Attributes scale in EA

[Hypnotist]

Is it just me or does it seem like the weight of the attribute scale has shifted to the middle (10 or 50) as "average" attribs. In 05 & 07 it really seemed like what was considered middle-of-the-road (for an NHLer) was in the 15/75 range (hence Green). With the last update, they changed the coloring scheme so green starts at 12/60 with 75+ just being brighter green. If this is the case, it DRASTICALLY alters what's acceptable for players at the different levels.

[KevT90]

Yes it was modified that way

[mpronger]

The only change was the coloring to highlight, but the attributes still have the same weight. A 10 player in EHMEA is not the equivalent of a 14 player in 2007. If you open the file Eastside Hockey Manager\data\skins\ehm_default.col you can edit the color for the attributes and this is how the game separates the attributes, in four categories: Low, Normal, Good and Excellent.

You guys are questioning the fact that the "Good" attribute that used to be lighter yellow in 2007 is now dark green.

If this is confusing you, then do what I did. I edited the mentioned file to have the dark green become light yellow, I posted this in the "attributes colors 1.1" post see how my EHM:EA looks:


With these values bellow you'll get a very similar look to EHM 2007, just find these entries in the file and modify the numbers which is nothing more than color in hex code:
# PROFILE_ATTRIB_LOW
88 200 0 0

# PROFILE_ATTRIB_NORMAL
89 211 135 9

# PROFILE_ATTRIB_GOOD
90 248 248 0

# PROFILE_ATTRIB_EXCELLENT
91 0 255 00

EDIT: Had switched 10 and 14 in the example provided

[Hypnotist]

A 14 player in EHMEA is not the equivalent of a 10 player in 2007.

I think you may have misunderstood. The above example should be switched; a 10 player in EA is not the equivalent of a 14 player in 2007. But I really think it is, or something proximate. There are WAY too many NHLers with speed ratings in the 40-55 range and only a select few have lightning speed. Either that or suddenly the vast majority of top tier talent all lost about 5 steps.

[Duranium]

The only change was the coloring to highlight, but the attributes still have the same weight. A 14 player in EHMEA is not the equivalent of a 10 player in 2007. If you open the file Eastside Hockey Manager\data\skins\ehm_default.col you can edit the color for the attributes and this is how the game separates the attributes, in four categories: Low, Normal, Good and Excellent.

You guys are questioning the fact that the "Good" attribute that used to be lighter yellow in 2007 is now dark green.

If this is confusing you, then do what I did. I edited the mentioned file to have the dark green become light yellow, I posted this in the "attributes colors 1.1" post see how my EHM:EA looks:

With these values bellow you'll get a very similar look to EHM 2007, just find these entries in the file and modify the numbers which is nothing more than color in hex code:
# PROFILE_ATTRIB_LOW
88 200 0 0

# PROFILE_ATTRIB_NORMAL
89 211 135 9

# PROFILE_ATTRIB_GOOD
90 248 248 0

# PROFILE_ATTRIB_EXCELLENT
91 0 255 00

thxalot for this tip - for everyone who´s interested in adjusting the attribute colors, this site is all you need

https://www.colorcodehex.com/

looking way better now

[mpronger]

The above example should be switched; a 10 player in EA is not the equivalent of a 14 player in 2007.

Yep, I inverted the number on the above example, edited and switched.

[batdad]

Only a few should have lightning speed hypno. In real life there is so little separation between players and their speed now, that everything is bunched to the middle. Parity among players has never been higher. Fewer and fewer guys stand out. A 20 (100) rating for skating is few and far between.

Game should not have 300 players or even more than 30 players who are super fast.

Training has allowed the great majority of guys in the NHL to be in the middle range of skating ability. For me that is 11-15, but 10 can work as well. The numbers themselves are irrelevant in many ways to me personally.

An NHL skater in general needs to be 12 or higher to me. But...there are some that have the tools to succeed at much lower than that because of experience or skill set. Guys such as Mark Streit who had 38 points for me at age 38 and 39 playing on my third pairing and on the PP. Skating was 9-8=7 and even 6 at the end of that run. Then scored 19 points as a 4th line LW and pp occasional guy at age 40.

No one attribute can be looked at in seclusion. You can find all the 20 skaters and sort them....sure easily done. But I am not seeing 10 skaters equal to 14 from EHM 2007. However I think as their should be there is less of a difference between a 10 skater and an 11 skater than there was in the past games. Maybe. Tighter skill levels.

