My Analysis of How The EA NCAA Game Uses Roster Ratings

[Art01]

I have written a document in Word to explain how I believe the EA NCAA Game Series uses Roster Ratings to impact the game. I consider the document to be too lengthy to post directly, so I have attached it as a Word ".doc" file, which can be read in almost any version of Word.

Comments are welcome, but not necessary. I hope all who read my analysis will be inspired to take a fresh look at the game and find new ways to enjoy working with and playing the game.

[khaliib]

Good stuff Art.

This applies to Madden as well with a few extra trinkets.

For NCAA I'm thinking of using this and making...

Top Tier schools:
Alabama type teams, were player ratings will use the scale of 99-127.

Mid Tier schools: (most of schools)
Arizona type teams, were player ratings will use the scale of 40-99.

Lower Tier schools: (many Mid-Major & below)
New Mexico St type teams, were player ratings will use the scale of 0-40.


Want to see what impact this would have on Gameplay.

Also, coach's ratings may fall in line with this concept also and impact in some manner.


Good stuff!!!

[Art01]

Thanks, Khaliib:

I will be interested to hear about your game-play results. I predict that your Mid-Tier and Low-Tier teams will not perform very well - especially the Low-Tier group because of the very low ratings.

Also, I predict that you'll see more Momentum impact when playing with the lower-rated teams because of the increased ratings Head Room.

For example, I think Punters and Kickers performance will be very poor with any Kick Accuracy and Kick Power ratings lower than 50.

One of the problems that exists with using ratings greater than 99 is that the software cannot accurately calculate OVR ratings, either for a team or for a player. So, it is difficult to compare teams when using ratings above 99. However, I consider that negative to be far less significant than the positives associated with using the increased ratings.

If you want to see what my edits look like, my latest Roster is in my 360 Locker (ArtD01).

[khaliib]

Yeah with the K/P it will be difficult to utilize this, but with other positions, the positional ratings may be able to use this.

The issue is that NCAA is built off of Madden, which is suppose to represent the best of the best, which fits the 40-99 scale because there is very little difference at this level.

But because the levels of abilities are much larger in College, the 40-99 scale that is used produces exactly what we hate, which is very little difference from a New Mexico St team/player vs an Alabama team/player.

Ex: Man/Zone Coverage
A 0-40 rating is possible to use.

Although the 99-127 scale leaves little to no upside room, my question is will it force the AI teams/players to perform at this range consistently?

There is an Hot n Cold mechanism in place for in-game performance.
If this is a piece that's used to create momentum swings, even a cold Alabama/player should perform better than my lower Tier school I'm controlling.

Another issue is that the User control negates most of the AI challenge because we negate any player ratings when we take control of a player, which outside Speed, allows a player to perform beyond the Team/player ratings.
That's why we're able to take a team with players with lower ratings than the AI counterpart and consistently beat them.

Just wondering with such a gap and positional ratings so low, if it will help negate the Hum player super seeding ratings.

It may not work, but just a thought to see.

Also, I've suggested to others to revisit some of the older versions of NCAA/Madden that we didn't have the Editor for ('09, '10 and '11) as these games performed much better than later releases but had some type of issue on the Offensive or Defensive side of the ball.

We shall see...

[Art01]

Because I only play on Coach Mode, I have no experience with how taking control of a player affects the game. But, I suspect that it makes a considerable difference - especially if you are controlling the QB.

Based upon my analysis, if I were going to play the game by taking control of one or more players, I think I would settle on one team as my team and then try to find a set of player ratings for my team which would equalize things on the field.

For example, I might experiment with the QB Accuracy rating for my QB to determine what setting makes it "pleasingly difficult" for me to complete passes at my current skill level - then be prepared to alter that setting as my skill level increased.

If I determined that the QB Accuracy rating did not matter when I am in control, then I would want to rethink my strategy about how to set up teams to defend against me. Perhaps dramatic reductions in Pass Blocking, or large increases in certain defensive player ratings to reduce the amount of time available for me to scan the field or get off a throw.

