game iterations, including a college version, that people have had a lot of fun with. That tells me the engine itself is fine for what it has been asked to do in the past. One reason to keep it simple is that people scream to high heaven if FOF generates .1 ypc more than the NFL, so it's important that the engine generate realistic OVERALL stats for this crowd.

The only reason people are starting to rag on it is not that the engine all of a sudden started stinking it up, but rather:

- Gameplanning got a new facelift that seems easier to create the gameplan you want, but is now creating complications trying to determine who will do what.
- Scout error was increased, causing a new focus on specific stats.

In FOF2k4, you could mostly trust your scout with veteran players (or at least we thought we could, maybe we were working on false assumptions), so stats were window dressing to all of us. It was cool to see who did what, but it was output from the game, period. So we didn't need to dig too deep.

In FOF2k7, stats became a feedback input to help determine if a scout was correct or not. So we started looking harder. And we started exposing weaknesses in them.

Remember back in 2k4 when people would bring up weird plays and Jim would play "Imagine this" games with us? They were fun and fine and we can all see weird bounces in real NFL games. Now we're going "But my Center sucks. How is he out doing these kinds of things? Does he really suck, is he better, or is the engine lying to me?" That's a key change in atmosphere, not in the engine itself.

[DougW]

Just my take, and I have no evidence to support.

I have always taken the ratings at value when it comes to what I think happens in coverage.

I play as if the ratings all mean something, so I play as if a safeties BnR ratings are not "useless".

That said, I assume that a safties BnR ratings are used when the defense is playing bump and run coverage, that his M2M ratings are used when the defense is in man 2 man, and zone is used when it's in a zone or mixed coverage.

Yeah, I know that usually he'll be in a zone even if the D is in a BnR; but I play as if his BnR rating is a rating of how well he plays/helps when his corners are BnRing.

Maybe I think too simple.

[Warhammer]

It amazes me that people have such an issue with safeties when I think the real problem is at the cornerback positions. Safeties will play in zone for the majority of the time. Whether it is 50% or 90%, I think it is safe to assume that zone is the primary coverage you need there.

The question about coverage should come into play at the CB position. If I play a 2 Deep Man to Man, I can assume that I need good MtM corners. But how much does zone factor into this. I think I can assume that it only factors a little into this. Let's say maybe 30%. What happens when I am in a nickel or dime though. Is the nickel or dime back in a zone or are they in MtM coverage? That makes a big difference because that MtM safety might be a better nickel back than my #3 CB.

Taking it one step further, when I move to a 3 Deep zone, how do the weights work out? Again, I can assume that zone is a big factor, but is MtM the rest of the factor? Or worse yet, what about 2 Deep zone, BnR? Are the corners in BnR 70% and MtM 30%, or is it BnR 70%, Zone 30%?

[Vinatieri for Prez]

Put it this way. This is now probably the 3rd in-depth thread on secondary coverages. I participated in one mostly with QS several months ago. I asked Jim a straight up question in a Q&A that was half-answered. Then 6.0e came out and while he claimed he was responding to people's requests for more information, his answer was still considerably lacking.

As sad as it is to say, I have basically given up trying to figure it out. I think I have a good handle on it, but I never truly will. It's an astounding developer decision to leave customers continually in the dark that really is quite baffling. It can only hurt future and continued interest in the game. It's too bad, really.

[Synovia]

As sad as it is to say, I have basically given up trying to figure it out. I think I have a good handle on it, but I never truly will. It's an astounding developer decision to leave customers continually in the dark that really is quite baffling. It can only hurt future and continued interest in the game. It's too bad, really.

I agree. There is absolutely no way in hell I will buy another version of the game without that changing.

[ace1914]

You guys are wrong in saying that the game's ambiguity is going to hurt sales. I just bought the game in January and while never owning any of the earlier versions, I love the game. This kind of game is like the perfect progression from all-day console gaming that I used to do for the last 20 years. I introduced my one of my younger brothers(he's 25 now) to the game last month, and he had a blast as well(we are both sports nuts). I don't look for the flaws in the game or an explanation for which coverage is used where. I take the game at face-value. A player that is rated well in a certain area, will more than likely do well in that area. Its not guaranteed, that he will, but MORE THAN LIKELY he will. Why not draft a safety that has great bnr and one that has great zone, and whoever exceeds, stays. Seems rather straight forward.

Also, how many of you guys would have rated Frank Gore do as well as he did? That's a perfect example of someone playing better than his "bars" would indicate. Why is everyone so hell bent on getting a step-by-step explanation for the game? From what I see in the leagues that I play in, most of you do pretty damn well most of the time and have a lot figured out anyway.

