Thoughts on Playbooks, Game-Planning, and Stats

[triplykely]

Background
Since I've started a new SP game/dynasty report and all of my testing has been focused on controlling and maximizing statistical output, I figured I'd open a new discussion on playbooks and game-planning to share what I know and maybe learn a thing or two.

Obvious caveat, I'm no expert and what I'm about to write could be wrong or off in some ways. While I did play an earlier version of FOF in the early 2000s I only recently rediscovered this, the best football sim ever. According to Steam I've spent over 1500 hours and a fair number of those have been running test simulations and I've yet to play MP. I typically don't keep detailed test records but have occasionally made notes. My general process is to draft a team, save it before training camp, and then make playbook changes and alternate saves. Due to some RNG in player progression, scout margin of error, etc., this doesn't lead to identical teams with each playbook iteration but it's close and you can always scrap it if a key player is drastically different. Then with each save I'll play with game-plans in an attempt to optimize stats in an effort to plan my game plans for SP games.

Assumptions
Given that FOF is a one man passion project of many years for Jim Gindin, and the recent discussion about the Quote of the Day, I'm inclined to believe 'Rex' plays a vital role in getting the most from your players. The amount of statistical analysis that's gone into FOF and the stat engine is beyond impressive and it seems straying too far from the mean is penalized. This doesn't mean you can't push things towards extremes, but generally speaking keeping this in mind leads to better efficiency on offense and consistency on defense.

Virtually every variable involved in FOF, if not every variable, is involved in stat generation. Taking advantage of chemistry, cohesion, coaching schemes, etc. all helps; but the playbook and game-plans are most important. Making sure to understand player ratings and scheme specific position requirements is critical to getting the most from your players.

Playbooks
Not to reinvent the wheel, but a lot of what I know I took and modified from this post from Corbes. With what's laid out in that thread, especially how to use the Play Creator and the wealth of real NFL stats on Sharp, anyone has a great foundation to build playbooks that increase expected wins. Keeping in mind the above assumptions regarding staying near the mean and Jim's statistical analysis, and with a goal of fluidity to game-planning in order to deal with injuries, utilizing the Guided Game-Plan Generator in conjunction with custom playbooks will yield the best results without being a burden to change on a game-plans weekly.

Given that the Generator uses down & distance ratios, broad pass distances, receiver targets normal/long, you need to craft a playbook that will best function with those inputs. Staying close to the mean on run/pass ratios, specific to down & distance, seems to lead to better stats. In fact letting Rex set the ratios seems to work well. If you have a QB with low INT bumping some distances up to long can help with TDs and YPA without costing you games to TOs. Since the target system is weighted you have to consider how many targets you'd want each player to get if everyone was set to 9. By doing this you ensure that as you tweak the weights you'll alter a players targets appropriately. You could make an even number of pass plays for each position but I find focusing on 2 WRs and a TE while having enough plays for everyone else to contribute is the best option. The game penalizes you for over targeting a player so utilizing everyone, even just a bit, helps. A quick note about team building, a receivers route running is the only individual rating displayed in the generator and is a strong factor in how Rex weights targets.

The point of this discussion is to maximize stats so its important to lay out your goals when creating a playbook. While I personally have been tweaking a universal playbook that works for any offense you can get nice results with Rex playbooks. For the sake of this, here's some basic thoughts I have on translating your offensive goals to creating a playbook.

Your game-plan will have somewhere around 55-65 plays give or take, so make sure you have enough plays of each type so you don't run into duplicate problems with the Generator when making game-plans. I've found 50 runs and 150 passes works well to play balanced or lean heavily either direction. In an alternate version of the playbook I use it only has 25 runs and they're all by the QB.

Theres a great section in the help file on the run game and how averages are determined. If you simply follow what Jim wrote you'll see a solid 4+ YPC average fairly easily, and if you get a little clever with formations and run direction you can push that even more. Misdirection runs are high risk/high reward and when used sparingly and coupled with a pass heavy offense can lead to big YPC averages with average RBs. Make sure to have a run play in every formation and look you have a pass play in to help avoid familiarity. Some formations are better for running and follow common sense (more TE/FB better for runs/play action).

