EHM The Blue Line

The Manimal v5.0 2012/13 needs your help!

Unfortunately, for various reasons, it has not been possible for researchers to update the player attributes for the NHL players. As you will no doubt be aware, it is really important that players' attributes are updated at least on an annual basis. Prospects can develop quickly over the course of a season, some players may have peaked and other players may have performed much worse/better than anticipated.

For this reason, we desperately need volunteers to help rate the NHL players. In order to make things easier, we are looking to concentrate on just a small number of key attributes (height, weight, CA, PA, Offensive Role and Defensive Role). We have also produced a simple guide (below) which will help you rate each of these key areas. We hope that this will make things simple enough for people to be interested in helping research. Unfortunately, if we do not get any volunteers, the NHL player attributes will not be updated for version 5 (rosters and contract details will however be correct). With only a limited amount of time and researchers, it is not possible for us to cover as many areas of updating as we would like - the NHL player attributes being one of these areas.


If you would like to help research, here is what you need to do:

1. Please post which team(s) you will research in this thread.

2. Have a read through the Research Guide below.

3. Download the Research Spreadsheet from here: http://sdrv.ms/Qsu97m

4. Open the NHL Research Sheet.xls and list all of the players currently on the NHL team roster(s) you plan to research. NOTE: You should go by the rosters as they are on NHL.com / the official NHL team site (and NOT according to the Manimal v4.2 Rosters). For each player, you must enter their name, date of birth, height (feet and inches go in individual cells in the spreadsheet) and weight (lbs).

5. Once you have entered all of the player details, you can now enter the CA, PA and Role attributes. If there are any other attributes which you think need changing from the Manimal v4.2 Rosters, use the Additional attributes to be changed column in the spreadsheet to list these.

NOTE: A second spreadsheet named NHL List.xls is included with the zip file you downloaded in step #3. This contains all of the key attributes for the NHL players currently on the Manimal v4.2 Rosters. If you feel that some/all of the attributes in the Manimal v4.2 Rosters remain correct, then just copy these over to your research spreadsheet.

6. Once completed, email your spreadsheet here: ehmtheblueline AT gmail DOT com. Please include your username and teams researched in the subject title of your email (e.g. "Archibalduk - Boston & Winnipeg").


Research Guide:

There are six key attributes we need researched for each player. It may well be that the data currently in the Manimal Database remains correct. If this is the case then please simply enter the same values into the Research Spreadsheet.

• Height
• Weight
• Current Ability
• Potential Ability
• Defensive Role
• Offensive Role

As an optional extra, if you feel that any of a player's attributes should be changed, then please feel free to also enter this into the Research Spreadsheet.

The following is a simple guide to help you rate the players in the key areas:


Height & Weight
Please enter the heights and weights using Imperial measurements (i.e. lbs for weight and feet & inches for height). We need these measurements because we will use these to calculate the Strength attiribute.


Current Ability
This is the overall rating of the player. The very best player to have ever played hockey (e.g. Wayne Gretzky) would be rated as 200 and the very worst player to have ever player hockey would be rated as 1 (this would probably equate to a 14 or 16 year old playing youth hockey). A rating of 190+ would normally only be given to the very best players to have ever played when at the peak of their career.

For the NHL, players should have a CA rating of between 100 and 190. The average player in the NHL should have a rating of 130. Additionally, if the average of the CAs of all players in the NHL were calculated, it ought to be approximately 130. The table below explains how the CAs should be awarded (the most important reference points have been highlighted in bold capital letters). Note that most players should be rated between 115 and 145 (i.e. between Low Average and High Average).

