New Look: Rating Overhaul

[AdamJones113]

As I and many others have discussed in previous posts, there is a dearth of differentiation between players in EA NHL. John Scott handles like Sidney Crosby, Phil Kessel hits like Zdeno Chara, and Chris Mason plays the pipes like Henrik Lundqvist. The series hasn't seen much overhaul rating-wise since potentials went from letter grades to stars for NHL 13, so here's a possible rating system that is designed to incorporate realism and reflect the ups and downs of an NHL season while remaining true to life.

Let's break it down step-by-step:

Letter Grades (F to A+)
Now, this is probably controversial. But the way I see it is, there is more of a difference between an A+ and an A than there is between a 99 and a 96. Rather than a scale of 36-99, which we all know is hardly used, the letter grade system encapsulates a scale of 13 different rankings, from A+ (Sidney Crosby, and only Sidney Crosby) to F (I'll get into that). Letter grades for overall and letter grades for the individual rating categories.

NHLScreen1.jpg

Letter Grade Fluctuation
This kills two birds with one stone: roster updates and Be A GM hot/cold streaks + fluctuation. For the former, a simple letter grade change is required. The latter is more complicated. When a player is on a hot streak, performing well, the letter grade is changed to a higher grade and the colour turns green, indicating improvement from the original grade. When a player is on a cold streak, performing poorly, the letter grade is changed to a lower grade and the colour turns red, indicating a worse grade than the original:

NHLScreen2.jpg

The rating will fluctuate after every game, but being on a long hot streak means the rating can go higher than normal while being on a long cold streak means the rating can decrease lower than normal. For injuries, it is essentially the same thing, with the rating turning orange to indicate injury. It is up to you to decide whether the rating is high enough for the player to play through the injury or whether his injury is too severe and the player should rest:

Untitled.jpg

Tangible Ratings vs. Intangible Ratings
The current iteration has six rating categories: Physical, Defense, Shooting, Puck Skills, Skating, Senses. I am proposing that ratings instead be divided into two categories, Tangibles and Intangibles. Tangible ratings, such as Speed, Passing, Wrist Shot Power, etc, are subject to the above fluctuation and will regress as a player gets older. Intangible ratings, such as Defensive Awareness, Discipline, Poise, etc, remain constant throughout a season and will only increase, at the end of the season, unless the player suffers a serious concussion or head injury. The Intangible ratings will increase as a player plays more games in the NHL:

NHLScreen3.jpg

Rating Category Additions: Skaters
Here are a few proposed new categories to further quantify players:

Offensive Positioning
Affects how well the player positions himself in the offensive zone and on the breakout. Determines how quickly he covers for a pinching defenseman at the point, how well he gains positioning in front of the net, and how well he finds open ice to set up in.

Defensive Positioning
Affects how well the player positions himself in the defensive zone and on the backcheck. Determines how well he defends passing lanes, how well he squares to the shooter/puck carrier, how well he picks up an open point man, and how well he gains position in the slot.

Vision
Affects how well the player sees the ice. Determines how well the player finds an open man to pass to, how well he sees incoming checks, and how well he reads the goalies.

Reaction
Affects how well the player reacts. Determines how quickly the player can get off one-times, how well he can change direction, and how quickly he reacts to a turnover or pass.


Sample Player Page

NHLScreen4.jpg

Potential

I think that potential has improved on the next-gen versions. Provided that playing a player on the NHL team or a lot of minutes in the AHL will raise their attributes (see Intangibles, above), potential doesn't have a lot to change. But here is a an attribute that works well as a measure of potential and well as a comparison between NHL players and players in other leagues.

The rating is how well a player's skills translate to the NHL. A player whose skills fit well into an NHL team—such as a sniper, or a hard-hitting defenseman, will have a higher grade. The grade encapsulates all of the player's stats, tangible and intangible. Therefore a player with a higher grade will have a better chance to become a good NHLer whereas a player with a lower grade will have a worse chance. Yet this rating includes another modifier: how ready the player is for the NHL. If they are ready or nearly ready, the grade will be coloured green; if they still have work to do, the colour will be yellow, and if they are far from NHL ready, the grade will be coloured red (white for an NHL player). To explain, here are two examples.

