Donny Moore, the 'Madden Ratings Czar', Leaving EA

[vannwolfhawk]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

I can't speak 4 the you tube community in general but I am aware that the guys from the simstandard have actively promoted use of ur ratings in the game and I'm sure would have made Rex , Clint et Al well aware of them
The only reason I personally haven't used them is that they don't work well for CFM , as you have yourself stated.
I still feel that rather than 2 separate ratings scales a more viable *compromise * would be to use your ratings for individual attributes and modifiers to give EA it's desired level of OVR for marketing purposes

Agreed! This is best way right here and Dan already stated he had a system to use to get higher overalls with his ratings. Problem solved! Suits happy and vannwolfhawk happy! Whats the hold up?

[Rashad19]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

That's exactly the point is with realistic ratings it's the same game and people will still love it. It will just eliminate speed rb's being chased down from behind by a DL or a lb running sideline to sideline to make a play. It will separate guys (wr vs a cb) with either speed differential or route running versus man coverage as well as add agility to that list.

In past nba2k's it was a complaint of the community when chris Paul was on a fast break and Tim Duncan would speed past him. They spread out ratings so this wouldn't happen and it made fast breaks more realistic. That's all this would do is make players stand apart from one another. It allows teams to truly have mismatches and it's now a strategy for a user to exploit, take advantage of it, or protect it from a defensive side. For example I could now put Richard Sherman on a island and shade my safety help to the other side which is realistic. As it is now really the only thing that matters is speed to create separation and there is not much difference in a rb or wr as there is with a LB or DL. It's why we see some frustrating things in the game that just isn't realistic. There will still be guys who are faster than others and still have an advantage. The biggest thing to me is less skating and in years past the game always felt way to fast which made it look bad where you see morphing, skating, etc. I'm not saying that 100% go's away but it's night and day difference and still looks 100% better.

Bottom line is just like in FIFA and nba2k players will learn to adjust and I can't see anyone complaining with rosters with real stats and backed up data to support the ratings. Even the tourney players I would assume would like this as it makes the game a true chess match now. It makes teams and players stand apart. Yes it makes you play slightly differently but it's still the same madden game just with less bs you would see on the field. Isn't that ultimately what we all want? It's a big difference in fbg ratings but it's not like it changes the game as we know it. It just makes it play sim and eliminates unrealistic things happening in the game that would never happen irl. It's truly a no brainer.

I just hope EA gets this right and makes the change to lower ratings, spread them out, and uses real data for ALL there ratings. It also eliminates any argument any player, and fan could ever have with ratings like speed, acceleration, strength, jumping, etc as it go's off what they did in the combine or whatever testing they did.

I agree we don't need 2 rosters nor would they ever probably do this, but we do need 2 different game plays with a sim setting. Like someone said before it would be great to get a system to use averages in drafts for the roster you are using. But if they insist on staying with what they got why not have a 2nd roster ? I get all the valid points ggsimmons made. I totally get it. And I get that 95% of players don't even know what fbg rated rosters are. I look at the threads in roster section and it surprises me they don't have the views like some of the nba2k roster makers have. Maybe a lot of that has to do with draft classes messing up in cfm. But if fbg and dan and ea had something set up where any updates dan made to a player or on his site automatically would be available for users it would be pretty cool and I'm sure it would catch some steam as people learned about them.

All my posts say the same thing I just hope EA see's the light and go's in the right direction. If they were to rest the rosters there is no doubt in my mind they would be like wow this game plays different and I didn't realize how much ratings held this game back.

Excellent post!! At this point of my life I'm just extremely frustrated with the direction of the game I love so much. Madden has not pushed the envelope innovation wise in 10 years. There have been slight improvements but playing games like FIFA and NBA2K opened my eyes to how current generation gaming should feel. They are not perfect games but I can see the growth with each new iteration. What is the last madden that has truly blown us away??