The passing game gets a bit more complicated during playbook creation due to the number of receiving options and routes. Like regular runs and misdirection runs, there are varying degrees of variance to the routes. Slants are fairly easy completions with the right personnel but do pose increased INT risk as they go over the middle, yet they also have a nice YAC chance. Screen passes have fairly high variance but are relatively low risk and high percentage. Out routes are also low risk and high percentage but don't net the greatest YAC. Curls and Comebacks are nice routes to use to keep a nice YPA as are medium passes in general so long as you protect your QB.

Naturally as you go deeper in the route tree you, especially over the middle, you increase the likelihood of INTs. To help counter this I typically include an extra blocker for all medium pass plays (none for short and screen) and always for deep and long (sometimes two but rarely). There is a benefit to thinking about the available coverages when designing/editing plays. Using real NFL concepts like floods, mesh or pick plays, etc. will help with efficiency and the generator typically uses them; although you may wish to change the concept of a specific play.

Another note on route combinations, thinking about and including check down options is critical. Make sure a player or two are available underneath on plays designed to throw deep in addition to having an extra blocker. Doing this will help avoid sacks, increase completion percentage and therefore YPA. Generally speaking a good number of short passes and screens, with a few medium pass and couple long and deep passes, per receiver is crucial to maximizing YPA by both maintaining a high completion percentage and average pass distance.

Because there is a near endless list of things to consider, another is your formation use and balance within the playbook. Again assuming you will use the Generator to game-plan, it's important to keep real NFL tendencies in mind to avoid defensive familiarity. The 113 is the most common formation with the 122 a distant 2nd. All other formations are essentially gimmicks but you can use those gimmicks for big plays. Deep play action passes from running formations after establishing the run can lead to big plays and the reverse is true of runs from passing formations. You can make unique offenses that heavily focus on a less used formation but generally speaking its easier leaning on the 113 and 122.

In the Generator you have pass distance inputs for down & distance as well as targets so it's important to balance your overall pass plays with your offensive goal. I have the ratios of distances worked out before creating plays so that when set to normal I will have a strong YPA, but can push it with the right QB by going to long.

Game-Planning
Because the Generator uses weights for target pass distance and pass distance is also included in the run/pass ratios I generally game-plan with the target distance weights the same. Rex will typically have the long weight one or two below the normal weight which leads to a lower YPA as each receivers available plays becomes skewed to shorter passes. Sometimes this may be to your advantage as you may have two otherwise equally rated receivers with one better on deep passes.

As mentioned above, staying close to the mean for run/pass ratios specific to down & distance is important. 1st and Possession and 1st Earned can be different, and both seem to get a slight boost by favoring the pass a touch, but staying near 50% is ideal for both YPC and YPA. The more you tilt your overall play balance towards run or pass the more you will improve the other. Throwing the ball 65% of the time will yield a lower YPA and higher YPC than throwing the ball 55% of the time.

You don't need to have the highest number of plays for your OC, as long as you're over 60 and followed the playbook info won't run into familiarity issues or any/many duplicate issues. If you want you can tweak the game-plan generated by moving/changing plays to increase redzone passing for example, but generally I find by building the right playbook you can get great stats without doing this.

Quick note on all passing game-plans. If you make a playbook with 25 runs (minimum required) and 175 passes, you can set the Generator to 15/0, 5/0, and 10/0 for the relevant downs to get a nice start to the game-plan. You will have to manually swap a few situational runs to passes but this is the quickest way I've found to have a fluid all passing game-plan that takes into account injuries.

Most of my testing has been on offense as thats where the player has the most control of the micro. On defense I've been able to create specialized defenses but found better consistency with using Rex to generate a new game-plan and the blitz weights.

I haven't dove into tweaking the game-plan settings too much, but aside from Rex'ing the FG distance I've found changing 'Go to Clock Draining Offense when leading with' to 1 minute instead of 4 minutes maximizes scoring and improves consistency.

Players and Important Ratings
Because efficiency is the ultimate goal arguably the most important player rating is Route Running and Getting Downfield for WRs/TEs. Theres good discussions on ratings but if you focus on these 2 only, and don't pick high OVR players with low RR or GD, the Generator really works wonders in controlling stats.

OL should lean towards the blocking of your scheme. Get run blockers if you want to pound the rock. You don't need studs to put up good numbers but one or two really help, especially LT to limit sacks and C for the run game.