[ace1914]

Sorry for the rant. But honestly, shouldn't more emphasis be placed on results rather than "scouts views" anyway? A safety that makes plays is what you should look for. But I guess I got off topic since this discussion is more based on gameplanning towards your teams strengths. After typing that above, I can see your point of view, but I'd rather try to base my playing time by who performs on the field, regardless of bars. Its more immersive that way, imho.

[Shard77]

Put it this way. This is now probably the 3rd in-depth thread on secondary coverages. I participated in one mostly with QS several months ago. I asked Jim a straight up question in a Q&A that was half-answered. Then 6.0e came out and while he claimed he was responding to people's requests for more information, his answer was still considerably lacking.

As sad as it is to say, I have basically given up trying to figure it out. I think I have a good handle on it, but I never truly will. It's an astounding developer decision to leave customers continually in the dark that really is quite baffling. It can only hurt future and continued interest in the game. It's too bad, really.

Sadly, you aren't the only one to feel this way. We recently had two veteran GM's (Who also happened to be great guys) quit our league because of their frustration over the seemingly random results of the sim engine. A little more insight into how the engine works could only help current and future players enjoy the game more.

[jbergey22]

The problem is if Jim comes out and tells us everything it becomes simple to manipulate the game.

[ace1914]

The problem is if Jim comes out and tells us everything it becomes simple to manipulate the game.

I agree. In the PFL, as a newbie and looking at everyone's records, the same teams seems to win. Do they understand how the game "sims" better than everyone else? I doubt it. I bet you guys find something that works and then stick with it instead of trying to beat the system by trying to predict the outcomes of games based on what you "feel" should have happened.

[Vinatieri for Prez]

Sorry for the rant. But honestly, shouldn't more emphasis be placed on results rather than "scouts views" anyway? A safety that makes plays is what you should look for. But I guess I got off topic since this discussion is more based on gameplanning towards your teams strengths. After typing that above, I can see your point of view, but I'd rather try to base my playing time by who performs on the field, regardless of bars. Its more immersive that way, imho.

One could argue that with the current stat engine, it is very difficult to determine whether the safety delivered "results." I'm not saying you can't, and I do believe that the stats are much better for evaluating secondary players. But for some positions, like O-line, you don't have much in terms of evaluating results.

[Vinatieri for Prez]

The problem is if Jim comes out and tells us everything it becomes simple to manipulate the game.

Can't agree with you. Nobody is asking Jim to divulge precise formulae of how, for example, coverages is computed. I would hardly call it manipulation if you were simply told (1) who the DB actually covers, and (2) what/how in general terms coverage ratings are used in different coverages so you could properly evaluate a player's ability. Right now, instead of manipulation, you are stumbling around in the dark.

[Vinatieri for Prez]

I should add this comment. I do not make these comments just to bash the game. I make them as constructive criticism in hopes of improving a very good game and make it a great game. It's still one of the best sports sims in current form. I just want it to be the best!

[Warhammer]

There are two problems with the situation. First, with the scouting error as it is, I don't know if the bars I am seeing are real. Fair enough, I need to pay attention to stats. But since I don't know how to evaluate when my DB is getting beat, I don't know what I am doing wrong.

I've also noticed anecdotally, that if I go with a "pure" system, mainly 3 Deep Zone or something, I get torn up. So you also have a gameplanning aspect in the results as well. So I am left wondering, did I have the wrong gameplan? Is my 50/50 player actually a 35/35 player? Or, is my guy just getting beat in certain coverages that he sucks at? Or, is this guy going up against their best WR?

That is where things get frustrating. There is no real feedback to determine what was happening. In FOF2k4, I could at least look at the bars and know what was expected to happen because scout error was much less. So you could be reasonably sure of what the situation was. When using a "pure" system you knew if you were getting "diced" or if it was a gameplanning issue.

The problem is where we don't know where ratings are used. Does BnR coverage use all three coverage ratings? Does M2M just use M2M and Zone ratings? What is each LBs responsibility in coverage? Etc.

[Synovia]

The problem is if Jim comes out and tells us everything it becomes simple to manipulate the game.

Only if the engine is flawed.

Also, how many of you guys would have rated Frank Gore do as well as he did? That's a perfect example of someone playing better than his "bars" would indicate.

Thats exactly what Frank Gore isnt. Gore was a stud in college, and blew out his knee. He should have been drafted high in the first, but wasnt because of the injury history.

Hes a stud, and hes performing like one. Think of him like a high bars, high combine red flag.

[ace1914]

Only if the engine is flawed.



Thats exactly what Frank Gore isnt. Gore was a stud in college, and blew out his knee. He should have been drafted high in the first, but wasnt because of the injury history.

Hes a stud, and hes performing like one. Think of him like a high bars, high combine red flag.