[table][tr][th]CA Rating[/th][th]Description[/th][th]Some very rough examples[/th][/tr]
[tr][td]190[/td][td]TOP[/td][td]Tim Thomas, Sidney Crosby, Alexander Ovechkin[/td][/tr]
[tr][td]180[/td][td]Very Good[/td][td][/td][/tr]
[tr][td]165[/td][td]GOOD[/td][td]Cam Ward, Dan Boyle, Ryan Callahan, Ryan Kesler, Jason Spezza[/td][/tr]
[tr][td]150[/td][td]Above Average[/td][td][/td][/tr]
[tr][td]145[/td][td]High Average[/td][td][/td][/tr]
[tr][td]130[/td][td]AVERAGE[/td][td]Jonas Hiller, Brad Stuart, Nik Antropov[/td][/tr]
[tr][td]115[/td][td]Low Average[/td][td][/td][/tr]
[tr][td]100[/td][td]POOR[/td][td]Jason LaBarbera, Matt Bartkowski, Mike Rupp[/td][/tr][/table]

Some players contracted to an NHL team may have played more in the AHL or ECHL than the NHL. If this is the case, it may be more appropriate to rate them using the CA ranges for the AHL/ECHL. In which case, below are the ranges you should use. You will see that the very best player in the AHL would be rated as 125 and that this is roughly the equivalent of a player who is rated as Average in the NHL.

[table][tr][th]AHL[/th][th]ECHL[/th][th]Description[/th][/tr]
[tr][td]125[/td][td]100[/td][td]TOP[/td][/tr]
[tr][td]117[/td][td]95[/td][td]Very Good[/td][/tr]
[tr][td]110[/td][td] 90[/td][td]GOOD[/td][/tr]
[tr][td]100[/td][td]85[/td][td]High Average[/td][/tr]
[tr][td]90[/td][td]75[/td][td]AVERAGE[/td][/tr]
[tr][td]80[/td][td]65[/td][td]Low Average[/td][/tr]
[tr][td]70[/td][td]55[/td][td]POOR[/td][/tr][/table]

Please note that the table really is just a guide. Don't feel that every High Average AHL player should be rated 100. You could equally rate them as 95 or 105. Just use your best judgement.


Potential Ability
This determines how good a player is likely to become. Players can be split into three categories depending on how old they are:


1. Older Players
If a Goaltender is approximately 31 years older, he should usually have his PA set to the same value as his CA. The only exception to this rule is if he is still likely to improve. If he is, set his PA between 5 and 20 higher than his CA depending on how significantly he will improve (in all likelihood, we would expect goalies not to improve by more than 10 above their CA at this age).

The same rule applies for Defencemen or Forwards who are at least 30 years old.


2. Experienced / Semi-Experienced Players
When setting the player's PA, you will need to take into account their CA. If the player is unlikely to improve much further, set his PA between 5 and 15 higher than his CA.

If the player is likely to improve quite a lot from his current ability (e.g. if he is 27 years old or younger), assign one of the following PAs depending on how good he is likely to become (see the table below). The table below provides details of the range of CA the player will likely achieve with each PA.

[table][tr][th]PA[/th][th]Probable CA & Description[/th][/tr]
[tr][td]-10[/td][td]170 – 200: Likely to become a top NHL player (very rare - e.g. Taylor Hall)[/td][/tr]
[tr][td]-9[/td][td]150 – 180: Likely to become a very good, or possibly top, NHL player[/td][/tr]
[tr][td]-8[/td][td]130 – 160: Likely to become an above average / good NHL player[/td][/tr]
[tr][td]-7[/td][td]110 – 140: Likely to become an average / below average NHL player[/td][/tr]
[tr][td]-6[/td][td]90 – 120: Likely to become an average AHL player / below average NHL player[/td][/tr]
[tr][td]-5[/td][td]70 – 100: Likely to become an average AHL player[/td][/tr]
[tr][td]-4[/td][td]50 – 80: Likely to become an average ECHL player[/td][/tr][/table]

3. Young Players
For young players who have only recently been drafted (i.e. within the last few years) and are harder to evaluate, use the following guide:

NHL Draft Spot: PA Rating
1 - 10: -9
11 - 30: -8
31 - 50: -7
51 - 80: -6
81+: -5


Defensive & Offensive Role
This determines how good the player is at playing as a defenceman (Defensive Role) and as a forward (Offensive Role). Use the following tables as a guide, depending on the player’s position / role within the team.

