Madden 19 Ratings Spreadsheet

[briz1046]

Quote: Originally Posted by Other Guy

Does the game necessarily need to have player ratings on irrelevant attributes? As you mentioned, the lower end of the ratings scale is reserved, for lack of a better word, for players who don’t rely on that skill (zone coverage for HBs). Is there a reason not to remove these ratings entirely? Thanks for any feedback, appreciate the insight.

I don't think it's "necessary" for all ratings on every player but some do matter to a point at least , HB passing for HB option plays ( are they still in the game at all?) , and defensive attributes for offensive players on turnovers , and of course ST plays , not saying they need be accurate ( they could never be ,per se ) but in the game and at lower levels than the scale used for positions more traditionally associated with those ratings

[khaliib]

Quote: Originally Posted by adembroski

I gotta do this because the conversations about ratings tend to include a lot of this kind of thing. An 80 short throwing accuracy is not a B. Well, it is in the scouting system, but the scouting system was designed with the idea that you're too stupid to understand that grades can be relative. It was indeed suggested by yours truly that the grades be relative to the scale at which the attribute is measured by position, but I was rebuffed and told my faith in the Madden user's intellect was unfounded. Also, I'm about 90% sure the guy who designed it has no idea what a median, a standard deviation, or a standard distribution is. So when I say what I'm about to say, I'm not talking about the scouting grades, I'm talking about a practical comparison of players on the roster. You have to keep in mind when looking at ratings (I'm talking about individual attributes, not OVR, that's a different beast and I'll get to it later) that every player regardless of position has every rating. QBs have ZCV, kickers have POW, LBs have RBS, and linemen have TAK. So the SAC attribute must account for the fact that both Dak Prescott and Ezekiel Elliott have a SAC attribute... so the lowest QBs can't be rated 1 in the category because they're QBs. And an NFL QB necessarily has exceptional throwing accuracy compared to the average HB or LT or whatever. So we have to look at the rating relative to the position. Among QBs in the game, the scale for SAC runs 73-99 with a median of 83. There are 41 players better than 83, and 54 players worse. Pretty fair to say 83 is a "C," not a "B." We tend to look at the numbers and decide kind of arbitrarily what we think a given number should mean, and we have no idea what it means in the game. An 83 seems like a high number on what we think of as a 99 point scale, but when you have to fit the entire rest of the league on the 99 point scale, you necessarily have to use only the very top of the scale to rate guys. So it doesn't really matter how high or low the numbers are. What matters is what they are compared to one another within a given position. A majority of the complaints that I see about ratings relate to the individual's own arbitrary idea of what the numbers should mean rather than EA's actual formulas in the game that translate the numbers into performance. They could conceivably rate every QB on a 5-point scale, from 94-99, and they would tune the formulate so that 96 performed with average accuracy. They could spread the scale out, and rate guys on the full 99 point scale (though it would be strange that every HB in the game threw as well as the worst QB), and adjust so that 50 performed with average accuracy. Brian Hoyer has average short accuracy, thus his rating is 83. In the 94-99 scale, he'd be changed to a 96. If they spread it out to cover the entire scale, he'd be a 50. Basically what I'm saying is stating that there are "too many" guys over 80 is an utterly meaningless statement because it's based only one what kind *feels* too big. According to the scale as it stands, Tom Brady's 99 is 60% better than Brian Hoyer because you measure from 73-99, not 1-99. As a side note, I'd be willing to bet the scale actually is 67-99, but they didn't feel anybody was JaMarcus Russell enough to go below 73 this year. The gameplay team decides what the given ratings levels mean, the guy who does the ratings works within those confines. OVRs are even less meaningful because they're completely arbitrary based on some dudes impression of how important each attribute compared to the next with little to no regard for what scale each attribute is on for that position (I was some dude for Madden 12). It's a broken, worthless formulate that only resembles a reflection of the ratings because they have an ineffective normalizing factor in it. Anyone who starts whining about OVRs is pretty much just mentally masturbating. Just for the record, there are 481 players on the roster at 80+. This is 20.3% of the total pool. I don't think that's bad at all. It makes 80 pretty much the precise cutoff for "starter quality." I think that's a pretty solid way to rate. Here's a breakdown you might find interesting. Players rated 95-99: 28, 1.2% Players rated 90-94: 61, 2.6% Players rated 85-89: 129, 5.4% Players rated 80-84: 263, 11.1% Players rated 75-79: 376, 15.9% Players rated 70-74: 515, 21.7% Players rated 65-69: 508, 21.5% Players rated 60-64: 414, 17.5% Players rated <60: 74, 3.1% Kinda gives one a new perspective compared to just deciding that there are too many 95s or whatever. It's one thing to count 28 95+players and say that's too many, but I seriously doubt if pressed the average fan couldn't name at least 1 or 2 players on each roster at least in the discussion for deserving it, and for every Cleveland, San Francisco, or New York that doesn't have a guy like that, there's a Pittsburgh, Los Angeles, or Philadelphia that might have 2 or 3.