I realize that this type of manipulation might seem to be "unrealistic", but then so is being able consistently dominate as a player.

In my write up, I intentionally stayed away from a specific mention of "AI" in the game. But, it is my belief that there is very little of what I will call "Interactive AI" - meaning AI that adjusts to game situations. For example, I just finished a game by Coaching Miami of Ohio against my beloved Crimson Tide. As expected, the Tide won handily. But, with less than a minute to go and with Bama in possession inside Miami's 10 yard line, there was no QB Kneel Down called by the CPU. Bama was running plays until time ran out. Also, Bama was still throwing the ball with very little time left.

I'll look forward to the results of your experiments.

[doncropper]

Art,

Thanks for this exceptional piece of analysis. I believe you have uncovered some very meaningful & immensely beneficial aspects of the logic engine of this game.

These findings and your analysis of them certainly will enhance my enjoyment of this game in the coming years.

I feel, with your approval of course, a need to copy and paste your great write up here in this forum so others can benefit from it as I have. If you disapprove please let me know and I will delete it immediately:

--------------------------------------

March 22, 2014 – Arthur Daniels, Page 1

MY VIEWS ON HOW NCAA14 USES ROSTER RATINGS TO MANAGE GAMEPLAY

Introduction:

After an extensive period of experimenting with Roster edits and playing Play Now type games in Coach Mode, I believe that I have developed an understanding of how the software in NCAA14 utilizes Roster Ratings to manipulate game play for User versus CPU types of games. My experiments and observations have resulted from use of Xanathol’s Generic Editor to implement Roster edits which are not possible to implement by editing Rosters within the game. I am presenting these views to the NCAA14 Community for consideration and further study and comment.

The Key Ratings Component:

Editing Rosters within the game limits all Roster Rating settings to a maximum of 99. However, editing Rosters externally via the Generic Editor permits the setting of a given Roster Rating to a maximum of 127. This 28-point differential is the key to understanding how the software uses Roster Ratings to manipulate game play.

The Impact of “Hidden” Roster Ratings Upon Momentum and Game Play:

For a number of years in the NCAA Football Series, much attention has been given to the impact of Momentum within the game. Momentum in real football is a fact of life and should play a role in a computer simulation. The questions have always been: How much should Momentum impact the simulation and how should Momentum be implemented and controlled?

Another topic of considerable discussion has been the relative impact of playing the game at different Difficulty Level settings and how the software uses Difficulty Level settings to manipulate game play.

I believe that the software uses the differential between the Static Roster Ratings (which are limited to a maximum of 99 when setting within the game) and the actual available maximum Roster Ratings of 127 to implement Momentum control and as a key component in affecting game play at the different Difficulty Levels. These higher ratings can be implemented into the Static Roster via the Editor and saved into the game.

Static Rosters, Dynasty Rosters and Dynamic Game Rosters:

Regardless of the mode of play (whether it be Play Now, Coach, Single Season or Dynasty), at game startup, the Static Roster is loaded into what I will call a “Dynamic Game Roster” file. For single Play Now type games, this Dynamic Game Roster is unloaded at the conclusion of the game and none of the Dynamic Roster changes, which might occur during the game, are saved to the Static Roster

For Single Season and Dynasty modes, there is a separate version of the Static Roster, which is initially a copy of the Static Roster loaded into the Dynasty database and which is saved as part of the Single Season or Dynasty modes (the “Dynasty Roster”). The Dynasty Roster is loaded into the Dynamic Game Roster file at the start of each game or game simulation in Single Season or Dynasty Mode. Certain changes to the Dynamic Game Roster which occur during a Dynasty-type game (such as Injuries) are retained and implemented into the Dynasty Roster at the close of a game or simulation. If we look at the Dynasty Roster during the course of a season, we can see the ratings changes which have been saved and note such events as Injuries. Ratings changes during the season are intended to reflect player-performance changes as the season progresses.

It is the manipulation of the Dynamic Game Roster that is the subject of this analysis.