Make sure to read the help file for the defensive scheme you run and move your players to best fit that scheme. Staying on top of weight training and paying attention to prototypical players for each position will maximize what stats your players generate. Creating a synergy with your offense seems to be the best way to allow Rex'ing the defensive game-plan to work. If you have a pass heavy offense build a strong pass defense, pound the rock then stuff the run.

QB is the most important/least important player. It feels like you need a stud QB but really strong receivers is the most important to putting up good numbers and averaging over 8 YPA. In order to break 9 YPA consistently you do need a stud at QB, and generally the stud QBs avoid INTs better as well allowing you to change Generator inputs to long.

Final Thoughts
That's all a bit ramble-y but its what I've done/used to consistently average over 4 YPC and 8 YPA with years as high as 10+ YPA and 6/7 YPC. It be great if any of the veteran players have any thoughts or info to add to this. I tried to keep this somewhat organized and readable but know its a bit of a mess to read. Again, I'm no expert, but hopefully any of this helps you with enjoying FOF.

[Ternvig]

Wow - enjoyed reading it!

Thank you for taking the time to write this.

[triplykely]

You bet, figured it may help and/or spark a conversation. I might add to this by going further into how I tweak a playbook based on the stats I get.

[Landshark44]

thanks for taking the time.....

can you explain how you set the sliders in the "guided gameplan"?

for instance....

the guy you want most targeted gets a 9, i'd imagine.... but what do the other guys get? 7? 5?

and guys i don't want catching the ball at all.. FB 1?

i like the post, though, and would like to learn some more about it..

[triplykely]

Originally Posted by Landshark44

thanks for taking the time.....

can you explain how you set the sliders in the "guided gameplan"?

for instance....

the guy you want most targeted gets a 9, i'd imagine.... but what do the other guys get? 7? 5?

and guys i don't want catching the ball at all.. FB 1?

i like the post, though, and would like to learn some more about it..

So the Guided Gameplan takes a snapshot of your playbook based on the inputs. Hypothetically (for easy math) you have 10 plays targeting your FL, SE, TE, and 5 plays for your R, T, HB, and 0 plays for the FB.

If you set every receiver to 9 and throw 30 pass in a game your FL, SE, and TE would receive around 6-7 targets each and your R, T, and HB would see around 3 each.

A lot of people don't get the FB involved in the passing game, and the best way to do this is to not make any plays where the FB is primary target. I still set them at 1 when this is my goal but its kind of redundant to do.

In the link I put in the OP theres a good breakdown on building a playbook based on real NFL teams that really helps making a custom book in general. When you set out to make a playbook figure out how many of your passes will go to each receiver. Hypothetically you want 75% of your passes going to your FL, SE, and TE split evenly, so 25% each. Maybe you've got a good pass catching RB you give 10% to, 10% to the slot WR, and 5% to the TE2.

Multiply the percentages by how many pass plays you put in your playbook and you will see roughly the distribution you want with everyone at 9. Normally what I do is set everyone to 9 and then back down who I want to limit. I will normally do 3 tiers with one or two players at 9, another one or two at 7, and the rest at 5 - but this obviously is flexible and dependent on what you want to do. Hope that clarifies a bit.

[Hammer]

Great post. Your understanding is very good as far as I am concerned and you come across as a veteran.

I'm inclined to believe 'Rex' plays a vital role in getting the most from your players

As far as giving you an indictor of balance, sure it is worth paying attention to. But there is a lot of room to do it better than Rex on both Offense and Defense. I know you go on to say "this doesn't mean you can't push things towards extremes". Well yes you can. In fact IMO the very best gameplanners out there throw Rex out the window and do their own thing. It is complex but there is a lot of room in the game to do that. Equally there is a lot of room to mess it all up and end up worse than Rex. It seems like your thoughts are well developed and make a lot of sense. I bet the more you tinker and the further you develop your own style the less Rex will seem a smart role model.

I haven't thrown a pass to a RB or FB in probably 100 seasons of MP. I have been spoilt with WRs and TEs and wouldn't rule it out, but I think it illustrates moving away from Rex can work as I have done very well. My targets are currently around 40% TE, 30% WR1, 30% WR2. Just 3 primaries.