Defensive Role
18 Top defensive skills (when compared with NHL defencemen)
17
16 Good defensive skills (when compared with NHL defencemen)
15 Average defensive skills (when compared with NHL defencemen)
14 Poor defensive skills (when compared with NHL defencemen)
13
12 A poor two-way forward
11
10 The typical NHL forward
9
8
7 An NHL forward who virtually never contributes to defence - i.e. a floater (very rare)

Offensive Role
18 A top forward who is often one of the top scorers in the NHL
17
16 Good offensive skills (when compared with NHL forwards)
15
14 Average offensive skills (when compared with NHL forwards)
13
12 Poor offensive skills (when compared with NHL forwards)
11
10
9
8
7 A stay at home defenceman

NOTE:
* An offensive defenceman should have a minimum of a 12 Offensive Role. Any less than 12 and he won't be regarded as offensive.
* A two-way forward should have a minimum of a 12 Defensive Role. Any less than 12 and he won't be regarded as a defensive forward.

I'm following the QMJHL a lot and I'm currently doing the rosters for every team in the league including creating the players who aren't in the game. I could send you what I did when I'll be done with it if you want.

c.demers91 wrote:I'm following the QMJHL a lot and I'm currently doing the rosters for every team in the league including creating the players who aren't in the game. I could send you what I did when I'll be done with it if you want.

Welcome to The Blue Line!

Do you have the rosters in csv form (i.e. by using the EHM Updater) or have you been editing the database (i.e. by using the Pre Game Editor)?

c.demers91 wrote:I'm following the QMJHL a lot and I'm currently doing the rosters for every team in the league including creating the players who aren't in the game. I could send you what I did when I'll be done with it if you want.

I have done the rosters but I have not updated the players. I have created some new(top two or three rounds from the draft) players but there are several missing.
Any help would be very much appreciated!
Send me a PM and we can discuss it further

archibalduk wrote:

c.demers91 wrote:I'm following the QMJHL a lot and I'm currently doing the rosters for every team in the league including creating the players who aren't in the game. I could send you what I did when I'll be done with it if you want.

Welcome to The Blue Line!

Do you have the rosters in csv form (i.e. by using the EHM Updater) or have you been editing the database (i.e. by using the Pre Game Editor)?

For now, I've only done a few teams in the Pre Game Editor cause I wanted to play with the new Sherbrooke Phoenix (my hometown team) but I can do it in the EHM updater from now on. I saw that there's a readme pdf so I'll read that to understand everything and then I'll be able to help you guys!

i cant open the excel file. i have microsoft office 2007. When i open them excel bug

kadri43 wrote:i cant open the excel file. i have microsoft office 2007. When i open them excel bug

What do you mean by "excel bug"?

- the Updater exports/imports csv files, not excel
- in csv files things like font/fill colour and formatting don't save



I open the csv file the Updater exports, and then use "save as" to save it as an excel - in excel I don't lose coloured font/fills and formatting

Once I'm done the editing (using excel), and I've removed columns/etc I may have added, I use "save as" again to return it to csv status

Then I import the csv file data using the Updater



Regards

it's okay it works now.
my question is that I dont understand the potential ability.

kadri43 wrote:my question is that I dont understand the potential ability

Current Ability (CA) and Potential Ability (PA) are essentially "overall" ratings

Current Ability (CA) is how good a player is at start-up...how good a player is on day 1 of a brand new game

Potential Ability (PA) is essentially "how good a player will be someday" with the right training/icetime/if injuries are avoided/etc


For really young players who are still developing, a "minus number" is given (like -9 or -7)...this means the player's PA will vary within a range

A -9 PA means a player's PA will end up somewhere between 150 and 180...each time you start a new game the PA will change, but it will always be within the range.

For example, Edmonton's Taylor Hall starts with a CA of 141 and a PA of -9...so each time you start a new game Taylor Hall's PA will change, but it will always be between 150 and 180

Some players have a "set PA" - for example Jeff Carter of the Kings starts with a CA of 160 and a PA of 170, Drew Doughty of the Kings starts with a CA of 160 and a PA of 175

So I'm now doing the Toronto Maple Leafs an d the montreal canadian.

tell me where you guys need help, I'm most familiar with the Toronto Maple Leafs organization. So I could help out with the Marlies. I'm also familiar with the Oilers and Flyers. So I can do those as well. Probably one at a time.

francis246 wrote:tell me where you guys need help, I'm most familiar with the Toronto Maple Leafs organization. So I could help out with the Marlies. I'm also familiar with the Oilers and Flyers. So I can do those as well. Probably one at a time.