(Continued next page)

March 22, 2014 – Arthur Daniels, Page 2

How The Software Manipulates the Dynamic Game Roster to Impact Momentum During Game Play:

I believe that the software utilizes the spread between the Static or Dynasty Roster Ratings and the Dynamic Game Roster available Ratings to simulate the effects of Momentum during the course of a game. The software expects that Static Ratings top out at 99, so with the available maximum ratings of 127, the software has a significant amount of “Head Room” to implement in-game ratings boosts to simulate increased Momentum. The software can also implement ratings reductions to simulate poor in-game performance (such as erratic QB play).

The sophistication of the software’s ability during game play to alter these ratings, coupled with the number and type of ratings changed and the amount of change implemented, will combine to simulate the level and frequency of Momentum swings during the course of the game.

The Relationship Between Dynamic Roster Ratings Manipulations, Difficulty Levels and Game Play:

I believe that the software also utilizes Dynamic Game Roster ratings manipulations to implement the differentiation in so-called play difficulty when playing the game with the various Difficulty Settings (Freshman, Varsity, All American or Heisman). I believe that the degree of manipulation is related to the Difficulty Level Settings as follows:

1. Freshman: User Dynamic Game Roster Ratings are given a significant in-game boost when the Static or Dynamic Roster is loaded at start-up and, as the game progresses, further performance-based alterations are made but generally favoring the User over the CPU.
2. Varsity: Same as Freshman, except that the Dynamic Ratings impacts are not as significant as in Freshman.
3. All American: Perhaps the “neutral” setting, wherein ratings manipulations are nearly equalized for both CPU and User teams, but still slightly favoring the User (particularly near the end of a close game).
4. Heisman: Perhaps some initial boosts to CPU ratings during start-up, but strongly favoring the CPU for in-game boosts (to the point at which the CPU becomes essentially unstoppable on both Offense and Defense). Probably employs both ratings boosts and ratings reductions to a significant degree.

For User versus CPU games, I also believe that the software will manipulate CPU play-calling to a greater or lesser extent, depending upon the Difficulty Setting. Such play-calling manipulation would result in a greater number of “dumb” CPU play calls during the course of a game at Freshman level, but a greater number of “smart” CPU play calls during a Heisman game.

Combining the effects of Dynamic Game Roster in-game ratings changes with “dumb” or “smart” CPU play-calling is, in my opinion, the methodology employed by the software to simulate play at different difficulty levels and to simulate the effects of Momentum.

The Impact of Player Sliders:

I believe that the impact of adjusting Player Sliders is to make initial mass alterations of certain key player ratings to the Dynamic Game Roster after it is loaded during start-up, but before game play begins. Because there are a small number of Player Slider types, when compared to the much larger array of Player Ratings types, the inevitable result of manipulating the game via the Player Sliders will be much less satisfactory than making ratings changes. And, because we cannot see the impacts of Player Slider adjustments, we can only speculate regarding their relative impact and any possible interactions between different Player Slider adjustments.

Note that the Setting of 50 is a Neutral setting, with increases or decreases in performance based upon the degree of departure from the neutral setting.

For almost all of my testing, I have left Player Sliders at the Default 50 Setting. Penalty Sliders are not a part of this analysis, although I set mine at slightly above 50.

(Continued next page)
March 22, 2014 – Arthur Daniels, Page 3

Validating My Observations and Conclusions:

I have used the Generic Editor to make significant Static Roster ratings changes and I have conducted extensive testing in the single-game Play-Now and Coach game modes to observe the results of these changes. Here are a couple of suggestions for modifications you can make to test these conclusions for yourself.

1. Using a Roster with normal player ratings (no player rating exceeding 99), start a Play-Now type game with Offense and Defense Difficulties set to Heisman. Choose two evenly-matched teams. As the game progresses, switch the Difficulty Levels to Freshman and notice the nearly immediate impact on game-play. This test is especially revealing in the case in which the CPU team seems to have captured significant Momentum – changing from Heisman to Freshman will almost immediately dramatically lower the CPU’s Momentum.