In fact Defensively I would say Rex is downright dumb and useless. The confusing bit is that defensive play calling matters very little, although it does play a part. I think many players get frustrated when they don't see the impact of calls. The impacts of making the right calls are subtle, but they are there. Rex just throws random shit in to the gameplan on D.

I think a video is round the corner. Much like the previous one it is a subject which can be talked about for hours and beyond the scope of my post.

The QB vs WR debate rolls on. I currently have a run first team 52% run last time I looked on a 22 game winning streak in an incredibly competitive MP league. There are many ways to win.

I still believe a QB is more important than a WR. You can DC a WR and shut him out pretty good. Yes the position group drives the yardage totals more so, but interceptions, sacks and fumbles raise the influence of the QB. The QB is just 1 guy though. That 2nd or 3rd WR or TE no doubt over powers the influence of the 1 QB.

Once again great material. I agree with the vast majority of what you said, but I guess going over that doesn't further the conversation.

[triplykely]

Originally Posted by Hammer

As far as giving you an indictor of balance, sure it is worth paying attention to. But there is a lot of room to do it better than Rex on both Offense and Defense. I know you go on to say "this doesn't mean you can't push things towards extremes". Well yes you can. In fact IMO the very best gameplanners out there throw Rex out the window and do their own thing. It is complex but there is a lot of room in the game to do that. Equally there is a lot of room to mess it all up and end up worse than Rex. It seems like your thoughts are well developed and make a lot of sense. I bet the more you tinker and the further you develop your own style the less Rex will seem a smart role model.

You're certainly right, and I may have overplayed what I think the importance of Rex is. Once thing I've noticed with Rex is that on offense there seems to be less variance simulating multiple seasons with the same team and defensive game-plan than manually setting the game-plan. As I write this though I realize that's probably more a result of poor plans by me opposed to good plans by Rex. I definitely agree about pushing away from Rex, I can of alluded to it but one of my fascinations is Tzachs all passing game-plans. One of the playbooks I have is specifically for this and works fairly well (not as good as his - I'm missing aspects of chemistry/cohesion/something) with most every QB. I think part of my thinking that Rex has less variance is my preference for passing over running. Rex will make a more balanced game-plan and that will help with offensive efficiency and decrease variance.

I haven't thrown a pass to a RB or FB in probably 100 seasons of MP. I have been spoilt with WRs and TEs and wouldn't rule it out, but I think it illustrates moving away from Rex can work as I have done very well. My targets are currently around 40% TE, 30% WR1, 30% WR2. Just 3 primaries.

In reading posts from other MP coaches like you, I think this is the norm, and it makes sense as you're focused on maintaining less players than I typically do in SP - obviously player talent is spread thinner in MP as humans beat the AI easily. It may be worth exploring the FB in MP as I believe there is value in the position given how cheap it is to have one or two studs at the position. In the playbook I use almost exclusively I can routinely get 8 YPT from the FB. Playing SP its fun for me to develop multiple fictional careers so I prefer to target multiple players, but with what you bring up that's most likely tainting my biases on the importance of spreading the ball around.

In fact Defensively I would say Rex is downright dumb and useless. The confusing bit is that defensive play calling matters very little, although it does play a part. I think many players get frustrated when they don't see the impact of calls. The impacts of making the right calls are subtle, but they are there. Rex just throws random shit in to the gameplan on D.

This is probably similar to my view on Rex on offense having less variance. Now what you say, and what I observe - but what goes against the help file - is that the Rex does random plays for defense. The help file makes it sees like its based on your team and the opponent but it sure looks random. Again when simulating multiple seasons with the same team I've noticed Rex game-plans seem to have lower variance but again I'm probably making mistakes on my manually crafted defensive game-plans. Its where I've spent the least time so I'm sure it presents the biggest opportunity.

The QB vs WR debate rolls on. I currently have a run first team 52% run last time I looked on a 22 game winning streak in an incredibly competitive MP league. There are many ways to win.

I still believe a QB is more important than a WR. You can DC a WR and shut him out pretty good. Yes the position group drives the yardage totals more so, but interceptions, sacks and fumbles raise the influence of the QB. The QB is just 1 guy though. That 2nd or 3rd WR or TE no doubt over powers the influence of the 1 QB.