To hear that individuals participated in Dan's meeting without having played the game with his updated ratings is like nails on a chalkboard. It screams of being unprepared to truly understand the capability of what your investigating. In 2015 we should not be screaming for ticker tapes, true pockets, flags on the fields, and players that are rated appropriately. All of these things have and can be done. As a community we need to stop making excuses as to why core basics of the game have not been implemented yet.

[Gman 18]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

Excellent post!! At this point of my life I'm just extremely frustrated with the direction of the game I love so much. Madden has not pushed the envelope innovation wise in 10 years. There have been slight improvements but playing games like FIFA and NBA2K opened my eyes to how current generation gaming should feel. They are not perfect games but I can see the growth with each new iteration. What is the last madden that has truly blown us away??


To hear that individuals participated in Dan's meeting without having played the game with his updated ratings is like nails on a chalkboard. It screams of being unprepared to truly understand the capability of what your investigating. In 2015 we should not be screaming for ticker tapes, true pockets, flags on the fields, and players that are rated appropriately. All of these things have and can be done. As a community we need to stop making excuses as to why core basics of the game have not been implemented yet.

Tellin' it like it is...


Sent from my iPhone using Tapatalk

[khaliib]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

They also need to rename some of their ratings to what they "actually" do during gameplay as appossed to just utilizing football termanality.

Pursuit is simply "Catch-up/Run Down" SPEED
- it's more about speed/acceleration of defensive players than actual pursuit angles.

Zone Coverage is "Ball Awareness" and has nothing to do with the ability to cover defined zones in the field.
- it's "Ball Reaction" on a pass no matter what type of coverage (Zone/Man) is called.
- this is the "Psych-DB" driver as it makes defenders react even before the ball has been released from the QB hands.
- this rating also helps kills "Downfield" throws by AI QB because QB ACC rating is programmed to only make throws to an opened WR and this rating affects the AI from attempting. (Zone Coverage + Strength rating for QB are the culprits for few downfield attempts by AI)

Jump is simply "Defensive Int" rating and has nothing to do with actual jumping ability.
- you will not see a defensive player in the Draft Classes come in rated lower than 75 for this reason.

Agility is "Momentum".
- more weight is applied to offensive players and is the reason why their momentum takes them out of bounds.
- defensive players are weighted differently (ie none) which is why they "turn-on-a-dime"
- also affects pulling OL pancake blocks on outside runs

Play Recognition is "Defensive Alertness" on runs.
- as soon as there is a ball carrier, EVERYONE turns to go make the tackle, even when their backs are to the play.

These are just a couple but part of the ratings confusion is just that, confusion between the football terms vs what they actually affect during gameplay.

[DCEBB2001]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

They also need to rename some of their ratings to what they "actually" do during gameplay as appossed to just utilizing football termanality.

Pursuit is simply "Catch-up/Run Down" SPEED
- it's more about speed/acceleration of defensive players than actual pursuit angles.

Zone Coverage is "Ball Awareness" and has nothing to do with the ability to cover defined zones in the field.
- it's "Ball Reaction" on a pass no matter what type of coverage (Zone/Man) is called.
- this is the "Psych-DB" driver as it makes defenders react even before the ball has been released from the QB hands.
- this rating also helps kills "Downfield" throws by AI QB because QB ACC rating is programmed to only make throws to an opened WR and this rating affects the AI from attempting. (Zone Coverage + Strength rating for QB are the culprits for few downfield attempts by AI)

Jump is simply "Defensive Int" rating and has nothing to do with actual jumping ability.
- you will not see a defensive player in the Draft Classes come in rated lower than 75 for this reason.

Agility is "Momentum".
- more weight is applied to offensive players and is the reason why their momentum takes them out of bounds.
- defensive players are weighted differently (ie none) which is why they "turn-on-a-dime"
- also affects pulling OL pancake blocks on outside runs

Play Recognition is "Defensive Alertness" on runs.
- as soon as there is a ball carrier, EVERYONE turns to go make the tackle, even when their backs are to the play.

These are just a couple but part of the ratings confusion is just that, confusion between the football terms vs what they actually affect during gameplay.