2. Create a “Test Roster”. Use the Generic Editor to set all ratings to 127 for any two teams. Save this Roster back into the game and play a User-CPU game between these two teams at Heisman Difficulty. Notice the fantastic performance by both teams, but, more importantly, notice the much-reduced impact of Momentum swings during the course of the game – demonstrating the lack of Head Room available for performance boosts during the game. If you also play a game at Freshman Level with these same two teams, you will probably see the results of the software reducing the dynamic ratings for the CPU team during the game.

Summary Comments and Suggestions:

1. I believe that every serious fan of NCAA14 should learn to use the Generic Editor in order to be able to modify the Rosters Ratings to achieve the most preferable game-play results.
2. Playing single games in Play-Now modes will yield the most satisfying results after extensive Static Roster edits because the base Static Roster remains unchanged no matter what happens within the game.
3. Playing a Single Season Dynasty is more likely to yield satisfying results with modified Static Rosters than playing multiple-season Dynasties. After several years in a multiple-season Dynasty, there are so many Roster player changes that the impacts of the initial edits will be severely reduced. Furthermore, it is likely that all new players brought into the game via Recruiting will have their respective ratings limited to a maximum of 99, thus necessitating repeated pre-season external editing to implement ratings greater than 99.
4. Use the Generic Editor to make Static Roster ratings changes and leave the Player Sliders at 50.
5. Learn to use the Export/Import feature of the Generic Editor to move a Roster into Excel for even more editing flexibility.
6. Advantages of using the Editor should also apply equally to User versus User games, although I have not yet tested my modifications in User versus User games.
7. Find the Difficulty Levels which suit your game play style and desires. For my conservative play style, my current preferences are Offense at Heisman and Defense at All American. I would like to use Heisman, but I have reduced Kick Power significantly in my Static Roster and, at Heisman Defense level, the CPU attempts long Field Goals which cannot be made with my Kick Power setting. Also, the CPU Offense Momentum increases at Heisman Defense can become unplayable at times.

Above all, enjoy the game and enjoy the journey of using these tools to enhance game performance. Within the OS Forums, there are on-going discussions about the Editors and using them to modify elements of the game beyond just manipulating Roster Player Ratings.

END

[herropreese]

