Help me refine a realistic progression system

[triplechin]

I want to get a solid basis for a progression system that is realistic, (and may be easy to code?) and use it to write a re-hauled franchise blog (DeuceDouglas for the love of God please clear your inbox or message me on reddit ) that all sim gamers could unify over and point to EA whenever they ask what we want. Maybe it ends up being a waste of time but it is the first time that I am not purchasing Madden so I am going to make an effort, I want to enjoy running and managing a football team, while also being able to play the games.

As another year of neglected franchise has gone by, I have seen more people on board for getting rid of XP. It just does not allow for reality and the more they build features off of it, the more time they are wasting both theirs and ours. This is really targeted at sim and realism from a coach/GM/owner standpoint, so if that does not interest you then I do not know if you would enjoy this. Give your input anyway if you please.

Here is a rough draft of a better system than XP.

http://forums.operationsports.com/fo...ssion-cfm.html

I have plans to have things such as positional coaches, coordinators, practice focus, vet leadership, and other things make an impact on the progression chances. That will be detailed in a rough draft of my franchise blog though.


Here are some concerns that I could not quite get a good concept formulated.

When to progress/how often. Initially, I imagined end of season progression. However, I am trying to emulate real life. Ex, Tom Brady. He was drafted as a 60? Then the next year, Bledsoe got hurt and Brady proved he was worthy of the job, lets say 80 something? But if Brady progresses from a 60 to an 80 in Madden, you can make decisions in your offseason based on that. I was imagining player overalls could be hidden, and get revealed through the preseason, with reports from scouts and coaches helping you make decisions in the meantime? Not sure how to tackle this situation, so definitely looking for help on this aspect. Same goes for an unexpected drop. Perhaps the year the Broncos won a superbowl, if they had known Peyton would drop from a 90 to a 78 or whatever, they may have made a move for a QB in that previous offseason. This happens often in real life, perhaps not as extreme as either of these cases, but I am looking for help in this aspect.

Other suggestions are welcomed as well!

[deu22ces]

Re: Help me refine a realistic progression system

I would suggest the vast majority of progression happen in the pre-season (80%+). IMO, and experience, this is when real skill and mental ratings progress occurs for players. I'd even argue that the only in-season progression should be AWR/PRC based on snap count/playing time (game experience). For the in-season AWR/PRC, I could see a system where snap count % in relation to total team snap count gives incremental bumps; the user could still customize whether 'redeeming' these occurs weekly/4 weeks/end season, etc. I'd give a premium to players who get 'starter' snap counts and regress down to a base line for role players and practice squad players.

** slight side note: I'd like to see a Special Teams position rating that took into account the players running, blocking, and tackling ability since many 'roster bubble' players make the 53 on that ability alone.** All of this would give some much needed meaning to pre-season.

The example you referenced with Brady; his "progression" occurred b4 taking over the starting job. No one outside the NE organization knew it yet tho. His scheme and playbook knowledge happened with repetition in practice and the film room. His accuracy and pocket movements could only be advanced in practice and pre-season games prior to getting the starting gig. IMO, player progression b/c of stats is the exact opposite of reality. Players' stats are b/c of prior work and effort put in in order to 'progress'. Other factors are at play as well, but definitely not stats.

The unexpected drop issue I'd have to think more on. Some kind of randomization factor needs to be in there tho. Not sure how that works in the coding world. B/c with your P.Manning example, he had 3 great season b4 the cliff drop. You could argue that there were definite 'hints' if you watched his film the year prior and the lower body injury was a big factor too. Maybe there's some clue in those factors.

Appreciate the effort you've put in already and what looks like continued effort you're about to continue with. Hopefully my 2 cents is worth slightly more to you.

[timhere1970]

Re: Help me refine a realistic progression system

I would suggest the vast majority of progression happen in the pre-season (80%+). IMO, and experience, this is when real skill and mental ratings progress occurs for players. I'd even argue that the only in-season progression should be AWR/PRC based on snap count/playing time (game experience). For the in-season AWR/PRC, I could see a system where snap count % in relation to total team snap count gives incremental bumps; the user could still customize whether 'redeeming' these occurs weekly/4 weeks/end season, etc. I'd give a premium to players who get 'starter' snap counts and regress down to a base line for role players and practice squad players.