Bro, you need to quit being “Stingy” with your insight.

I ain’t going anywhere in this 103 degree weather in Fresno and my spaghetti is almost done, so loosen them fingers and get-to-snichin on that keyboard with some more info. lol

[Sphinx]

Quote: Originally Posted by khaliib

Bro, you need to quit being “Stingy” with your insight. I ain’t going anywhere in this 103 degree weather in Fresno and my spaghetti is almost done, so loosen them fingers and get-to-snichin on that keyboard with some more info. lol

Exactly, I want more insight! More more!

[tdawg3782]

Quote: Originally Posted by adembroski

I gotta do this because the conversations about ratings tend to include a lot of this kind of thing. An 80 short throwing accuracy is not a B. Well, it is in the scouting system, but the scouting system was designed with the idea that you're too stupid to understand that grades can be relative. It was indeed suggested by yours truly that the grades be relative to the scale at which the attribute is measured by position, but I was rebuffed and told my faith in the Madden user's intellect was unfounded. Also, I'm about 90% sure the guy who designed it has no idea what a median, a standard deviation, or a standard distribution is. So when I say what I'm about to say, I'm not talking about the scouting grades, I'm talking about a practical comparison of players on the roster. You have to keep in mind when looking at ratings (I'm talking about individual attributes, not OVR, that's a different beast and I'll get to it later) that every player regardless of position has every rating. QBs have ZCV, kickers have POW, LBs have RBS, and linemen have TAK. So the SAC attribute must account for the fact that both Dak Prescott and Ezekiel Elliott have a SAC attribute... so the lowest QBs can't be rated 1 in the category because they're QBs. And an NFL QB necessarily has exceptional throwing accuracy compared to the average HB or LT or whatever. So we have to look at the rating relative to the position. Among QBs in the game, the scale for SAC runs 73-99 with a median of 83. There are 41 players better than 83, and 54 players worse. Pretty fair to say 83 is a "C," not a "B." We tend to look at the numbers and decide kind of arbitrarily what we think a given number should mean, and we have no idea what it means in the game. An 83 seems like a high number on what we think of as a 99 point scale, but when you have to fit the entire rest of the league on the 99 point scale, you necessarily have to use only the very top of the scale to rate guys. So it doesn't really matter how high or low the numbers are. What matters is what they are compared to one another within a given position. A majority of the complaints that I see about ratings relate to the individual's own arbitrary idea of what the numbers should mean rather than EA's actual formulas in the game that translate the numbers into performance. They could conceivably rate every QB on a 5-point scale, from 94-99, and they would tune the formulate so that 96 performed with average accuracy. They could spread the scale out, and rate guys on the full 99 point scale (though it would be strange that every HB in the game threw as well as the worst QB), and adjust so that 50 performed with average accuracy. Brian Hoyer has average short accuracy, thus his rating is 83. In the 94-99 scale, he'd be changed to a 96. If they spread it out to cover the entire scale, he'd be a 50. Basically what I'm saying is stating that there are "too many" guys over 80 is an utterly meaningless statement because it's based only one what kind *feels* too big. According to the scale as it stands, Tom Brady's 99 is 60% better than Brian Hoyer because you measure from 73-99, not 1-99. As a side note, I'd be willing to bet the scale actually is 67-99, but they didn't feel anybody was JaMarcus Russell enough to go below 73 this year. The gameplay team decides what the given ratings levels mean, the guy who does the ratings works within those confines. OVRs are even less meaningful because they're completely arbitrary based on some dudes impression of how important each attribute compared to the next with little to no regard for what scale each attribute is on for that position (I was some dude for Madden 12). It's a broken, worthless formulate that only resembles a reflection of the ratings because they have an ineffective normalizing factor in it. Anyone who starts whining about OVRs is pretty much just mentally masturbating. Just for the record, there are 481 players on the roster at 80+. This is 20.3% of the total pool. I don't think that's bad at all. It makes 80 pretty much the precise cutoff for "starter quality." I think that's a pretty solid way to rate. Here's a breakdown you might find interesting. Players rated 95-99: 28, 1.2% Players rated 90-94: 61, 2.6% Players rated 85-89: 129, 5.4% Players rated 80-84: 263, 11.1% Players rated 75-79: 376, 15.9% Players rated 70-74: 515, 21.7% Players rated 65-69: 508, 21.5% Players rated 60-64: 414, 17.5% Players rated <60: 74, 3.1% Kinda gives one a new perspective compared to just deciding that there are too many 95s or whatever. It's one thing to count 28 95+players and say that's too many, but I seriously doubt if pressed the average fan couldn't name at least 1 or 2 players on each roster at least in the discussion for deserving it, and for every Cleveland, San Francisco, or New York that doesn't have a guy like that, there's a Pittsburgh, Los Angeles, or Philadelphia that might have 2 or 3.