In all honesty, if Madden is going to use proper source material, like scouting data, they need to use the actual attribute names that scouts look for. Then, again just being totally honest here, they need to use the same grading system. They can keep the OVR on a 0-99 scale if they want, but in all reality, most of these grading systems are not on a 0-99 point scale. The one that drives my ratings is 0.0-5.0. BLESTO uses 2.40-1.00, and National uses 2.40-10.00. The 0.0-5.0 is best, in my opinion, because each tenth of a point (0.1) is essentially 2 points on Madden's scale. So, you can either use even numbers to represent this on a 0-100 points scale, OR you can use the 0.0-5.0 scale with increments of one tenth.

Something to also consider - the highest grade I have ever seen for any attribute was Junior Seau's "Football Sense" grade of 4.80 on this 0.0-5.0 scale. That would mean that the highest AWR grade available for any player as far back as this data goes is 96. That really says something. That means that even the scouts believed that the highest grade in any category at any time still was 0.2 points off from being perfect.

The lowest grade I have ever seen for the same attribute at any position is 0.1. That means, if we converted this straight to Madden, the lowest awareness a player had would be a 2. The average is about 1.7, or 34. The scouts really think that the difference between an average player's awareness and the player with the highest awareness (or football sense) ever is 62 Madden points. That is amazing, and just goes to show you how average some of these guys really are when it comes to football sense.

If you look at other attributes, you find the same thing - players are nowhere near as close to perfect (100) as EA would lead us to believe. Let's look at the ever controversial Arm Strength grade for QBs. The average grade for all players is 2.0. That would be an average THP rating of 40. Believe it or not, the low is 0.1, so some QB out there actually has a THP of 2. Yup, a 2. The highest ever has a THP of 3.8. That means that the highest QB THP in the game would only be a 76.

So what does all of this mean?

1. EA needs to modify their OVR formulas. QBs with average THPs of 40 would really hurt those OVR scores because 1/3 of the OVR is based on THP! That is WAY OFF of how the scouts grade the importance of arm strength. The trade-off in having accurate trait grades is having inaccurate OVR grades. EA needs to change this, and I have the formula to do it.

2. We need to make people understand that even the best trait of the best player ever at his position is not going to be perfect. If the highest grade ever found was 4.8 out of 5.0, then you know that getting to a 5.0 (100 in Madden) is essentially impossible.

3. We need to break the habit having the average NFL player being rated close to an 80 OVR. That just is not the case in reality. According to the data, the average NFL player or college prospect has an average overall grade of 2.60. That would be a 26 in Madden. The guys in the league or on the street as free agents are closer to an average grade of 3.70 or 37. Guys on rosters today have an average grade of 5.0 or 50. Using this scale, you can see just how elite these players are. Here is the distribution from my latest batch of data straight from the scouts:

9.0s: 4
8.0s: 13
7.0s: 64
6.0s: 293
5.0s: 879
4.0s: 1670
3.0s: 6908
2.0s: 8280
1.0s: 5636
0.0s: 2164

As you can see, there are fewer players rated 5.0+ than there are rated less than 1.0! The average grade for this batch: 2.61. THAT is how rare the elite talent is. Madden needs to adjust this and use the distribution to show just how elite the guys at the top are.


What I would want to do with EA is use this data to drive the attributes, overall grades, and the draft classes. I would use all of the historical combine and pro day data (going back to 1996 is what I have) to make sure that the physical attributes are on point for every position. Then, I would use the scouting data to get the proper distribution for each position-specific skill (attribute). All would be mapped out on the bell curve, so you can still get those elite players that are 3 standard deviations away from the mean for each position (the QB with a 3.8 arm strength and the DE with a 4.40 40 yard dash, etc). I have to be brought in to do this, however, even at the cost of massive backlash.

I will pose the same question to everyone here that I would pose to the development team. Are you trying to make an accurate game or not? Are you trying to mimic the NFL? It's an honest question. I hope that they can use this information to make the game better while also maintaining that competitive balance of USER v. USER and USER v. CPU.