I finished the leafs and the montreal canadien. Plz do the oilers and the flyers

alright working on Oilers and flyers right now

Just a question to the administrators.
Did you received my updates?

kadri43 wrote:Just a question to the administrators.
Did you received my updates?

All received! Thank you!

I will take a look this weekend.

Hi,

I have been playing the EHM game since the beginning and really appreciate and love the work that is being done on this site.

I have "studied" or maybe rather been "obsessed" with getting the CA/PA/Attributes ratings as accurate as possible, for every roster update I have played (all of them). I've gotten a real good, realistic understanding of the CA/PA ratings over the years.

What I've always wondered is why don't we use a "baseline" to determine the CA/PA ratings? I see that for every roster update that comes out, people submit their own take on players from a specific team. This will obviously yield differing and biased results. I have used those roster updates over the years and appreciate and enjoy the work that has been done. So I am not critisicing anyone here, on the contrary.

However, I do believe we could make it a little easier. Instead of people evaluating each player individually and assigning a CA/PA that may or may not be a realistic setting, then we could have people match those players up against a baseline. Let me explain:

I work with IT Security and do alot of work with standards and baselines. Take server hardening for instance, say Windows Server 2008 R2, where we take GPO's of industry best practice and match up against what we have and see the difference. We could employ the same "idea" here by using an "industry best practice" for rating Hockey players. This means we could let proffesional people do the rating for us and all we would have to do is match up the player from the "best practice" to a reference table. This best practice could be the one used at .http://www.hockeysfuture.com/playerprojections/

They rate prospects (guys like Mike richards and the class of 2003 are all in their database still, so not only prospects apply) and apply a potential score 1-10 along with a probability of success A-F. They have examples of players for each level of the potential score. That is half our work done for us right there. All we would have to do is come up with a reference table that aligns with the hockeyfutures.com potential score. i.e. Match up 1-10 to 1-200. for instance:
10 would be 190-200
9 would be 170-190
8 would be 155-170
7 would be 140-155
6 would be 130-140
etc

Say we have agreed on a reference table (something along the ones above), the we would do the following:
1. Take a player to update (Brayden Schenn)
2. Look him up in http://www.hockeysfuture.com/prospects/brayden_schenn/
3. They have given him a potential rating of 8.5
4. 8.5 in our reference table is 170
5. Schenn's PA in EHM would then be 170

Idunno, what do you guys think?

Cool idea I think (I love reference materials), but it may not always easily convert/connect...


In Manimal 4.2 Scheen has a PA of -8 (130-160)

If a 9 rating starts at 170, then wouldn’t Scheen’s 8.5 result in a max of about 163?
And his probability of success is a B, meaning he could end up a 7.5, or about 148…

The problem is EHM doesn’t give a PA rating that allows for a 148-165 range…while a -8 (130-160) may be a tad low at the high end and may go to low, the next option (-9) results in a 150-180 range, which is much to high - it seems 21 year old Scheen's current PA of -8 may be the best option available…the point is EHM’s editing limitations need to be considered too IMO

You are absolutely correct. That is why if such an idea was to be adopted, we would need to come up with the best possible reference table (alignment with hockeysfuture.com).

There are many variations, problems and solutions of the issue you are highlighting.

However, if we were to come up with a good alignment (reference table) then we would have a pretty easy system for future roster updates to re-classify players.

One solution could be to "expand" the potential scores to include the .5's:
10
9.5
9
8.5
8
etc
and to be more rigid with the conversion so the range does not wary.