So much awesome in this post. Much of it is how I've always thought the ratings work. Give us more please lol.

Sent from my SM-G950U using Tapatalk

[adembroski]

Quote: Originally Posted by Other Guy

Great stuff man, truly! I was doing some rudimentary statistical analysis on the ratings as well, and pretty much came to the same conclusions you did. There really isn’t anything fundamentally wrong with ratings or the ratings “spread” within positions. If there is an issue, I’d say it’s that the ratings differences don’t show through in actual gameplay. That’s what leads most people into the “ratings are too close” train of thought. It’s a gameplay issue rather than a ratings issue, imo.

Completely different discussion. Do the ratings matter enough in gameplay? is there a great enough difference? I could answer either way depending on when you ask. I could say I can't *feel* the difference one day, the next day I'm watching a game and reminded of just how little difference there really is between NFL players. These are the top 1% of top college players, remember. This is why we get slideritis. See my sig.

Quote:

I did have a few questions that maybe you could shed some light on. Do the devs use, or do you think there should be, a distribution to the ratings? I didn’t find any specific rating that fit a normal, or any other commonly-used, distribution curve. The only pattern I noticed was that most ratings are positively skewed.

Let me ask you this; by what model would you distribute "talent" through the league, and how strict would you be with its imposition? How many statistical anomalies do you allow?

Currently, the 65% of players are between 62 and 75 OVR, or a half a standard deviation of the mean (which is 71, the standard deviation is 8.8). If you're comparing to the standard distribution, you'd actually want about 38% of the roster in there.

A standard distribution, assuming no change in the mean, would look like this;

2 players from 98-99 OVR
12 players from 94-97 OVR
40 players from 90-93 OVR
104 players from 85-89 OVR
217 players from 81-84 OVR
355 players from 76-80 OVR
452 players from 72-75 OVR

I'm not sure the average Madden fan would accept that. I think they'd say the number of 99s is much better, but they'd go nuts over 54 players ovr 90. You could go Pareto distribution, which is probably a more realistic model, but then you'd be making the "everyone feels the same" problem even more pronounced.

Quote:

Does the game necessarily need to have player ratings on irrelevant attributes? As you mentioned, the lower end of the ratings scale is reserved, for lack of a better word, for players who don’t rely on that skill (zone coverage for HBs). Is there a reason not to remove these ratings entirely? Thanks for any feedback, appreciate the insight.

No, it probably doesn't need them. It has them because you have Denard Robinsons and J.J. Watts that occasionally play out of position or have a history at another position. I think if I were designing it I'd just have personality traits like in NFL Head Coach and attach special flags to players who have talents outside their position, that would free-up a ton of confusing chart data and memory as well.

[adembroski]

Rereading my earlier post: How the hell did I end up with formulate where I meant formula? Grammarly can be stupid sometimes.

[adembroski]

What's the matter with me? I come up in this thread and start arguing and tossing out wisdom like I'm some kinda messiah and I didn't even thank the OP;

JKennedy87, you're a superhero. Thanks so much for this.

