EHM1 Regen Testing (2016)

Discuss all aspects of editing the data and databases in EHM here. Have a question about the EHM Editor, EHM Assistant, editing the .cfg files, hex editing the .dat or .db files? Want to tweak the EHM exe file to change league rules/structure, start date etc? This is the place!
Forum rules
This is the forum to discuss all aspects of editing the EHM data and tweaking the game.

Have a bug or feature request for the EHM Editor? Post them in the EHM Editor thread. Please start a new thread or post in another thread if you have a question about how to use the EHM Editor.

Given the large number of questions on similar topics, we ask that you start a new thread for a new question unless you can locate a similar question in an existing thread. This will hopefully ensure that similar questions do not get buried in large threads.

Useful links: EHM 1 Assistant (Download) | EHM 1 Editor (Download) | EHM 1 Editor Tutorials | Editing Rules & Structures Guide | Converting EHM 2004 / 2005 DBs to EHM 1 | Converting an EHM 2007 DB to EHM 1 | Extra_config.cfg | Import_config.cfg | Player Roles
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

EHM1 Regen Testing (2016)

Post by nino33 »

Here's the College/Junior League data from Test 1 (no players/staff added) at 2025
.

CA/PA averages at 2025 by League (as of August 29, 2025)
The left hand side is all players Contracted To (so includes signed minor pros/juniors and is much higher than the number that "play" in the League)
The right hand side is the number of teams x 23 players, so it is reasonably representative of those actually playing in the League)
LEAGUE Players CA average CA range PA average PA range ---------- Players (23 per team) CA average CA range PA average PA range
Atlantic Hockey (NCAA) 188 28.5 3-55 45.4 10-69 ---------- 253
Big Ten (NCAA) 173 45.0 2-113 66.4 10-154 ---------- 138 53.8 20-113 77.1 28-154
ECAC (NCAA) 280 52.4 1-126 75.0 7-146 ---------- 276 53.1 4-126 80.0 10-146
WHL 684 60.8 25-121 89.0 40-166 ---------- 506 66.3 52-121 94.9 61-166
OHL 627 64.2 21-116 92.7 35-172 ---------- 460 69.9 55-116 99.1 73-172
QMJHL 517 63.3 25-123 89.9 41-174 ---------- 414 67.5 53-123 94.2 62-174
USHL 326 46.2 2-118 73.5 10-166 ---------- 391
J20 SuperElit 558 47.2 3-115 75.1 10-160 ---------- 460 52.8 30-115 81.8 42-160
MHL 590 46.1 3-100 75.5 10-159 ---------- 621
Fin A-Jr SM-Liiga 357 46.2 1-95 72.6 10-148 ---------- 368
NOTE - I left the "right hand side" blank if the League didn't have excess players
NOTE - only "found" 27 MHL teams (3 different Leagues are combined under "MHL" in the EHM Assistant!)


I tried to add some Bantam/Midget data too, but it got overwhelming for my brain! HaHa
The Assistant's abbreviations often don't match what the database shows in the Pregame Editor, and to frequently multiple leagues are combined under one abbreviation.....so I'm done for now!

My only comment at this time (I'm tired) is.....I simply don't understand the incredibly low CAs that are in so many leagues - like literally so many leagues that are "normal/average" include players that by their CA are the worst hockey players on the planet! Why do Leagues where the CA average is 50-80 have players with CAs in the 10-20 range, and players below 10 too! - it makes no sense to me! Players in the NLA, HockeyAllsvenskan, the DEL, the Austrian League and more with CAs under 10.....I can't see such players EVER being invited to even a try out

And the Junior/College Leagues too! I can't see players with CAs below 20 even being allowed on the ice, let alone below 10...
Last edited by nino33 on Sun Apr 21, 2019 4:35 pm, edited 2 times in total.
User avatar
Primis
Freeware Moderator
Posts: 1698
Joined: Tue Aug 03, 2010 7:46 pm
Location: Michigan, USA

Re: EHM1 Regen Testing (2016)

Post by Primis »

I would be curious how the WCHA and NCHA collegiate conferences would fare in analysis. WCHA is traditionally one of the strongest and deepest of the NCAA conferences, and the NCHA is a strong one now as well. And Hockey East was left out as well (Boston U, Boston College, etc). ECAC is the very academic arm of NCAA hockey (the Ivy League schools, etc) and is no doubt "weaker". Atlantic Hockey isn't really any stronger. So they're probably not the bets examples for comparison to Major Junior.

As to the CA issues, I agree. I think there should be a large disparity at the level of course, but lots of CA's of 10 seems extreme. However, I'd say for a really good NCAA program that has a good year, a good 1/3 up to even 1/2 of the players on that team will go on to still play hockey *somewhere* else, whether it's ECHL, SPHL, or lower-level overseas.