Quote: Originally Posted by doncropper

Art, Thanks for this exceptional piece of analysis. I believe you have uncovered some very meaningful & immensely beneficial aspects of the logic engine of this game. These findings and your analysis of them certainly will enhance my enjoyment of this game in the coming years. I feel, with your approval of course, a need to copy and paste your great write up here in this forum so others can benefit from it as I have. If you disapprove please let me know and I will delete it immediately: -------------------------------------- March 22, 2014 – Arthur Daniels, Page 1 MY VIEWS ON HOW NCAA14 USES ROSTER RATINGS TO MANAGE GAMEPLAY Introduction: After an extensive period of experimenting with Roster edits and playing Play Now type games in Coach Mode, I believe that I have developed an understanding of how the software in NCAA14 utilizes Roster Ratings to manipulate game play for User versus CPU types of games. My experiments and observations have resulted from use of Xanathol’s Generic Editor to implement Roster edits which are not possible to implement by editing Rosters within the game. I am presenting these views to the NCAA14 Community for consideration and further study and comment. The Key Ratings Component: Editing Rosters within the game limits all Roster Rating settings to a maximum of 99. However, editing Rosters externally via the Generic Editor permits the setting of a given Roster Rating to a maximum of 127. This 28-point differential is the key to understanding how the software uses Roster Ratings to manipulate game play. The Impact of “Hidden” Roster Ratings Upon Momentum and Game Play: For a number of years in the NCAA Football Series, much attention has been given to the impact of Momentum within the game. Momentum in real football is a fact of life and should play a role in a computer simulation. The questions have always been: How much should Momentum impact the simulation and how should Momentum be implemented and controlled? Another topic of considerable discussion has been the relative impact of playing the game at different Difficulty Level settings and how the software uses Difficulty Level settings to manipulate game play. I believe that the software uses the differential between the Static Roster Ratings (which are limited to a maximum of 99 when setting within the game) and the actual available maximum Roster Ratings of 127 to implement Momentum control and as a key component in affecting game play at the different Difficulty Levels. These higher ratings can be implemented into the Static Roster via the Editor and saved into the game. Static Rosters, Dynasty Rosters and Dynamic Game Rosters: Regardless of the mode of play (whether it be Play Now, Coach, Single Season or Dynasty), at game startup, the Static Roster is loaded into what I will call a “Dynamic Game Roster” file. For single Play Now type games, this Dynamic Game Roster is unloaded at the conclusion of the game and none of the Dynamic Roster changes, which might occur during the game, are saved to the Static Roster For Single Season and Dynasty modes, there is a separate version of the Static Roster, which is initially a copy of the Static Roster loaded into the Dynasty database and which is saved as part of the Single Season or Dynasty modes (the “Dynasty Roster”). The Dynasty Roster is loaded into the Dynamic Game Roster file at the start of each game or game simulation in Single Season or Dynasty Mode. Certain changes to the Dynamic Game Roster which occur during a Dynasty-type game (such as Injuries) are retained and implemented into the Dynasty Roster at the close of a game or simulation. If we look at the Dynasty Roster during the course of a season, we can see the ratings changes which have been saved and note such events as Injuries. Ratings changes during the season are intended to reflect player-performance changes as the season progresses. It is the manipulation of the Dynamic Game Roster that is the subject of this analysis. (Continued next page) March 22, 2014 – Arthur Daniels, Page 2 How The Software Manipulates the Dynamic Game Roster to Impact Momentum During Game Play: I believe that the software utilizes the spread between the Static or Dynasty Roster Ratings and the Dynamic Game Roster available Ratings to simulate the effects of Momentum during the course of a game. The software expects that Static Ratings top out at 99, so with the available maximum ratings of 127, the software has a significant amount of “Head Room” to implement in-game ratings boosts to simulate increased Momentum. The software can also implement ratings reductions to simulate poor in-game performance (such as erratic QB play). The sophistication of the software’s ability during game play to alter these ratings, coupled with the number and type of ratings changed and the amount of change implemented, will combine to simulate the level and frequency of Momentum swings during the course of the game. The Relationship Between Dynamic Roster Ratings Manipulations, Difficulty Levels and Game Play: I believe that the software also utilizes Dynamic Game Roster ratings manipulations to implement the differentiation in so-called play difficulty when playing the game with the various Difficulty Settings (Freshman, Varsity, All American or Heisman). I believe that the degree of manipulation is related to the Difficulty Level Settings as follows: 1. Freshman: User Dynamic Game Roster Ratings are given a significant in-game boost when the Static or Dynamic Roster is loaded at start-up and, as the game progresses, further performance-based alterations are made but generally favoring the User over the CPU. 2. Varsity: Same as Freshman, except that the Dynamic Ratings impacts are not as significant as in Freshman. 3. All American: Perhaps the “neutral” setting, wherein ratings manipulations are nearly equalized for both CPU and User teams, but still slightly favoring the User (particularly near the end of a close game). 