[SyncereBlackout]

Quote: Originally Posted by adembroski

I gotta do this because the conversations about ratings tend to include a lot of this kind of thing. An 80 short throwing accuracy is not a B. Well, it is in the scouting system, but the scouting system was designed with the idea that you're too stupid to understand that grades can be relative. It was indeed suggested by yours truly that the grades be relative to the scale at which the attribute is measured by position, but I was rebuffed and told my faith in the Madden user's intellect was unfounded. Also, I'm about 90% sure the guy who designed it has no idea what a median, a standard deviation, or a standard distribution is. So when I say what I'm about to say, I'm not talking about the scouting grades, I'm talking about a practical comparison of players on the roster. You have to keep in mind when looking at ratings (I'm talking about individual attributes, not OVR, that's a different beast and I'll get to it later) that every player regardless of position has every rating. QBs have ZCV, kickers have POW, LBs have RBS, and linemen have TAK. So the SAC attribute must account for the fact that both Dak Prescott and Ezekiel Elliott have a SAC attribute... so the lowest QBs can't be rated 1 in the category because they're QBs. And an NFL QB necessarily has exceptional throwing accuracy compared to the average HB or LT or whatever. So we have to look at the rating relative to the position. Among QBs in the game, the scale for SAC runs 73-99 with a median of 83. There are 41 players better than 83, and 54 players worse. Pretty fair to say 83 is a "C," not a "B." We tend to look at the numbers and decide kind of arbitrarily what we think a given number should mean, and we have no idea what it means in the game. An 83 seems like a high number on what we think of as a 99 point scale, but when you have to fit the entire rest of the league on the 99 point scale, you necessarily have to use only the very top of the scale to rate guys. So it doesn't really matter how high or low the numbers are. What matters is what they are compared to one another within a given position. A majority of the complaints that I see about ratings relate to the individual's own arbitrary idea of what the numbers should mean rather than EA's actual formulas in the game that translate the numbers into performance. They could conceivably rate every QB on a 5-point scale, from 94-99, and they would tune the formulate so that 96 performed with average accuracy. They could spread the scale out, and rate guys on the full 99 point scale (though it would be strange that every HB in the game threw as well as the worst QB), and adjust so that 50 performed with average accuracy. Brian Hoyer has average short accuracy, thus his rating is 83. In the 94-99 scale, he'd be changed to a 96. If they spread it out to cover the entire scale, he'd be a 50. Basically what I'm saying is stating that there are "too many" guys over 80 is an utterly meaningless statement because it's based only one what kind *feels* too big. According to the scale as it stands, Tom Brady's 99 is 60% better than Brian Hoyer because you measure from 73-99, not 1-99. As a side note, I'd be willing to bet the scale actually is 67-99, but they didn't feel anybody was JaMarcus Russell enough to go below 73 this year. The gameplay team decides what the given ratings levels mean, the guy who does the ratings works within those confines. OVRs are even less meaningful because they're completely arbitrary based on some dudes impression of how important each attribute compared to the next with little to no regard for what scale each attribute is on for that position (I was some dude for Madden 12). It's a broken, worthless formulate that only resembles a reflection of the ratings because they have an ineffective normalizing factor in it. Anyone who starts whining about OVRs is pretty much just mentally masturbating. Just for the record, there are 481 players on the roster at 80+. This is 20.3% of the total pool. I don't think that's bad at all. It makes 80 pretty much the precise cutoff for "starter quality." I think that's a pretty solid way to rate. Here's a breakdown you might find interesting. Players rated 95-99: 28, 1.2% Players rated 90-94: 61, 2.6% Players rated 85-89: 129, 5.4% Players rated 80-84: 263, 11.1% Players rated 75-79: 376, 15.9% Players rated 70-74: 515, 21.7% Players rated 65-69: 508, 21.5% Players rated 60-64: 414, 17.5% Players rated <60: 74, 3.1% Kinda gives one a new perspective compared to just deciding that there are too many 95s or whatever. It's one thing to count 28 95+players and say that's too many, but I seriously doubt if pressed the average fan couldn't name at least 1 or 2 players on each roster at least in the discussion for deserving it, and for every Cleveland, San Francisco, or New York that doesn't have a guy like that, there's a Pittsburgh, Los Angeles, or Philadelphia that might have 2 or 3.

I love this forum



Sent from my iPhone using Tapatalk