Another way to get around the issue you are describing (which I agree with completely) is to look at it differently. We take Schenn as the example. Hockeysfuture predict him to become somewhere between first line forward/elite talent (8.5). If we use Manimal 4.2 then the -8 (130-160) is too low and not indicative of his real potential. In my opinion -9 (150-180) better fits the first line forward/elite talent potential. Add in the fact he got a "B" for probability of success, meaning he could drop to 7.5 at the very least (still be a first/second line forward), according to my reference table in the previous post a 7.5 would land him at around 150 at the very least. So instead of only looking at the numbers, we would need to look at the definition of the numbers to help classify the player.

I think we actually did something like this when we were updating prospects for an earlier version of the DB. We made a chart where you could match up a HF rating (with both the number and the letter) to a PA. It seemed like a decent method for prospects.

I'm pretty sure I've got it at work. I'll try to remember to post it tomorrow.

I think I know what you mean, B72. I think you're referring to the Manimal Research Sheet. There is a table on that sheet which converts the Hockey's Future ratings to the EHM PA ratings. I have posted the table further below. The table takes into account both the potential score and the success score. I think this probably does the job quite well - but Manimal would be able to confirm this.

I completed agree with citd (welcome to TBL, btw!); I think converting a professional standardisation into EHM's CA/PA system would be a very good way of doing things. This is particularly the case seeing as we have had little interest in helping update the NHL ratings (but thanks to Kadri, cdemers91 and Francis for their kind help - they have been our sole volunteers). What we really need is a way of calculating the CA (and ideally the Offensive and Defensive Role ratings). Is there any standard scoring system from a reputable source which rates players' current ability?

Maybe even something like ESPN Fantasy Hockey? They have a scoring system (see this) but the problem is it only rates current form rather than a player's current ability. For example, Crosby is ranked #147 for forwards because he played so few games last season whereas we all know that in terms of actual ability/skill he'd be easily up in the top 5. TSN also runs a similar ranking system (here) but this is also based on form rather than skill.

Here's the Hockey's Future table I mentioned:

[table][tr][th]Age ->[/th][th]Under 22[/th][th]22+[/th][/tr]
[tr][th]HF Rating[/th][th]PA[/th][th]PA[/th][/tr]
[tr][td]10.0A[/td][td]-10[/td][td]-10[/td][/tr]
[tr][td]10.0B[/td][td]-10[/td][td]-10[/td][/tr]
[tr][td]10.0C[/td][td]-10[/td][td]-10[/td][/tr]
[tr][td]10.0D[/td][td]-9[/td][td]-9[/td][/tr]
[tr][td]9.5A[/td][td]-10[/td][td]-10[/td][/tr]
[tr][td]9.5B[/td][td]-10[/td][td]-10[/td][/tr]
[tr][td]9.5C[/td][td]-9[/td][td]-9[/td][/tr]
[tr][td]9.5D[/td][td]-14[/td][td]-9[/td][/tr]
[tr][td]9.0A[/td][td]-9[/td][td]-9[/td][/tr]
[tr][td]9.0B[/td][td]-9[/td][td]-9[/td][/tr]
[tr][td]9.0C[/td][td]-14[/td][td]-9[/td][/tr]
[tr][td]9.0D[/td][td]-14[/td][td]-8[/td][/tr]
[tr][td]8.5A[/td][td]-9[/td][td]-9[/td][/tr]
[tr][td]8.5B[/td][td]-9[/td][td]-9[/td][/tr]
[tr][td]8.5C[/td][td]-14[/td][td]-8[/td][/tr]
[tr][td]8.5D[/td][td]-14[/td][td]-8[/td][/tr]
[tr][td]8.0A[/td][td]-8[/td][td]-8[/td][/tr]
[tr][td]8.0B[/td][td]-8[/td][td]-8[/td][/tr]
[tr][td]8.0C[/td][td]-14[/td][td]-8[/td][/tr]
[tr][td]8.0D[/td][td]-14[/td][td]-7[/td][/tr]
[tr][td]7.5A[/td][td]-8[/td][td]-8[/td][/tr]
[tr][td]7.5B[/td][td]-8[/td][td]-8[/td][/tr]
[tr][td]7.5C[/td][td]-14[/td][td]-8[/td][/tr]
[tr][td]7.5D[/td][td]-14[/td][td]-7[/td][/tr]
[tr][td]7.0A[/td][td]-7[/td][td]-7[/td][/tr]
[tr][td]7.0B[/td][td]-7[/td][td]-7[/td][/tr]
[tr][td]7.0C[/td][td]-13[/td][td]-7[/td][/tr]
[tr][td]7.0D[/td][td]-13[/td][td]-6[/td][/tr]
[tr][td]6.5A[/td][td]-7[/td][td]-7[/td][/tr]
[tr][td]6.5B[/td][td]-7[/td][td]-7[/td][/tr]
[tr][td]6.5C[/td][td]-13[/td][td]-6[/td][/tr]
[tr][td]6.5D[/td][td]-13[/td][td]-6[/td][/tr]
[tr][td]6.0A[/td][td]-6[/td][td]-6[/td][/tr]
[tr][td]6.0B[/td][td]-6[/td][td]-6[/td][/tr]
[tr][td]6.0C[/td][td]-13[/td][td]-6[/td][/tr]
[tr][td]6.0D[/td][td]-13[/td][td]-4[/td][/tr]
[tr][td]5.5A[/td][td]-6[/td][td]-6[/td][/tr]
[tr][td]5.5B[/td][td]-6[/td][td]-6[/td][/tr]
[tr][td]5.5C[/td][td]-12[/td][td]-6[/td][/tr]
[tr][td]5.5D[/td][td]-12[/td][td]-5[/td][/tr]
[tr][td]5.0A[/td][td]-5[/td][td]-5[/td][/tr]
[tr][td]5.0B[/td][td]-5[/td][td]-5[/td][/tr]
[tr][td]5.0C[/td][td]-12[/td][td]-4[/td][/tr]
[tr][td]5.0D[/td][td]-12[/td][td]-4[/td][/tr][/table]