4. Heisman: Perhaps some initial boosts to CPU ratings during start-up, but strongly favoring the CPU for in-game boosts (to the point at which the CPU becomes essentially unstoppable on both Offense and Defense). Probably employs both ratings boosts and ratings reductions to a significant degree. For User versus CPU games, I also believe that the software will manipulate CPU play-calling to a greater or lesser extent, depending upon the Difficulty Setting. Such play-calling manipulation would result in a greater number of “dumb” CPU play calls during the course of a game at Freshman level, but a greater number of “smart” CPU play calls during a Heisman game. Combining the effects of Dynamic Game Roster in-game ratings changes with “dumb” or “smart” CPU play-calling is, in my opinion, the methodology employed by the software to simulate play at different difficulty levels and to simulate the effects of Momentum. The Impact of Player Sliders: I believe that the impact of adjusting Player Sliders is to make initial mass alterations of certain key player ratings to the Dynamic Game Roster after it is loaded during start-up, but before game play begins. Because there are a small number of Player Slider types, when compared to the much larger array of Player Ratings types, the inevitable result of manipulating the game via the Player Sliders will be much less satisfactory than making ratings changes. And, because we cannot see the impacts of Player Slider adjustments, we can only speculate regarding their relative impact and any possible interactions between different Player Slider adjustments. Note that the Setting of 50 is a Neutral setting, with increases or decreases in performance based upon the degree of departure from the neutral setting. For almost all of my testing, I have left Player Sliders at the Default 50 Setting. Penalty Sliders are not a part of this analysis, although I set mine at slightly above 50. (Continued next page) March 22, 2014 – Arthur Daniels, Page 3 Validating My Observations and Conclusions: I have used the Generic Editor to make significant Static Roster ratings changes and I have conducted extensive testing in the single-game Play-Now and Coach game modes to observe the results of these changes. Here are a couple of suggestions for modifications you can make to test these conclusions for yourself. 1. Using a Roster with normal player ratings (no player rating exceeding 99), start a Play-Now type game with Offense and Defense Difficulties set to Heisman. Choose two evenly-matched teams. As the game progresses, switch the Difficulty Levels to Freshman and notice the nearly immediate impact on game-play. This test is especially revealing in the case in which the CPU team seems to have captured significant Momentum – changing from Heisman to Freshman will almost immediately dramatically lower the CPU’s Momentum. 2. Create a “Test Roster”. Use the Generic Editor to set all ratings to 127 for any two teams. Save this Roster back into the game and play a User-CPU game between these two teams at Heisman Difficulty. Notice the fantastic performance by both teams, but, more importantly, notice the much-reduced impact of Momentum swings during the course of the game – demonstrating the lack of Head Room available for performance boosts during the game. If you also play a game at Freshman Level with these same two teams, you will probably see the results of the software reducing the dynamic ratings for the CPU team during the game. Summary Comments and Suggestions: 1. I believe that every serious fan of NCAA14 should learn to use the Generic Editor in order to be able to modify the Rosters Ratings to achieve the most preferable game-play results. 2. Playing single games in Play-Now modes will yield the most satisfying results after extensive Static Roster edits because the base Static Roster remains unchanged no matter what happens within the game. 3. Playing a Single Season Dynasty is more likely to yield satisfying results with modified Static Rosters than playing multiple-season Dynasties. After several years in a multiple-season Dynasty, there are so many Roster player changes that the impacts of the initial edits will be severely reduced. Furthermore, it is likely that all new players brought into the game via Recruiting will have their respective ratings limited to a maximum of 99, thus necessitating repeated pre-season external editing to implement ratings greater than 99. 4. Use the Generic Editor to make Static Roster ratings changes and leave the Player Sliders at 50. 5. Learn to use the Export/Import feature of the Generic Editor to move a Roster into Excel for even more editing flexibility. 6. Advantages of using the Editor should also apply equally to User versus User games, although I have not yet tested my modifications in User versus User games. 7. Find the Difficulty Levels which suit your game play style and desires. For my conservative play style, my current preferences are Offense at Heisman and Defense at All American. I would like to use Heisman, but I have reduced Kick Power significantly in my Static Roster and, at Heisman Defense level, the CPU attempts long Field Goals which cannot be made with my Kick Power setting. Also, the CPU Offense Momentum increases at Heisman Defense can become unplayable at times. Above all, enjoy the game and enjoy the journey of using these tools to enhance game performance. Within the OS Forums, there are on-going discussions about the Editors and using them to modify elements of the game beyond just manipulating Roster Player Ratings. END

Could you post the file of your tommycoa/playmakers combined roster so we can port it to 360?

[Art01]

My modified Roster is a version of the Yukon46 Roster. It is a 360 Roster and is currently in my 360 Locker under gamer tag "ArtD01" . I do not own a PS system.