** slight side note: I'd like to see a Special Teams position rating that took into account the players running, blocking, and tackling ability since many 'roster bubble' players make the 53 on that ability alone.** All of this would give some much needed meaning to pre-season.

The example you referenced with Brady; his "progression" occurred b4 taking over the starting job. No one outside the NE organization knew it yet tho. His scheme and playbook knowledge happened with repetition in practice and the film room. His accuracy and pocket movements could only be advanced in practice and pre-season games prior to getting the starting gig. IMO, player progression b/c of stats is the exact opposite of reality. Players' stats are b/c of prior work and effort put in in order to 'progress'. Other factors are at play as well, but definitely not stats.

The unexpected drop issue I'd have to think more on. Some kind of randomization factor needs to be in there tho. Not sure how that works in the coding world. B/c with your P.Manning example, he had 3 great season b4 the cliff drop. You could argue that there were definite 'hints' if you watched his film the year prior and the lower body injury was a big factor too. Maybe there's some clue in those factors.

Appreciate the effort you've put in already and what looks like continued effort you're about to continue with. Hopefully my 2 cents is worth slightly more to you.

I would add a trait so that all players do not progress the same with the same amount of snaps. There are are lot of players in the nfl who stick around year after year due to their smarts while players with suerior athletic ability are cut. Madden does a poor job of reflecting this because it is easy to improve awareness and make anybody smart. IRL only a small percentage of those physically gifted prospects ever become nfl starters.

[adembroski]

Re: Help me refine a realistic progression system

OK, what I'm about to spill out of my head onto this page is a vague rememberance of a system I designed in detail and excel-prototyped when I was at EA. The much more complete and detailed document is probably sitting on a sharepoint at EA right now, and given the number of my basic ideas that went into Madden 17, I'm guessing they're aware of it. That said, I don't know if I put it on sharepoint, so I might have just had it on my personal drive. If that's the case, it's gone.

Anyways, being that I'm just remembering this, I'm not going to plug in a bunch of numbers, just give an overview of the basic concept.

A few prerequisites for this system to work, which were also designed but never saw light. I'm not going to outline these systems, only give you the basics so I can refer to them within the progression system I designed.

1. Full coaching staffs, including 3-levels of staff for each player; a specialist (Strength and Conditioning Coach) or position coach, a coordinator, and a head coach.

2. A somewhat NFL 2k5 practice/training system wherein specific drills are scheduled.

3. A character flag system which marks players as having specific traits (as opposed to the existing trait system. I used flag as an alternative term because trait was already in use). This would be similar to the trait system from NFL Head Coach 09. Specific flags tied to progression might be "Great Work Ethic" or "Hitch in Throwing Motion."

OK, now the progression system.

1. Each attribute at each position is assigned an annual potential improvement amount. This is the most that a given attribute can improve for a given position in one year.

2. There are a number of factors that play into progression for a specific attribute; the coaching ability in descending order of importance (for instance, the position coach is 2x as important as the coordinator, who is 2x as important as the head coach), the breakdown of time spent on drills that develop that attribute, AGE (of course) the progression and work ethic related flags (this could use Motor and Development traits if no flag system were implemented), and a modifier for events, such as injuries.

Each category produces a number between 1 and 100, and these should be tuned so the mean is 50. Progression of the attribute in question = weighted average of all progression factors (as a percentage) * progression potential.

This locks everything into a range so it's predictable. It opens a lot of options for the user to have input into how each player progresses. It's simple, straightforward, easy to understand, easy to communicate to the user, and makes sense in a real world way.

Now, what about the non-work ethic related flags? Why do we have a hitch in our throwing motion, and how do we deal with that.

Every coach has the bandwidth to focus on one guy within his specialty arena. Your QB has a hitch in his throwing motion, which translates in-game to a slower release and occasional poor throws regardless of his accuracy rating.