I'll detract my statement on Gore. But I still disagree on the only being manipulated if its flawed. Again as I stated earlier, I don't understand how you guys see the game as being flawed. Its sounds like you guys want to know if a guy has goo bnr skill bars he "has" to succeed at corner, or if he has good zone bars he "has" to succeed in zone. That is not how real life works and there is a lot more grey than black and white. Is the game pefect, no, but its a very good representation. But I guess since I don't have any other txt sims to base it off, its more enjoyable than most console sports games.

[Ajaxab]

Its sounds like you guys want to know if a guy has goo bnr skill bars he "has" to succeed at corner, or if he has good zone bars he "has" to succeed in zone. That is not how real life works and there is a lot more grey than black and white.

This is precisely the problem. We have to be able to explain why certain players succeed and others don't. Right now, it's extremely difficult to figure this out. All we have to go on are: 1) bars (that may or may not be reliable); 2) stats (that may come from a player's performance, the gameplan, cohesion, opposition skills, opposition gameplan, etc.). We have to work with what we can see. In real life, we can see a corner get beat time and again and plausibly identify why it might be happening. We don't have this vision in FOF. So we have to work with what we can see, the bars and the stats. Unfortunately, the bars can't be trusted, so we have to go to stats. But how do we know a player's stats reflect their talent level? Theoretically, you go to their bars to see if the performance matches their apparent skill level. But again, the bars can't be trusted so it turns into an exercise of a dog chasing his tail.

2k7 makes me an agnostic gamer. I don't really know if there's anything I can plausibly trust in this game that enables sensible decision-making. For whatever reason, the designer won't allow me to see some of the things that would really help in this sense-making process. Like others, I'm not looking for all the answers. I just want to have faith that what I'm doing is making a difference. Right now, that faith is rapidly fading. It's unfortunate, but the game's lack of useful feedback on key components like coverage ratings has shelved the game for me.

[ace1914]

This is precisely the problem. We have to be able to explain why certain players succeed and others don't. Right now, it's extremely difficult to figure this out. All we have to go on are: 1) bars (that may or may not be reliable); 2) stats (that may come from a player's performance, the gameplan, cohesion, opposition skills, opposition gameplan, etc.). We have to work with what we can see. In real life, we can see a corner get beat time and again and plausibly identify why it might be happening. We don't have this vision in FOF. So we have to work with what we can see, the bars and the stats. Unfortunately, the bars can't be trusted, so we have to go to stats. But how do we know a player's stats reflect their talent level? Theoretically, you go to their bars to see if the performance matches their apparent skill level. But again, the bars can't be trusted so it turns into an exercise of a dog chasing his tail.

2k7 makes me an agnostic gamer. I don't really know if there's anything I can plausibly trust in this game that enables sensible decision-making. For whatever reason, the designer won't allow me to see some of the things that would really help in this sense-making process. Like others, I'm not looking for all the answers. I just want to have faith that what I'm doing is making a difference. Right now, that faith is rapidly fading. It's unfortunate, but the game's lack of useful feedback on key components like coverage ratings has shelved the game for me.

If the bars are an automatic verification for performance on the field, where's the challenge or the surprise of the overachiever or underachiever? If you can predict the outcome of each of your players on every play there is no replay value in my opinion.

[Ajaxab]

If the bars are an automatic verification for performance on the field, where's the challenge or the surprise of the overachiever or underachiever? If you can predict the outcome of each of your players on every play there is no replay value in my opinion.

As others have said in this thread, no one is asking for a magic bullet. No one wants bars that are an automatic verification. I know I don't. But what I do want is a correlation that is trustworthy. Trustworthy every single time? No. But certainly trustworthy enough that I am not fumbling around in the dark every time I go to evaluate a player. And trustworthy enough that there is some correlation between bars and performance. In this version, that correlation just doesn't occur often enough to make the bars reliable at all.

[Warhammer]

If the bars are an automatic verification for performance on the field, where's the challenge or the surprise of the overachiever or underachiever? If you can predict the outcome of each of your players on every play there is no replay value in my opinion.

Any good sim is going to have a random number added to the value of the bar to determine the result. So, assuming a D100 and players rated 50 and 100 in corresponding skills, the guy rated 50 is going to outperform the 100 rated guy about 25% of the time (I'm winging this here so that could be high). You'd have 50 + d100 vs. 100 + d100. So there is still a significant random factor, and depending on how the roll is, you can have a guy with a 1 rating beating a guy with a maxed out rating, but it would be rare, etc.

What we want and what we need is a feedback method that allows us to pick people out. Right now, we could have a potential breakout 25 rated player vs. a 50 rated guy going the other way. Depending on the position, we may never have a chance or opportunity to figure out the 25 rated guy is the one to keep.