I wonder if a 1-200 range simply isn't enough to properly mimic the disparity and balance across so many different levels.

Also looking at this, to me, it seems the Q, WHL, and OHL are fairly overrated when compared to NCAA. Hate to be blunt, but the CA and PA averages should NOT be as big a difference as that. There's some bias at work there. The Major Junior teams and leagues are being overly favored there IMHO, which makes me eyeroll because it seems to forever be a struggle. The idea that everyone across the board in Major Junior (at only 18 and 19) is that much better than the 20, 21, and 22 year olds in NCAA across the board is a problem. I'm speaking purely CA-wise.
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Primis wrote:I would be curious how the WCHA and NCHA collegiate conferences would fare in analysis. WCHA is traditionally one of the strongest and deepest of the NCAA conferences, and the NCHA is a strong one now as well. And Hockey East was left out as well (Boston U, Boston College, etc). ECAC is the very academic arm of NCAA hockey (the Ivy League schools, etc) and is no doubt "weaker". Atlantic Hockey isn't really any stronger. So they're probably not the bets examples for comparison to Major Junior.
They were never meant as a comparison, they're only in the same "group" because they're not Pro Hockey
I spent quite awhile looking for Hockey East, couldn't find it and was tired
As I tried to note the EHM Assistant does not make such work easy to do!

Primis wrote:I wonder if a 1-200 range simply isn't enough to properly mimic the disparity and balance across so many different levels.
I suspect such an idea isn't even open for discussion (can't radically change/recode the game)


Primis wrote:Also looking at this, to me, it seems the Q, WHL, and OHL are fairly overrated when compared to NCAA. Hate to be blunt, but the CA and PA averages should NOT be as big a difference as that. There's some bias at work there. The Major Junior teams and leagues are being overly favored there IMHO, which makes me eyeroll because it seems to forever be a struggle. The idea that everyone across the board in Major Junior (at only 18 and 19) is that much better than the 20, 21, and 22 year olds in NCAA across the board is a problem. I'm speaking purely CA-wise.
If there's bias, then I think it's yours :-D

I spent 5-6 months a year on Major Junior for years.....I'm not sure we ever had an NCAA researcher during that time
And I think you've taken some general data and extrapolated it in a way I never would

I really don't think claiming bias and eyerolling are helpful in any way, and they are for sure NOT helpful
I get that you're trying, as you've actually offered to try and help me!
But I think I'm the only one that does this, and makes an effort to be public.....there's a reason why the community hasn't gotten this kind of information from others IMO (including Riz), as "the flack" is just not worth it
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

I wish people would realize that if somebody (me!) spends dozens/hundreds of hours on something, and explains repeatedly about the difficulties and challenges encountered while doing so, and repeatedly points out the lack of researchers, and repeatedly expresses the frustration of false accusations and ignorant claims (check out the flack I've received in recent days at SI) and personal attacks, and repeatedly responds negatively to people "expressing" their "rights/opinions" as if they too had put in dozens/hundreds of hours - the end result could very well be I'll just give up

Thanks to Manimal/CJ/Alessandro for the support I've received, it's helped more than you'll likely ever know

I'm sorry if people think I'm bad/rude for posting like this...but I just don't understand you/your views
I would NEVER treat someone like me the way I'm to often treated
When someone is volunteering lots of their time and produces significantly positive/helpful results, and all they're asking for is a little thoughtfulness/compassion/understanding (and stop with the hyperbole/negativity), why can't they (me!) have it?

I have "stepped away" in recent weeks, as I asked to be removed from the Researchers/Veterans Forums to try and "help myself" avoid negative feelings (and may have lost my "connection" to Riz as a result, I don't know), and I've pretty much given up on EHM reddit and thinking maybe I should on SI as well.....but these are not good things IMO; I want to help and have a lot to offer, and it's not intrinsically better/good if I do less


I felt so much better lately, as I don't feel "connected" (obligated, my issue!) to others in a way that makes me feel uncomfortable/dishonest - I like to communicate about EHM , and help with EHM (I think I do TONS for the community), and I like that now I can "say my peace" and if people choose to reject it/me maybe that'll help me be in less pain if/when I do quit all things EHM (because I do like helping/I think I am helpful & I think what I'm asking for is IMO excessively reasonable)

FYI - the "dishonest" references to how I feel when I try to do what others seem so much better at then me...forget about things/don't care/don't respond/don't be bothered - I try to work things out, and work THROUGH things & don't seem capable of letting things not bother me "like water of a duck's back" (and I feel dishonest/uncomfortable trying to be someone I'm not)


So to the community....your support/understanding would be appreciated :-)