Say you're the 49ers and you drafted a 1st round QB with a hitch. You can assign, if you so choose, your head coach (Shannahan, a QB guru) to work specifically to eliminate that flag, thus eliminating the negative gameplay effects of it. You do this a week at a time during training camp, and at the end of the week, you find out whether you've succeeded. When a coach is focused on fixing one player's major problem, he does not apply his progression toward anybody else on the roster. So it's risk/reward. Are you willing to devote the time necessary to fix that QBs hitch, or live with it to continue building up the rest of your team.

There would be other ways to use this as well, if you want to focus progression in a specific direction. If you're rookie linebacker is especially bad in play recognition, maybe you focus your LB Coach on him for a week to get him up to speed at the expense of other players' progression.

Anyways, the basic idea is a mathematically sound system that maintains roster integrity all the way through 30 years of franchise mode, is easily manipulated without being abusable, realistic and sim-friendly, and has some tangible risk-reward to it.

[triplechin]

Re: Help me refine a realistic progression system

I would suggest the vast majority of progression happen in the pre-season (80%+). IMO, and experience, this is when real skill and mental ratings progress occurs for players. I'd even argue that the only in-season progression should be AWR/PRC based on snap count/playing time (game experience). For the in-season AWR/PRC, I could see a system where snap count % in relation to total team snap count gives incremental bumps; the user could still customize whether 'redeeming' these occurs weekly/4 weeks/end season, etc. I'd give a premium to players who get 'starter' snap counts and regress down to a base line for role players and practice squad players.

** slight side note: I'd like to see a Special Teams position rating that took into account the players running, blocking, and tackling ability since many 'roster bubble' players make the 53 on that ability alone.** All of this would give some much needed meaning to pre-season.

The example you referenced with Brady; his "progression" occurred b4 taking over the starting job. No one outside the NE organization knew it yet tho. His scheme and playbook knowledge happened with repetition in practice and the film room. His accuracy and pocket movements could only be advanced in practice and pre-season games prior to getting the starting gig. IMO, player progression b/c of stats is the exact opposite of reality. Players' stats are b/c of prior work and effort put in in order to 'progress'. Other factors are at play as well, but definitely not stats.

The unexpected drop issue I'd have to think more on. Some kind of randomization factor needs to be in there tho. Not sure how that works in the coding world. B/c with your P.Manning example, he had 3 great season b4 the cliff drop. You could argue that there were definite 'hints' if you watched his film the year prior and the lower body injury was a big factor too. Maybe there's some clue in those factors.

Appreciate the effort you've put in already and what looks like continued effort you're about to continue with. Hopefully my 2 cents is worth slightly more to you.

Thanks for the kind words. Yeah, during preseason makes some sense to me. It would be interesting to have certain attributes that could improve during the season, while most only improve in the offseason.

The main problem I am trying to tackle is from a gameplay standpoint. Season is over, it is free agency. Perhaps Blake Bortles is expected to take a big step after throwing 35 TDs so you dont pursue a QB, but then it turns out he stinks and now you are stuck with him for a year. At what point does it make sense from a gamer's standpoint of how much to know and when? Should scouts and coaches give you opinions on projections and expectations during the offseason, then his overall would surprisingly stagnate or dip in the preseason? Preseason seems like it makes the most sense, but definitely need to think on it more.

[triplechin]

Re: Help me refine a realistic progression system

Say you're the 49ers and you drafted a 1st round QB with a hitch. You can assign, if you so choose, your head coach (Shannahan, a QB guru) to work specifically to eliminate that flag, thus eliminating the negative gameplay effects of it. You do this a week at a time during training camp, and at the end of the week, you find out whether you've succeeded. When a coach is focused on fixing one player's major problem, he does not apply his progression toward anybody else on the roster. So it's risk/reward. Are you willing to devote the time necessary to fix that QBs hitch, or live with it to continue building up the rest of your team.

There would be other ways to use this as well, if you want to focus progression in a specific direction. If you're rookie linebacker is especially bad in play recognition, maybe you focus your LB Coach on him for a week to get him up to speed at the expense of other players' progression.