Case in point, my DL guys typically have between 20 and 40 tackles per year. Over the course of 4 preseason games, how am I to tell the difference between normal production, good rolls, small sample size, etc. Especially when over the long haul, these guys might get 2-3 tackles a game.

[Julio Riddols]

Did the Jets know Jeff Blake was going to be a star for a few seasons? I can't see why they would cut him if they did.

I think the game is less of a crapshoot than most people think, I guess, because I have never felt "cheated" by results of a game unless I messed up somewhere in my gameplanning and put TE in two spots on the depth chart in a certain formation or something.

Anyone can beat anyone on a given day. Outside factors such as luck, temperature, humidity, wind, turf condition, crowd noise levels, and several other things besides a players "talent" affect every play.

I believe the bars we see are much less misleading than most people seem to think- Instead I think maybe we rely too much on something like a rating being a good determinant of a players ability to play. There are things like team chemistry and coaching ability to go on top of the changing conditions during a game that affect everything that happens on the field.

I think maybe thats why some players can excel in Denver's running system but be an average back for other teams. I think thats why I look at the receivers for Atlanta as largely an unknown quantity, because they have had one of the most unorthodox QB's in the league throwing them the ball. That puts them in far different situations than a receiver playing with a pocket QB who does things entirely differently.

Due to all the above, whether it is a flawed or simple system or not, I still find that the game delivers an experience (to me) that is extremely close to simulating what it is like in real life to try and find that Marques Colston in the 7th round. I agree with the notion that knowing too much about the inner workings of the game would probably ruin the game entirely rather than making it more fun to play, because players in real life don't even come with ratings. They come with emotions, autonomy, and humanity, which the characters in this game can't supply. The world has their opinions, there are people that say Eddie George was a great back while I believe he was an overrated runner who benefitted from having the ball shoved into his stomach 30 times a game. I always had that opinion, and always considered him to have less talent than his contemporaries. Who knows whether he actually truly was one of the better backs of his time or not? There is no concrete thing that can give the world a unanimous answer to that question. There is opinion, conjecture, fact based arguments.. But if Eddie George played his career with another team, who is to say he ever becomes a consistent starter, let alone a player people considered to be highly qualified at his position? Maybe with another team he puts his name alongside the greatest of all time.

I think when it comes down to it, every player is a "system player", some to a greater extent than others. Maybe instead of assuming the game is flawed or oversimplified, we should consider that there is no way to know how a specific player will do at any time, given any specific set of circumstances.

I think the much greater "flaw" in the game is found when playing single player games and watching computer teams run into ridiculous cap trouble or make some seriously questionable moves. This whole thing about bars remaining "hidden", etc. seems to be much less based on desire for "realness" than it is based on our desire to have an advantage over our opponents.

After all, if the ratings were as much of a crapshoot as people are making them out to be, I doubt we would see teams competing at top levels consistently in multiplayer leagues while others struggle to make the playoffs.

I think achieving victory is based much more on the strategy of the battle than it is the ability of the warriors fighting it.

[Julio Riddols]

As an example of a system being a major factor in player success, look at my WLB in the OSFL, Otis Vaughn.

He is a 6 year vet, fully developed, rated 68/68. (Which is within a point or two of where he was when I signed him in FA)

In the season prior to our signing him, he played 944 snaps in 16 games at SLB (441 run, 503 pass)

He recorded the following stats: 100 tackles, 25 assists, 1 sack, 0 blocks, 5 hurries, 4 QB knockdowns. His PRpct and Tkpct were 1.2 and 13.3 respectively.

we signed him, moved him to WLB, and the following season, he played 772 snaps in 14 games (303 run, 469 pass)

He recorded the following stats: 73 tackles, 30 assists, 20.5 sacks, 3 blocks, 20 hurries, and 43 QB knockdowns. His PRpct and Tkpct were 9.3 and 16.0 respectively.

I understand I changed his position, and that gave him different stats, but I think he probably would not have made the FA market if he had just disrupted
80 plus of his opponents pass plays the season before. He also had a tackle percentage almost 3 points higher even though he was on the field for a far lower percentage of run plays than pass plays, and he wasn't covering receivers as much, as he only had 22 balls caught on him as opposed to 40 the previous year on only 34 more pass plays. Where did his tackles come from?

I believe it was the system.

My point being, the way he was used made him one of the most devastating defenders in the game. He single handedly made my defense what it was and I would consider him irreplaceable. Apparently the team who had him before me didn't think he was, because he was let go in FA.