yeah, that lists works for PA, I think
Are all players rated by HF or only the prospects?

I think it's only prospects that are rated.

Is the new NHL 13 out yet? Could we lift the player ratings from that and figure out a way to convert them to CA?

bruins72 wrote:I think it's only prospects that are rated.

Is the new NHL 13 out yet? Could we lift the player ratings from that and figure out a way to convert them to CA?

Yeah, I purchased a copy recently.

That's a good point. EA is renowned for over-rating players, but perhaps we could compensate for this by reducing the rating (e.g. if a player is rated 100/100 in NHL 13, we could rate him 185 or whatever). So long as the players are correctly ranked by EA then maybe it would be worth taking a look. IIRC the game also has a decent range of role types for each player which probably could help for the basis of Defensive/Offensive Role ratings.

Player ratings can be found here: Western Conf & Eastern Conf (halfway down each page there are links for each team - click on those links and you'll see the list of player ratings).

However, there is mention here that "The ratings were met with high criticism because certain players were given very generous ratings while others were really undersold."

Here's an example of Boston's ratings:

There's a list of most of the NHL 13 ratings here: http://www.ultimatehockey.info/nhl13ratings

Guys,

This is really interesting. We could use the Hockeysfuture ratings for calculating PA and the NHL 13 ratings for CA ability. So, by looking at http://www.ultimatehockey.info/nhl13ratings we can basically say the conversion to CA in EHM would be to take the NHL13 rating and multiply by 2 (since they are using a range of 0-100 and EHM is using 0-200). so we get a table like this for CA:
200 = 100
190 = 95
180 = 90
170 = 85
160 = 80
etc

We can do this for every roster update for EHM, lookup the CA from the newest NHL game and the PA from hockeysfuture.

By the way, hockeysfuture is not only for prospects:
http://www.hockeysfuture.com/prospects/mike_richards
http://www.hockeysfuture.com/prospects/jeff_carter
http://www.hockeysfuture.com/prospects/bobby_ryan

But should probably just be used for players under 22/23 or something along those lines. And it sounds like a good idea to "modify" the nhl ratings slightly. we could just export the ratings and apply a general rule to all, such as "subtract 1 or 2."