I like this idea of players having issues that need to be worked on, and may or may not get better at, but it requires more attention which will show by other parts of the roster being neglected. Interesting

[SyncereBlackout]

Re: Help me refine a realistic progression system

OK, what I'm about to spill out of my head onto this page is a vague rememberance of a system I designed in detail and excel-prototyped when I was at EA. The much more complete and detailed document is probably sitting on a sharepoint at EA right now, and given the number of my basic ideas that went into Madden 17, I'm guessing they're aware of it. That said, I don't know if I put it on sharepoint, so I might have just had it on my personal drive. If that's the case, it's gone.



Anyways, being that I'm just remembering this, I'm not going to plug in a bunch of numbers, just give an overview of the basic concept.



A few prerequisites for this system to work, which were also designed but never saw light. I'm not going to outline these systems, only give you the basics so I can refer to them within the progression system I designed.



1. Full coaching staffs, including 3-levels of staff for each player; a specialist (Strength and Conditioning Coach) or position coach, a coordinator, and a head coach.



2. A somewhat NFL 2k5 practice/training system wherein specific drills are scheduled.



3. A character flag system which marks players as having specific traits (as opposed to the existing trait system. I used flag as an alternative term because trait was already in use). This would be similar to the trait system from NFL Head Coach 09. Specific flags tied to progression might be "Great Work Ethic" or "Hitch in Throwing Motion."



OK, now the progression system.



1. Each attribute at each position is assigned an annual potential improvement amount. This is the most that a given attribute can improve for a given position in one year.



2. There are a number of factors that play into progression for a specific attribute; the coaching ability in descending order of importance (for instance, the position coach is 2x as important as the coordinator, who is 2x as important as the head coach), the breakdown of time spent on drills that develop that attribute, AGE (of course) the progression and work ethic related flags (this could use Motor and Development traits if no flag system were implemented), and a modifier for events, such as injuries.



Each category produces a number between 1 and 100, and these should be tuned so the mean is 50. Progression of the attribute in question = weighted average of all progression factors (as a percentage) * progression potential.



This locks everything into a range so it's predictable. It opens a lot of options for the user to have input into how each player progresses. It's simple, straightforward, easy to understand, easy to communicate to the user, and makes sense in a real world way.



Now, what about the non-work ethic related flags? Why do we have a hitch in our throwing motion, and how do we deal with that.



Every coach has the bandwidth to focus on one guy within his specialty arena. Your QB has a hitch in his throwing motion, which translates in-game to a slower release and occasional poor throws regardless of his accuracy rating.



Say you're the 49ers and you drafted a 1st round QB with a hitch. You can assign, if you so choose, your head coach (Shannahan, a QB guru) to work specifically to eliminate that flag, thus eliminating the negative gameplay effects of it. You do this a week at a time during training camp, and at the end of the week, you find out whether you've succeeded. When a coach is focused on fixing one player's major problem, he does not apply his progression toward anybody else on the roster. So it's risk/reward. Are you willing to devote the time necessary to fix that QBs hitch, or live with it to continue building up the rest of your team.



There would be other ways to use this as well, if you want to focus progression in a specific direction. If you're rookie linebacker is especially bad in play recognition, maybe you focus your LB Coach on him for a week to get him up to speed at the expense of other players' progression.



Anyways, the basic idea is a mathematically sound system that maintains roster integrity all the way through 30 years of franchise mode, is easily manipulated without being abusable, realistic and sim-friendly, and has some tangible risk-reward to it.

How is progression potential determined?

How are assistant coaches and coordinators evaluated? (Frequency and based on what factors)

What attributes , if any, cannot be improved upon


Sent from my iPhone using Tapatalk

[adembroski]

Re: Help me refine a realistic progression system

How is progression potential determined?

How are assistant coaches and coordinators evaluated? (Frequency and based on what factors)

What attributes , if any, cannot be improved upon


Sent from my iPhone using Tapatalk

1. Tuning. That's not variable. It's a league-wide baseline designed to ensure league roster stability.

2. Completely different design that would require its own thread.

3. Any and all. I'm a firm believer in designing consiquence over restriction. Wanna waste your head coach's time turning JJ Watt into a QB, have at it. When he gets to 60 THP and refuses to resign with your dumb ***, live with your mistake.

Sent from my SM-G930P using Operation Sports mobile app