If you go back to the grassy knoll thread about ratings masking, the first thing I think you have to think about is how long SkyDog has been dissecting this game and trying different gameplans and combinations of players. I doubt there are many SP seasons where he doesn't have a team that is among the leaders of the league in multiple categories. The Quarterback in question in that thread was reeling off tons of TD's a year, just going crazy, while other QB's with seemingly superior ability were performing worse.. I believe the argument there was that the A.I. needs to be able to see a "creeper" creep or know that a guy is getting better so they don't just go and release that player when they could make perfect use of him.

Here the argument seems to be that the game itself runs in such a simple manner that on a given play your players are simply weighted percentages waiting to be assigned stats based on a variable enhanced outcome. I have a feeling it isn't quite that simple.. Who knows though, maybe it is.. But whether it is or isn't I still feel as though I have control over what my team does on the field. I created the offense and defense I used in the OSFL entirely from the ground up with no A.I. involvement and fine tuned them over a season or two with a similar team in SP, then was amongst the best defenses in the league once I moved that gameplan to the MP league. I tweaked small things like blitzing percentages or play expectations in common situations occasionally (1st and 10, 2nd and 10, etc.) but basically it was the same defense all year.

The offense was also among the top in many categories, however, in the positional strength ratings, my team steadily falls around the middle of the pack at most positions.

Given that I have logged between 20 and 30 seasons of SP action in FOF 2007, I don't think I can attribute much of my success to knowledge of the inner workings of the game. Instead, this leads me to believe that the gameplanning interface is intuitive enough that a relative novice can get the most out of his team with some careful consideration. I probably spent about 4 to 6 hours all told with the offensive and defensive plans I used for the season.

Another thing I think worth mentioning is that my defensive game planning ability seemed to get a huge boost when the last patch came out. Jim said the game was using defensive game plans wrong, and there definitely seemed to be some significant differences when that patch was released. I think I was able to go from having an inexplicably mediocre offense and defense to having a strong team on both sides of the ball because I actually felt as though the teams strengths were being utilized by my game planning where it was much more of a crapshoot beforehand.

Thats all I really have to say about the topic. I think the game is fun and it works the way it was intended more so than almost any game I have ever played. I think by over analyzing the inner workings of the game, people might become discouraged by things here and there. On the other hand, maybe I am selling myself short by being satisfied with the game in its current iteration, with its imperfections being dissected in front of me over time.. But I don't really care how the game works inside. I feel like I have the control I want and I don't feel like I miss anything when it comes to playing against 31 other humans.

I do agree the SP A.I. still needs to be perfected in its ability to recognize improving players and keep good players signed, and it would be nice to be able to eliminate MP "midweeks" without using roster workarounds, but really those are my only small gripes about an otherwise outstanding game. I guess I just don't understand why others feel as though they're being duped when I feel like everything plays out about how it should in most situations.

[Ajaxab]

As an example of a system being a major factor in player success, look at my WLB in the OSFL, Otis Vaughn.

He is a 6 year vet, fully developed, rated 68/68. (Which is within a point or two of where he was when I signed him in FA)

In the season prior to our signing him, he played 944 snaps in 16 games at SLB (441 run, 503 pass)

He recorded the following stats: 100 tackles, 25 assists, 1 sack, 0 blocks, 5 hurries, 4 QB knockdowns. His PRpct and Tkpct were 1.2 and 13.3 respectively.

we signed him, moved him to WLB, and the following season, he played 772 snaps in 14 games (303 run, 469 pass)

He recorded the following stats: 73 tackles, 30 assists, 20.5 sacks, 3 blocks, 20 hurries, and 43 QB knockdowns. His PRpct and Tkpct were 9.3 and 16.0 respectively.

I understand I changed his position, and that gave him different stats, but I think he probably would not have made the FA market if he had just disrupted
80 plus of his opponents pass plays the season before. He also had a tackle percentage almost 3 points higher even though he was on the field for a far lower percentage of run plays than pass plays, and he wasn't covering receivers as much, as he only had 22 balls caught on him as opposed to 40 the previous year on only 34 more pass plays. Where did his tackles come from?

I believe it was the system.

First, I appreciate the articulate, reasoned response here. There are definitely some useful points to consider. As I read this, I wonder how you knew Vaughn would succeed in your system. Do you look at his stats over and against his ratings and see a disconnect, roll the dice and sign him anyways? What was it about this player that caused you to pick him up? How did you know you could convert him to WLB and use him such that he would become one of the most devastating defenders in the game? You were evidently using some feedback that the other GM who released him was not using.

Given that I have logged between 20 and 30 seasons of SP action in FOF 2007, I don't think I can attribute much of my success to knowledge of the inner workings of the game.

This is an interesting comment. How would you explain your success?

I think I was able to go from having an inexplicably mediocre offense and defense to having a strong team on both sides of the ball because I actually felt as though the teams strengths were being utilized by my game planning where it was much more of a crapshoot beforehand.