This first example is an example of what Connor McDavid's rating would look like. McDavid is projected to go 1st overall and start in the NHL the first season after being drafted. His skills as a playmaker translate well to the NHL, and being a #1 and a generational talent he is close to NHL ready.

Here is a lower-round two way forward, with skills translating poorly to the NHL and a player that (being picked late) is far from NHL ready:

Here is a foreign player, whose skill as a sniper translates well but who is not NHL-ready, playing against inferior competition:

Player Roles
Rather than having vague definitions for snipers, playmakers, grinders, and the ever-mystifying 3rdSCRs, I am proposing a kind new player type that combines both a player's type and their line. For example, Sidney Crosby would be a 1stPLY; Mason Raymond a 3rdSNP; Zac Rinaldo a 4thGRN. The role is the players' SUGGESTED role, based on their skills and the team they are on. The advantages of combining the two are that it means both create separation (rather than having Steven Stamkos and Raymond fall under the SNP category) and gives a better indication of where a player is best suited to play.

Role Performance
Ah yes, finally the thing I've been alluding to for a while.

This is a fluctuating (based on player performance) grade that determines how well Player A is playing into his role in comparison with players across the league with the same role as Player A. Players with a high personal grade will play better in their role: playmakers with high grades will pass more precisely; snipers with high grades will shoot more accurately, etc. Players will a low personal grade play worse in their role: playmakers will low grades will pass poorly, snipers with low grades will shoot less accurately, etc. This has the potential to both create player separation and also make players play more accurately as higher grades means the player is playing up to his skill better.

As stated, the attribute itself fluctuates based on performance. The colour scheme is the same as the already discussed tangible attributes: green for better than original, red for worse, white for right there. The addition for this attribute is a second letter grade, in parenthesis, to the right of the main grade. This one indicates the league average for players of that type. It also fluctuates with the same colour scheme. This way, if a player is performing poorly you can still judge him to be better than the league average at his position.

Here are three examples: (bonus points if you can guess which player I was thinking of)

In this example, the player is playing above his default role performance rating. Being a sniper, his shooting attributes are currently boosted. The league average grade is also higher than normal.

In this example, the player is playing lower than his default rating. His skills are not performing into his role. However, the player is still above the league average, which is right at its default rating.

In this example, the player is playing right at his default rating, but the league's average role performance rating is lower than normal. Therefore the player does not gain boosts or lose skill, but should still play better than most players in his role.

It also works for juniors, both in-season and as a way to compare for the NHL—they would have a lower grade to begin with, having less experience and skill than NHL players.

Instituted the correct way, this rating would solve a number of problems (some mentioned in this thread by Bigwill and Simple) and provide a good second-in-command rating, as it were, that could take over from overall in Be A GM.



Whew, long and complicated. Hope you all got through it OK. Break for a bit, then starting to work on goalie ratings next (yes, this has been about defensemen too. Just used forwards 'cause I felt like it.) I know the whole system needs work, but I think it was important to set up the base. Refinements will come with time.

Keep the feedback coming and keep the questions and concerns coming!

[Shirazz]

Really like this idea!

[bigwill33]

Don't take this the wrong way or get discouraged in your works; but I don't like this system that you are suggesting.

The main reasons are because you are limiting the ratings to 13 potential grades. That seems like far less seperation versus the numbered rating system currently in place (or even the older one that stretched from 30's-90's).

I am also going to say something that may not be popular around here. I don't think that the separation in player skill is screwed up how most of the people at OS are claiming it to be. This could be misinterpreted but try and follow what I mean.

The players in EA's NHL series are all capable of pulling off dekes and the speed threshold needs to be adjusted to show more dynamic. That being said, I feel that there is separation in talent in this series, however. If you play a season or even online you often see the difference based on stats alone.