Or are you saying this is the game or database? Because if you are speaking about database, it has been compared and contrasted over and over and over, and the average skater vs top end skaters have been changed. THere were way too many guys with elite speed and elite skills at one point in the DB, and those had to be changed to move everyone closer to the middle, save for the few really elite guys.

[Peter_Doherty]

Agree with most (maybe even everything) of what batdad wrote there

Honestly, there should be 0 skaters in the NHL with 20-20-20 (acceleration, speed, agility), Hagelin would probably be the closest with something like 20 acc 20 agi 18 or 19 speed, maybe someone in the West is as fast as him but my point is that noone should have a 60/60 there.

[Obalobex]

game genned players are generating with all 20s in physical stats though, like, lots of them. its kind of unsettling

[Peter_Doherty]

Yeah, that's a big problem... 4 seasons in i had over 70 Hagelins in my game

[Alessandro]

I reported this at the sigames forums http://community.sigames.com/showthread ... s-too-much

[CJ]

I have modified the attribute colouring and made a thread for it here, if interested:
http://www.ehmtheblueline.com/forums/vi ... 32&t=14055

[Hypnotist]

Wow! this went WAY deeper than my initial question. LOL

FTR: I'm not against having the "average" stat towards the middle of the scale. I prefer it. I'm a bell-curve believer (which is why I don't teach anymore). I just wanted to know if anyone else noticed this because, as I said, it drastically changes the way attribs would be read and analyzed. If the weight shift was not intended, then it would probably go in the bug list since the algorithm is making everybody a lot crappier. If it was intended then I need to revise what I consider good enough for any particular level of competition. Thanks for all the feed back.

[Obalobex]

unfortunately you're using a 1-20 or 1-100 scale to not just model the NHL (which would make it trivial to assign numbers to skills) but the entire world and division of leagues. someone with a 1 in any physical trait or technical trait would probably not be an nhler and then it becomes a grey area of trying to decide if the scale is 6-20 or 11-20, and that needs to change based on stat (a player with a 6 in balance would probably not succeed professionally, a player with 6 checking could be an all-star)

[batdad]

Huh? Not sure I understand you here Obalobex, but all the attributes are based off of the TOP LEVEL being 20, that means that some players with 1s in some categories could succeed in some levels of hockey, but not in higher ones. It is a top down scale, just as it is in real life. Maybe I misunderstand you completely. But yes, some attributes are more important to success of individual players than others, just as it is MUCH harder for small players to be successful than it is for bigger players. It totally makes sense to me that a guy with great balance but no ability to do things iwth the puck or check players does not succeed in hockey. They are known as figure skaters, and are very good at their game.

[Obalobex]

the game has to model 1-20 because it is modelling multiple levels of hockey, 1 being literal children and 20 being an elite nhler. if 1 was simply "bottom 5th percentile of NHLers" it would be easier to use the scale to model the nhl

[Peter_Doherty]

the game has to model 1-20 because it is modelling multiple levels of hockey, 1 being literal children and 20 being an elite nhler. if 1 was simply "bottom 5th percentile of NHLers" it would be easier to use the scale to model the nhl

And that's why we have CA aswell, a player with 100 CA 20 wristshot will have a worse shot then a player with 200 CA 20 wristshot, atleast that's how i understand it?

[batdad]

Not necessarily Peter in my understanding.

Attribute distribution comes into play. Each CA has a total # of attribute points given to it, and it is spread around through the player profile depending on player role within it, and how that player is designed. It is possible for a guy with a 50 CA to have a better shot than a guy with a 200 CA in the game. Both can be 20, and one may have a better shot than the other. It may be the guy with 50 CA, just he will not be able to do anything else. Kind of like the dudes that win the long drive in golf.

Ovechkin with a 20 shot could have a weaker shot than say some random 16 year old. HIghly improbable, but is possible. That random 16 year old would have alot to work on in other areas.


And yes O it might be, except then you cannot model the rest of the world, you have to base the whole thing. This is not just an NHL game as I am sure you are aware. You would then be getting into ridiculous numbers for attributes in the negatives, or changing the entire attribute scale. What we have now makes total sense--modelling an entire world of hockey, not just one league.

REmember attributes are NOT the be all and end all of a player. Guys with lower ones may peform well depending on a ton of things.

[Peter_Doherty]

Aight, thought CA weighed in on the attribute itself not only the attribute distribution, guess i was wrong.

[batdad]

Do not think it does, but it is possible I suppose.