How do you identify your team's strengths given the feedback mechanisms we have in place? Crapshoot implies randomness. How did you move from feeling as though things were random to feeling as though you had some control?

All of these questions have to do with the golden feedback question. I completely understand how someone might make a case for saying that I am looking for a constantly predictable x+y=z kind of a simulation. I could see how someone might accuse me of trying to get MP secrets. But I am interested in neither. I am merely interested in trying to figure out how others are making sense of a game that appears tremendously opaque. I am only interested in discerning how people conclude that they have some control over game outcomes. Can we say with Julio that the outcomes we receive are plausible consequences of our decisions? I want to believe they are, but I'm still a 2k7 agnostic.

[Vinatieri for Prez]

I also appreciate JR's thoughtful response. And as I have said before, it's a great game. I think JR has added a few points we can't forget about which also exist in real life. Some guys are system players - you only have to look at FA busts in the NFL that happen all the time. And we also tend to forget the multitude of factors that can affect performance (cohesion, chemistry, gameplan, coaching skills, etc.)

With all that said, then, it means that with the limited feedback that we do get in the game, judging player ability by stats starts to become a pointless task -- meaning the more you offer the additional factors as reasons why guys may not perform as expected, the more impossible it is to predict a player's potential for success on another team. And since you can't actually watch a guy play, then you are left solely with ratings - which arguable aren't totally accurate to begin with.

[ReiperXHC]

If the bars are an automatic verification for performance on the field, where's the challenge or the surprise of the overachiever or underachiever? If you can predict the outcome of each of your players on every play there is no replay value in my opinion.

Not for the bars to be an automatic verification. But an explanation of what the bars are "supposed" to mean. For example. In the help file a list of all bars and what each one represents...specifically.

Something like.

Zone Coverage - This players projected ability to handle zone coverage situations. (but have it more specifically stating if it means when "he's" in zone or when the "team" is in a zone coverage line up)

or

Zone Coverage - This players projected ability to play well in zone coverage defensive schemes. (This would mean more as to which schemes you're using...much more specific)

At the end of the list something could me mentioned about the actual scout error or the fact that the player "practices well but becomes nervous and inconsistent in real world situations." etc...

Nobody's asking for "this" means your player will only do "that" type of thing. We only want a nudge in the right direction.

And I believe others are also looking for a way to diagnose why certain things are working and why certain things aren't.

In the real world, when a player is doing poorly, a coach "usually" can figure out why, even if he thought origionally that the player would be good. This all makes sense to me. There could be better ways to implement it.

Now as to Jim's ability to make a great game and me buying it? I'm sold already. It took me a long time to find a good football game that has realistic enough stats to be plausible. (Yeah, I think Madden's "sim" engine could borrow some concepts from him) I'd definately buy more of his games....and I probably will. But I hope and would like to think that Jim sees all this and would like to implement it himself.

Sometimes we think that something as simple as "adding this stat" or "doing that with the sim engine" is simple work, but even things that may seem simple looking from the outside may take a long time to implement, especially without bugs and the other commen problems associated with writing programs. (not that I'm a programmer) But I'd like to think that it's not just as simple as sticking it in there.

Okay, I'll end my post here, I have a bad habit of putting up long drawn out posts...he he...I do the same thing when I'm actually talking to someone...I never shut up.

-----shutting up.

-Reip

[flair1234]

Just noticed last night that the help file on the update addresses this question.

Seems to back the line that common sense rules the day. So for instance in a cover2 bump and run; the safetys zone rating and the CBs and LBs B&R ratings would come heavily into play.

There is some jargon however to the effect of the way the ratings work is complex so some of the other factors would come into play.

BTW, I think one of the reasons this answer is important is because it allows you to really save money on quality back ups and borderline starters. For instance if you are going to run a predominantly B&R scheme, you can sign relatively one dimensional corners without incurring a big performance problem or paying for attributes that are not as important.

[Stevebsfan]

Nobody's asking for "this" means your player will only do "that" type of thing. We only want a nudge in the right direction.

And I believe others are also looking for a way to diagnose why certain things are working and why certain things aren't.

This is what i'm looking for. I'm not looking for a walk through on how to create the most dominant team in my league, but I do agree he should offer more on the defensive gameplanning side of it.

If a real coach gets a guy who is fantastic at reading offensive routes, strong with the bump and run style of defense, but poor at covering his zone, he has the ability to put that player in positions to maximize his strong points and minimize his weak points. There are still tons of other factors involved to make a team successful on defense, or even that player successful. For all we know, the scouting report on your player was off and your player isn't as good with bump and run or m2m as it seemed initially.

Only way to really find out though is to put them in pure bump and run situations and see how they do.