[TheDominator273]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

Question for you Dan, if a QB has throw power that low do you feel the game accurately represents the velocity they throw and the distance they can throw as well?

[DCEBB2001]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

Question for you Dan, if a QB has throw power that low do you feel the game accurately represents the velocity they throw and the distance they can throw as well?

At present, it does not. That is why the ratings that are present on the FBG site are converted to the Madden rating ranges. The fact that I have to convert them at all is the issue here. To me, the engine would have to be reworked so that a QB with an Arm Strength grade of 5.0 would throw at a level of velocity that is unobtainable, but close to theoretically obtainable (4.9). Therefore, what I have to do instead, because of having to work inside of Madden's ratings "box" is convert the data I have into ratings that are usable in the game.

In this sense, a QB with an Arm Strength grade of 3.8 would be set to equal 99 in Madden instead of having a 5.0 being set to 100. The other issue is that since every player has a rating for every attribute (example: DEs having a THP rating) you have to consider all players and their positions, even though it is very unlikely that a DE will ever be graded for his arm strength/throwing power. To this effect, I have to pick a logical range within the game for each attribute to properly account for having players from other positions that have lesser talents in a particular skill or ability.

To EA's credit, and this is something that I do agree with them on, every player SHOULD be rated in every attribute to account for position changes. The user should be allowed to place his DE at QB and suffer the consequences of having a QB with an arm strength of 20 and an accuracy of 5. Realistic or not, we should have that option. If a GM/owner (maybe that guy down here in North Texas) really wants to move all his DEs to QBs, he can do so. It wouldn't be recommended, but he could have them do it. We should be able to do the same thing. That is why I am such a fan of doing away with the position change penalties - the penalty lies in placing a player at the wrong position and suffering dealing with his inadequate abilities for his new position. We don't need EA to penalize us again do we?

[mcadley]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

I wasn't a huge fan of this guy... being a Raiders fan, he made them a lot worse than they already were and are. Raider *****!

[MustangLife03]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

They definately need to update elusiveness to how the big scouters like pro football focus and the others do. Since madden throws elusiveness with how often/good they are at using juke/spin plus slippery. But pro football focus rates elusiveness on breaking out of tackles exclusively. Eddie lacey, jonathon stewart, lynch, mccoy, are the top most elusive guys from 2014 according to pff. Head fake needs to be added by juke and spin, head fake is real and happens.

But as far as gameplay and saying nba2k, etc are so much more advanced. How can Maddens gameplay have more features other than a updated ranking system. And more accurate penalties and sliders that need to be tweaked. Coach glass, all the plays aside from missing combo plays. I forgot the title of the plays teams run that have like 4 plays in 1. Option, hb handoff, and 2 pass routes. That needs to be implemented to the game. And maybe more realistic zone blocking and zone runs.

[capa]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

Great analysis and work DCEBB.

But you are correct. It will be a tough sell to get Madden to change the way their engine works in all their algorithms for individual player performance on the field. Not that it is a huge undertaking if designed and approached correctly...just that it will be a tough sell to get them to take this on I think.

They have an engine that "works" (in their opinion) with the current ratings system. We all know it is severely flawed and not realistic at all. But it works. Convincing them that there needs be such a huge design change will be a challenge.

I hope you can pull it off because this is the only way we will get the game finely tuned. It can't and won't happen with sliders no matter how they tweak them. I have always believed that the path to realism on the field is in the player ratings...for ALL sports games.

C

[DCEBB2001]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

They definately need to update elusiveness to how the big scouters like pro football focus and the others do. Since madden throws elusiveness with how often/good they are at using juke/spin plus slippery. But pro football focus rates elusiveness on breaking out of tackles exclusively. Eddie lacey, jonathon stewart, lynch, mccoy, are the top most elusive guys from 2014 according to pff. Head fake needs to be added by juke and spin, head fake is real and happens.

I agree that there is a better way to rate the ELU attribute, but PFF is not it. They are not scouts. Everything they do is performance based. They will not tell you what an individual is actually capable of because they only look at the result of the play. They look at production, not traits. I would not recommend using PFF for anything aside from assigning tendencies.