Definitely agree on many ways to win, your approach of a balanced offense, from my testing, has the lowest variance and best YPC, YPA, and expected wins. You can push the needle more by passing, but as you mention increase the risk of TOs. The only times I've broke 9/10 YPA for a season was with no more than 55% passing. Part of that issue might be my desire to throw to ball to more targets than MP coaches, who given the competition should be trusted over my SP experience.

Once again great material. I agree with the vast majority of what you said, but I guess going over that doesn't further the conversation.

Thanks for taking the time to spell out some thoughts and add input. I was hoping by writing it someone would offer a better understanding of some of my anecdotal conclusions. The quote of the day is either giving me insight to the game or driving me nuts thinking I'm uncovering the under-workings of the sim engine but who knows.

[Hammer]

It may be worth exploring the FB in MP as I believe there is value in the position given how cheap it is to have one or two studs at the position.

I absolutely agree with you here. A couple of guys in the RZB are throwing to stud FBs. It isn't a fashion, but I think it could become one shortly. I only play in the 1 league, and have been stacked at WR and TE for a while. I haven't had to go this route I guess. For me it is something I would look at when the great TE and WR are not accessible. Sooner or later I think that will happen. Although I expect these stud FBs will begin to creep up in popularity. As you say these guys are also more cap friendly. I have only seen out and out stud FBs perform well in the passing game to this point.

I don't know if you have seen the video's produced by a couple of the RZB GM's. They tend to go on for a couple of hours and at times dig deep in to the game. Experienced GMs giving their thoughts. Some are league specific, but others more game specific. You sound like a guy who might enjoy this stuff with your appetite to understand the game. I think you are well on the way in any case, I bet you would do well in MP.

Moving on to the topic of result variance. I have a team which can run or pass well. Last season I ran heavy during the regular season, as I believe it gives less result variance. We tend to use the term "dice rolls", being exposed to luck if you will. I think running the ball exposes you less to luck. If you are the better team you are more likely to win if you have a strong running game. If you pass the ball the variance is higher, you may destroy your opponent, but equally the effect of luck is increased and the dice may roll against you. I could be wrong but that is how I see it.

Last season we went 15-1, yet only scored more than 40 twice. Scoring 41 and 42. Our points dif was +299, pretty good but not earth shattering. We played what I felt was low turnover football which gave the dice minimal opportunity to mess things up for us.

Moving on to the playoffs we were faced with better teams. What struck me is that they were teams with strong run defenses. We switched the gameplan and played much more aggressively and scored 68, 57 and 44. I think the switch perhaps gave us an advantage as we had been running hard and heavy all year, but I am not convinced the gameplan was "better". I think we probably caught some good dice, the luck went for us. I have since tested that same gameplan and it certainly has a tendency to crash and burn. My gut says we would not have gone 15-1 in the regular season with it as the dice would of swang against us. Our RBs don't tend to fumble much as all so a bad roll might be a loss of 2. Against the pass it may be a pick, sack or fumble.

With the same line of thinking when playing a better team I think airing it out may make more sense, put the game in the hands of the dice. I do think the run is easier to stop. When you are playing great run defenses I don't think my regular season 52% running gameplan is going to succeed. On the flip side I am more confident passing the ball and getting good results against a great pass defense. The tools don't seem there to stop the pass the way you can the run.

[triplykely]

Originally Posted by Hammer

I absolutely agree with you here. A couple of guys in the RZB are throwing to stud FBs. It isn't a fashion, but I think it could become one shortly. I only play in the 1 league, and have been stacked at WR and TE for a while. I haven't had to go this route I guess. For me it is something I would look at when the great TE and WR are not accessible. Sooner or later I think that will happen. Although I expect these stud FBs will begin to creep up in popularity. As you say these guys are also more cap friendly. I have only seen out and out stud FBs perform well in the passing game to this point.

This is a big part of my thinking with making my playbook varied enough to utilize any position, but like everything it has its pros and cons. With a stud TE/WRs it definitely isn't going to swing things much/at all throwing to the FB (I also never have running success with them). One of the late round positions I target sometimes are undersized TEs with good RR&GD because they normally see a bump when moved to FB.

I don't know if you have seen the video's produced by a couple of the RZB GM's. They tend to go on for a couple of hours and at times dig deep in to the game. Experienced GMs giving their thoughts. Some are league specific, but others more game specific. You sound like a guy who might enjoy this stuff with your appetite to understand the game. I think you are well on the way in any case, I bet you would do well in MP.