The goals typically come from players that score goals. Meaning players with high shooting accuracy and shot power score more goals than the 4th liner that doesn't have those attributes. If I pull up and curl into the slot with a Crosby I can expect a reasonable chance to pick that top corner. Where as if I do the same with a Zach Sill I don't expect to have that prowess to score.

The problems in my opinion arise in the fact that the goalies all play the same exact way and that their animations are just sped up based on their ratings.

And also that skill sets for skaters are not properly displayed in the series. If you can dangle you should be able to deke. Playmakers should be able to pass through pinholes drop pucks on to sticks. And snipers should be able to pick a spot and aim for it and hit it more often than other players. The list can go on and on. Role players and 3rd liners should be able to win battles in the corners and cycle (which is nearly non existent in this game).

So, overall the point I am trying to make is this; The actual ratings and system do not matter as much as the game play. You could rate everyone 86-99 with your system and it might not be any different than A+ through F. BUT you could also keep the current rating system and adjust the discrepancy in how the players play to their strengths and weaknesses even more.

That is what is truly needed to me. Because it doesn't matter what you label the ratings as. It matters that there is more seperation in how players play and more dynamic in style and signature movements. The only reason we want to see the numbers reflect these changes is because it makes it easier for our brains to comprehend the differences when we see it with tangibles such as a numbered rating system. But in all honesty it is not needed to fix this game and is not as important what the numbers say as much as it is what we see during the game play and the statistics that result from it.

[AdamJones113]

Quote: Originally Posted by bigwill33

Don't take this the wrong way or get discouraged in your works; but I don't like this system that you are suggesting. The main reasons are because you are limiting the ratings to 13 potential grades. That seems like far less seperation versus the numbered rating system currently in place (or even the older one that stretched from 30's-90's). I am also going to say something that may not be popular around here. I don't think that the separation in player skill is screwed up how most of the people at OS are claiming it to be. This could be misinterpreted but try and follow what I mean. The players in EA's NHL series are all capable of pulling off dekes and the speed threshold needs to be adjusted to show more dynamic. That being said, I feel that there is separation in talent in this series, however. If you play a season or even online you often see the difference based on stats alone. The goals typically come from players that score goals. Meaning players with high shooting accuracy and shot power score more goals than the 4th liner that doesn't have those attributes. If I pull up and curl into the slot with a Crosby I can expect a reasonable chance to pick that top corner. Where as if I do the same with a Zach Sill I don't expect to have that prowess to score. The problems in my opinion arise in the fact that the goalies all play the same exact way and that their animations are just sped up based on their ratings. And also that skill sets for skaters are not properly displayed in the series. If you can dangle you should be able to deke. Playmakers should be able to pass through pinholes drop pucks on to sticks. And snipers should be able to pick a spot and aim for it and hit it more often than other players. The list can go on and on. Role players and 3rd liners should be able to win battles in the corners and cycle (which is nearly non existent in this game). So, overall the point I am trying to make is this; The actual ratings and system do not matter as much as the game play. You could rate everyone 86-99 with your system and it might not be any different than A+ through F. BUT you could also keep the current rating system and adjust the discrepancy in how the players play to their strengths and weaknesses even more. That is what is truly needed to me. Because it doesn't matter what you label the ratings as. It matters that there is more seperation in how players play and more dynamic in style and signature movements. The only reason we want to see the numbers reflect these changes is because it makes it easier for our brains to comprehend the differences when we see it with tangibles such as a numbered rating system. But in all honesty it is not needed to fix this game and is not as important what the numbers say as much as it is what we see during the game play and the statistics that result from it.

I see what you are saying. However, I think you will be pleasantly surprised when you see my next two aces that I haven't played yet. This was to outline the basis which will let them work properly. Yes, 13 is a much lower scale than 36-99, but again, it's the basis for what's yet to come.

To provide a hint/teaser, one of the next two will address EXACTLY what you are referring to, and it's something that in my mind translates much easier to letter grades than number grades.

I'll release that one tomorrow, maybe the other one later tonight.

Keep the feedback coming!

[bigwill33]

Cool, cool. I like your ideas and will definitely keep reading what you come up with.