Gameplanning defense in this game is like trying to figure out an offense when your only options are the formation you want to use and how often you want to run/pass. Why am I allowed to tell my RB whether or not to run inside or outside the LT, or how often I want my QB to throw screen passes, but I have so little control over what's going on with my defense. I really wish Jim would clear some stuff up.

Someone also mentioned how there are other teams in leagues that happen to have good seasons year after year and it's probably because they stick with whatever works. Well, maybe it's because they figured out some inside tidbit about defensive gameplanning which should be public knowledge and they dominate that way. I think leagues would be more balanced the more information we had on how the game works. Then the better GM's can work on managing the cap, making good draft picks, making good trades, and managing the roster in order to have a successful franchise.

I mean this is front office football, not coaches corner football. The gameplanning aspect shouldn't be so cryptic where an avg football team with an avg head coach and coaching staff in game can go 14-2 because the GM has inner working knowledge about how gameplanning goes.

Just my 2c

[Ben E Lou]

I agree with much of your post. I do, however, want to comment on one thing that's interesting here...
I mean this is front office football, not coaches corner football.I found my receipt for Front Office Football the other day. It is dated November 17, 1998. Point being, that's a name Jim came up with 9 years ago. Since then, the game has evolved, in good part because of community input. And a significant part of that input has been requesting more coaching features--especially since MP was added. Yeah, it's still called "Front Office" Football, but the fact of the matter is that part of success in the game is on the "coaching" side of things. You can't really turn an 8-8 team into 14-2, as you said, but the "coaching" side of things does make a difference, and should. If you couldn't improve/hurt your team with your "coach" decisions, those features shouldn't be in the game, because that would just make them a waste of time.

That being said, let me reiterate that in this particular case, I would agree that we need to know *how* to use our players. For example, in real life, a coach or GM knows whether his weakside corner in the strong-zone, weak-man is playing tight bump coverage or loose man coverage. In real life, a coach or GM knows what kind of players and defensive schemes are best suited to stop a team that likes to throw a lot of 10-to-15-yard passes. FOF's documentation doesn't really answer those questions.

[Stevebsfan]

Yea, I see that point and I do agree with it. I'm not one who is strictly looking for a GM only type of game. I'd just like a game where I feel like i'm in completely control of my team from Monday through Saturday, giving my team the best chance it can at winning without actually having a hands on approach.

On the game planning side of the game, i'd like to be able to get a little more detailed w/o getting overwhelming. By overwhelming, I mean giving the option of what to do on 1st, 2nd, 3rd and 4th down from the 1, 2, 3, 4-5, 6-7, 8-10, 11-16, or 17+ yard lines while choosing between run, short pass or long pass on an interface that has to be the sloppiest way possible. I like the fine tuning, I hate the interface. I'm not a game designer, so I couldn't even begin to suggest how to make that more presentable, but there has to be a better way than 96 boxes and no way to make mass adjustments to any (or lock any).

Then the adjustment screen is even more clunky. 160 drop down menus all bunched together to adjust whether my team is leading by 3 points at my own 7 or my own 6 seems a bit extreme. Stuff like that is why you need a real quality head coach in game and just rely on his in game decisions.

Speaking of coach. What exactly are they for? On the help screen they spend a tiny bit talking about what they do, but then spend the rest talking about the clunky coach signing process (which btw is still missing a key re-sign option). Is a head coach only required to have the highlighted red sections, or should they be good in all areas? Does a head coach with high QB skill and an Off-Coord with high QB skills mean your QB is going to what? Live up to his potential? Go beyond his potential? Get hot gf's? Drive fast cars? Why are the top 4 penalized teams in my league have coaches with 'Good' to 'Excellent' discipline ratings? Do coaches really matter at all?

If not, or if they matter very little, then that means the game planning really falls on our shoulders and we're either left in the dark with defensive game planning, or overwhelmed by tons of minor options/tweaks that aren't really going to have a huge impact in the game. Most of those tweaks are variable situations anyway. In a 7-3 game, facing a team that can barely move the ball because their QB was injured early in the game, a coach may be prone to punting it towards the end of the game instead of kicking 45+ yard FG's. However, if it's 30-26, getting those 3 points may be worth the gamble.

I love a lot of aspects about the game, but the game planning side has always been on the annoying side, only to get 10x worse in 07. Instead of getting more detailed and easier to use, it got more confusing, cryptic and overwhelming. I really hope Jim does something about the game planning pages instead of just burying slight tips of information deep inside the help files.

[Yoda]

I agree. In the PFL, as a newbie and looking at everyone's records, the same teams seems to win. Do they understand how the game "sims" better than everyone else? I doubt it. I bet you guys find something that works and then stick with it instead of trying to beat the system by trying to predict the outcomes of games based on what you "feel" should have happened.