I may or may not have, I watched another video that was suggested after one of the ones posted here but am not sure which league it was. It would be something I'd probably enjoy, will search YT for RZB. I'm tempted to try a MP league but I play FOF in spurts so I've just not committed. No real reason not to.

Moving on to the topic of result variance. I have a team which can run or pass well. Last season I ran heavy during the regular season, as I believe it gives less result variance. We tend to use the term "dice rolls", being exposed to luck if you will. I think running the ball exposes you less to luck. If you are the better team you are more likely to win if you have a strong running game. If you pass the ball the variance is higher, you may destroy your opponent, but equally the effect of luck is increased and the dice may roll against you. I could be wrong but that is how I see it.

Exactly how I see it. One thing I have done in SP, which obviously allows you to get way more talent then in MP, is utilize screens and other high percentage passes to keep YAC high while minimizing INTs/sacks from throwing deep. It is definitely easier to keep YAC high and avoid bad dice with a run game, but as you point out you won't blow the lid off of people as much.

With the same line of thinking when playing a better team I think airing it out may make more sense, put the game in the hands of the dice. I do think the run is easier to stop. When you are playing great run defenses I don't think my regular season 52% running gameplan is going to succeed. On the flip side I am more confident passing the ball and getting good results against a great pass defense. The tools don't seem there to stop the pass the way you can the run.

Its like the Quote of the Day regarding a stud RB not finding success against 8 in the box. You can definitely make a defense than doesn't allow high YPC pretty easily but its much harder to stop the pass. In case I join a league I don't want to tip my hand too much, but I've found focusing on the secondary and pass rush while passing somewhere around 55-60% passing, the right down specific run/pass ratio, and a good mix of play action deep passes as well as screens to high RR/GD players is the best of both worlds. If you face a run team you tend to outscore them and if you face a passing team you tend to generate sacks and TOs.

[Hammer]

One last thing I will say. I have put 100% run or pass in situations with no ill effect. I think it is the overall balance rather than situational balance the AI is worried about.

It would also be interesting to see if Avg Att went up in Tzach 's all pass Offense when runs were mixed in, or whether ideal balancing can nerf that out. Would have to be careful to swop out a balanced variety of passing lengths.

In a standard GP I agree rushing, or perhaps successful rushing, raises passing Avg Att. Not sure if the quality of the rushing plays in.

[QuikSand]

Thanks to you both for the big info drops here. I have not yet committed to digesting it, but I plan to. I don't fee I have good command of offensive gameplans, and have run out of patience quickly when fiddling with them on my own. I'd be delighted if I can glean enough here to get back into it a bit, and feel like it's time well spent.

[triplykely]

That's interesting Hammer, and I may just be confirming my bias with how I interpret my testing. I've never broke 9 YPA for a season when passing more than 60% but have when passing less, even with mediocre QBs. Obviously there are a number of reasons why this could be the case that have nothing to do with run/pass ratio. The other aspect of my testing is the down & distance ratios. My thinking behind them is sort of a sequenced drive with the goal a new 1st down. If you pass on 1st down you have a higher likelihood of 0 yards and to face a 2nd and 10 but also a bigger chance of gaining a new 1st down. Following this thinking and testing some ideas its seems like the ratios of down & distance affect YPC and YPA based on how real NFL defenses normally act in those situations. In the NFL On 3rd and long may see a draw play, but probably not any other type of run, because a draw plays off of what the defense is expecting with a pass. In the help file Jim mentions situational football and although anecdotal I've had success improving YPC and YPA by tweaking down & distance ratios while leaving the overall run/pass balance the same.

And Quik, thanks to you and the other guys for doing those videos. It gave me a lot of great insight to start experimenting with. Given some of the outlier seasons I've seen, I think a strong chemistry team with the right playbook/game-plan could put up silly numbers. I've only focused on chemistry to avoid conflicts, never to make sure I have affinities, so I know I'm missing out on YPA and YPC.

Since I accidentally deleted my new SP game I may start a new one and specifically draft for chemistry while using my thoughts on playbooks to start a testing thread with the goal maximizing yards/play and wins.