[AdamJones113]

Potential

I think that potential has improved on the next-gen versions. Provided that playing a player on the NHL team or a lot of minutes in the AHL will raise their attributes (see Intangibles, above), potential doesn't have a lot to change. But here is a an attribute that works well as a measure of potential and well as a comparison between NHL players and players in other leagues.

The rating is how well a player's skills translate to the NHL. A player whose skills fit well into an NHL team—such as a sniper, or a hard-hitting defenseman, will have a higher grade. The grade encapsulates all of the player's stats, tangible and intangible. Therefore a player with a higher grade will have a better chance to become a good NHLer whereas a player with a lower grade will have a worse chance. Yet this rating includes another modifier: how ready the player is for the NHL. If they are ready or nearly ready, the grade will be coloured green; if they still have work to do, the colour will be yellow, and if they are far from NHL ready, the grade will be coloured red (white for an NHL player). To explain, here are two examples.

This first example is an example of what Connor McDavid's rating would look like. McDavid is projected to go 1st overall and start in the NHL the first season after being drafted. His skills as a playmaker translate well to the NHL, and being a #1 and a generational talent he is close to NHL ready.

PotentialScreen1.jpg

Here is a lower-round two way forward, with skills translating poorly to the NHL and a player that (being picked late) is far from NHL ready:

PotentialScreen2.jpg

Here is a foreign player, whose skill as a sniper translates well but who is not NHL-ready, playing against inferior competition:

PotentialScreen3.jpg

Well, this is complicated and definitely needs work. But I think it's a good idea, at least to consider. Keep the feedback coming and don't hesitate to critique!

Third step coming tomorrow.

[Simple Mathematics]

I like this idea, especially the potential system and the hot and cold streaks. Just 2 things:

1. I don't want them to limit the ratings anymore by being limited to F to A+. I want them to use a full 1 to 99 scale. I want John Scott to have a 1 in shooting accuracy and Stamkos to have a 99. I want there to be a system where a player can have an 80 rating in a certain skill in the AHL, but if you brought him up to the NHL it would drop a certain amount so you can utilize more player separation.

2. I would want regression in all aspects - including both tangibles and intangibles. Look at Dany Heatley. His offensive awareness has dropped a ton in real life, so I want this to be able to happen in the game.

[AdamJones113]

Quote: Originally Posted by Simple Mathematics

I like this idea, especially the potential system and the hot and cold streaks. Just 2 things: 1. I don't want them to limit the ratings anymore by being limited to F to A+. I want them to use a full 1 to 99 scale. I want John Scott to have a 1 in shooting accuracy and Stamkos to have a 99. I want there to be a system where a player can have an 80 rating in a certain skill in the AHL, but if you brought him up to the NHL it would drop a certain amount so you can utilize more player separation. 2. I would want regression in all aspects - including both tangibles and intangibles. Look at Dany Heatley. His offensive awareness has dropped a ton in real life, so I want this to be able to happen in the game.

In reverse order:

Heatley suffered a multiple injuries, something that the game cannot replicate. Something like that could absolutely lower intangibles (as I think about it now, a concussion could could work too to lower intangibles-good call), but since the game cannot include multiple simultaneous injuries I didn't consider it for inclusion. I'll edit the OP for concussions though. (Also, his regression has been mostly due to getting older + being injury prone. But then again, this is what roster editing is for!)


My problem with the bolded is that it doesn't work both ways and it doesn't work well if a player is actually good in a specific category. If John Doe scores 25% of his shots and 60 goals in the AHL with a 90 Wrist Shot Accuracy ("A" grade), but is called up to the NHL, his Wrist Shot Accuracy should not go down to a 75 because now he is playing against NHL players! And what if a player gets sent down? Does he suddenly become a 99 overall? What about Juniors, arguably a step below the AHL? If John Doe is drafted 1st overall, at 70 overall, if he gets sent down does he become a 99? A goal of sending players to the AHL is so they can get more experience and therefore become better, not so they can tear it up, making you seriously consider promoting them.


Next step—the big one—later today.