Well, looking at my team in the PFL, you'll notice I've gone up and down.

But, as for as gameplanning goes, there are a couple of things I take into consideration.

First of all, is the team. What is the build/make-up of the team? Right now, in several leagues (PFL, eNFL, FFL), I have offenses that are producing really well. And they are all similar in someways.

But for defense, I am at a complete loss. I play against a team that does like 70% of their passes short, and I make my defense mainly BnR, and get torched. I'll shut down the leagues best passing teams in the playoffs (Hi Shard ;) ), and get killed by a high school girls flag football team.

What I would like to see, is this.

1. The gamelog to tell you who is being DC'ed and when.
2. For the best WR to be BEST WR, the one with the highest OVR.
3. Specific Team defensive game planning. I want Deion Sanders to ALWAYS cover Jerry Rice, I don't care where he lines up.

See, if I set it to DC each WR 100%, then it should do that.

[flair1234]

Well, looking at my team in the PFL, you'll notice I've gone up and down.

But, as for as gameplanning goes, there are a couple of things I take into consideration.

First of all, is the team. What is the build/make-up of the team? Right now, in several leagues (PFL, eNFL, FFL), I have offenses that are producing really well. And they are all similar in someways.

But for defense, I am at a complete loss. I play against a team that does like 70% of their passes short, and I make my defense mainly BnR, and get torched. I'll shut down the leagues best passing teams in the playoffs (Hi Shard ;) ), and get killed by a high school girls flag football team.

What I would like to see, is this.

1. The gamelog to tell you who is being DC'ed and when.
2. For the best WR to be BEST WR, the one with the highest OVR.
3. Specific Team defensive game planning. I want Deion Sanders to ALWAYS cover Jerry Rice, I don't care where he lines up.

See, if I set it to DC each WR 100%, then it should do that.

One thing I have found just from single player is, it seems to me that the game starts penalising you when you become one dimensional.

For my coverage options, blitzing, and double coverage I will rarely go above 50%. I read somewhere that the teams offensive coordinator makes adjustments if you do something too often. I switched my SP and am switching my MP team out of BnR, into the man schemes, because then I have more options.

Another no no that I have discovered is you can't seem to do alot of double coverage and alot of blitzing. I have started making many of these decisions based on the QB.... what is the use in sending the blitz if he has a high sense rush... why double cover heavily if he has a high read defense.

I have some parameters I have been using and have been getting better results with my SP team. We will have to see in MP.

BTW, I don't do any testing, but I do watch the game which tips you off here and their.

[Yoda]

I am a little different. If my team is a solid blitzing team with good Dline and LB's, I don't care who the QB is, I am going to be blitzing and rushing him. The good QB's are going to get their yardage anyway.

The 2015 season in the PFL, I won the bowl game. My top WR's were all injured and out for the playoffs. My highest rated one was a 39/39. I knew it was going to be my defense to carry me (and Carson Palmer playing god-like helped). We went in and went through 3 of the top passing attacks in the league. And each time, I shut them down, because I went with the strength of my defense. Even though they had solid skills in SR and RD, I did what the strengths of my players were.

I don't gameplan (defensively) around what I see their bars are, I look at what my defense is good at, and what their tendancies are as an offense.

But either way, I am blitzing. ;)

[Dave Hansen]

I don't gameplan...call all my plays myself. Only have about 3 and half seasons under my belt...so my 2c:

* I've learned even with stength in say Band R coverage, I have to mix in MTM and Zone. As others point out, the AI is good at spotting trends in coverage, blitzing, AR/AP. This can be turned into your advantage by remembering what you have called in similiar situtions and doing a 180.

One of my favorites is to show a blitz or two during an AI possession. On 3rd and 6, I won't blitz, play MTM, watch the AI throw a screen pass to the FB that goes for 3 yards. I set up a very deep pass or two by running on first down from a 2 TE form , have the AI adjust to AR and send the FL on a deep fly.

The enjoyment for me is to be HTH with a good AI and equal talent...then having to out think/adjust/take play calling risks in order to win. To me that takes place on the sidelines with a headset calling each specific play vs. going the gameplan route. It also allows me to have a HC with strong motivational, discipline, injury avoidance talents.

I find it easier to defend against teams with great wideouts than teams that utilize the TE's and RB/FB...particularly around my goal line. My last two training camps I have scheduled 4 hours for defense. I find INT's go up a little, blitzing to be a little stronger. But the biggest payback I've noticed is the AI's 3rd down success has dropped from 42% to 35% over the last two years.. What I seem to notice is with this extra TC defense, is when the AI needs 7 yards, they get 5/6...when they need 13 they get 11. No scientific proof but a sense that is what was helped me.