[Steel4Reel]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

I agree that there is a better way to rate the ELU attribute, but PFF is not it. They are not scouts. Everything they do is performance based. They will not tell you what an individual is actually capable of because they only look at the result of the play. They look at production, not traits. I would not recommend using PFF for anything aside from assigning tendencies.

Brilliantly put about PFF.

Too many people cite PFF as some kind of all knowing source. They arent. Their grades are often utter garbage and are often not a good reflection of how good/bad certain players were.

[DCEBB2001]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

Brilliantly put about PFF.

Too many people cite PFF as some kind of all knowing source. They arent. Their grades are often utter garbage and are often not a good reflection of how good/bad certain players were.

All you need to do is read their FAQ page. Lot's of stuff that should raise red flags in there. Here are some quotes:

"All of the data is collected in order to build a detailed picture of each player’s performance and production over the course of a season."

"We offer a different type of scouting, strictly based on performance and not technique or upside"

"We are looking for the result of that poor technique, not the poor technique itself. If poor technique results in a positive play, that is graded at the same level as good technique yielding a positive play."

"Essentially, we’ve created a new type of scouting that strictly looks at performance, not necessarily the process that gets there. In our dealings with NFL clients, we’ve referred to this as supplying the “what” as they supply the “why.” We can tell a team that an offensive tackle gives up an inordinate amount of bullrush pressure and they can determine if it’s a lack of technique, functional strength, or perhaps a combination of the two."



To me, PFF doesn't tell you how good a player is. Instead it tells us how well a player produces. There is a big difference between the two. They only look at the result of the play. Anytime someone attempts to justify rating players in Madden with PFF stats, I just roll my eyes and ignore it.

[charter04]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

All you need to do is read their FAQ page. Lot's of stuff that should raise red flags in there. Here are some quotes:

"All of the data is collected in order to build a detailed picture of each player’s performance and production over the course of a season."

"We offer a different type of scouting, strictly based on performance and not technique or upside"

"We are looking for the result of that poor technique, not the poor technique itself. If poor technique results in a positive play, that is graded at the same level as good technique yielding a positive play."

"Essentially, we’ve created a new type of scouting that strictly looks at performance, not necessarily the process that gets there. In our dealings with NFL clients, we’ve referred to this as supplying the “what” as they supply the “why.” We can tell a team that an offensive tackle gives up an inordinate amount of bullrush pressure and they can determine if it’s a lack of technique, functional strength, or perhaps a combination of the two."



To me, PFF doesn't tell you how good a player is. Instead it tells us how well a player produces. There is a big difference between the two. They only look at the result of the play. Anytime someone attempts to justify rating players in Madden with PFF stats, I just roll my eyes and ignore it.

I'm sure PFF is one of the sources that was used to rate players by Donny Moore. Madden ratings usually change based on production just like PFF. It's one of the reasons the ratings aren't accurate.

Example: DeMarco Murray got huge ratings boosts as the season went on last year.

But, did he really get better? Or was the line and his carries what raised production?

He even got a one point raise in his speed rating?! Lol so as the year went on he even got faster?! Makes no sense.

EA has been doing ratings backward for a while.

[fballturkey]

Re: Donny Moore, the 'Madden Ratings Czar', Leaving EA

I'm sure PFF is one of the sources that was used to rate players by Donny Moore. Madden ratings usually change based on production just like PFF. It's one of the reasons the ratings aren't accurate.

Example: DeMarco Murray got huge ratings boosts as the season went on last year.

But, did he really get better? Or was the line and his carries what raised production?

He even got a one point raise in his speed rating?! Lol so as the year went on he even got faster?! Makes no sense.

EA has been doing ratings backward for a while.

Agreed 100% and this is the problem I have with the whole XP system as well. A player doesn't get better by making big plays on the field. They make big plays on the field because they're good. It makes no sense to reward players in a CCM that play well with ratings boosts.