P.S. FYI I'm not looking for a "written response" so no one has to feel they need to give one IMO (I suspect it'd be best if there wasn't)
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Primis wrote:I would be curious how the WCHA and NCHA collegiate conferences would fare in analysis. WCHA is traditionally one of the strongest and deepest of the NCAA conferences, and the NCHA is a strong one now as well. And Hockey East was left out as well (Boston U, Boston College, etc).
I'll do my best to include them with the 2035 testing results! :thup:

I think the 2035 testing results will be even more representative of what the game does (which is what we're largely looking for/some guidance for the researchers).....2025 is mostly players from the database
User avatar
Primis
Freeware Moderator
Posts: 1698
Joined: Tue Aug 03, 2010 7:46 pm
Location: Michigan, USA

Re: EHM1 Regen Testing (2016)

Post by Primis »

nino33 wrote:If there's bias, then I think it's yours :-D

I spent 5-6 months a year on Major Junior for years.....I'm not sure we ever had an NCAA researcher during that time
And I think you've taken some general data and extrapolated it in a way I never would

I really don't think claiming bias and eyerolling are helpful in any way, and they are for sure NOT helpful
I get that you're trying, as you've actually offered to try and help me!
But I think I'm the only one that does this, and makes an effort to be public.....there's a reason why the community hasn't gotten this kind of information from others IMO (including Riz), as "the flack" is just not worth it
I don't think you're at all understanding what I'm saying, nino.

You're into 2025. You took a snapshot of rosters in 2025. There should be no human-created players at these junior levels now, right? The game is generating them (for reference, two entire classes of freshmen in NCAA could have come and gone/graduated, plus still another couple of years... two "generations" of NCAA players essentially). There shouldn't be anything to attack or discuss researcher-wise. League research has nothing to do with it. At this point 10 years in, it's all game-gen, right? You don't have a bunch of 7 and 8 year old Americans in the starting DB to 10 years down the road populate NCAA's do you?

The bias I'm noting here is game bias: that the major junior levels are for some reason showing a much higher game-gen average CA than NCAA, despite the fact that NCAA teams can feature players that are sometimes years older (and therefore have had more development time on them). That doesn't really make sense to me. I don't see it as a DB thing, then.

If it were 18 and 19 year olds vs 18 and 19 year olds, I could maybe see it. But NCAA teams can sometimes have 23 and 24 year olds on them depending on how things all sort out (injuries gaining an extra year of eligibility, etc) and regularly feature 21 an 22 year olds, and I'm sorry.... to me it's just sheer logic that even if their PA's are somehow lower they still have had extra years of development. We're not talking PA, we're talking CA, and through sheer age/having more years to develop in many cases, the NCAA ones should be a lot closer to the major junior ones. I could see the major junior ones being marginally better or something, but the idea that the disparity in average player CA is that big seems very, very wrong.

Do you see what I'm getting at? It's also why I was curious to see what the WCHA and Hockey East was like. Because maybe their CA average is higher still than the others. I can't comment on the european ones and how they fit, because I have no frame of reference. I do for the North American ones though.
User avatar
Manimal
TBL Admin Team
Posts: 6344
Joined: Thu Apr 24, 2008 4:01 am
Custom Rank: EHM Rosters Man
Favourite Team: Djurgårdens IF
Location: Karlstad, Sweden

Re: EHM1 Regen Testing (2016)

Post by Manimal »

The things we could look at from a database point is league and team reputations. However, comparing the KHL and SHL averages shows it might not matter. Strange to see the KHL averages so low.
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Primis wrote:I don't think you're at all understanding what I'm saying, nino.
No, I think you're missing my point
It was late, I was tired, and the online site I found with "top NCAA hockey leagues" listed leagues I had a hard time finding because of the challenges in using the Assistant (and I recognized the names in the Hockey East and that's why I spent awhile trying to find it...I had to redo the whole exporting data process to find anything for the UK and NCAA because I hadn't included teams initially (which me involves multiple computers/OneDrive and long waits because my old laptop's so slow and then cleaning up the data because of issues with the Assistant)

By your own admission the top NCAA Leagues weren't included, and so I simply wouldn't extrapolate the significant meaning that you did ("eyeroll" & "seems to forever be a struggle" & "bias" are all "loaded words" don't you think?)

And you outright said "it seems the Q, WHL, and OHL are fairly overrated when compared to NCAA"
#1 that's my work being referenced as "overrated" (another loaded word)
#2 there was NEVER any intention of comparing them (so no reason to call them overrated)
#3 the NCAA leagues are by your own admission not represantitive

Primis wrote:Do you see what I'm getting at?
Yes, I just don't agree
I don't think your comments connect with how the game actually works
The whole concept of the "underdatabase" is based on this view I have http://www.ehmtheblueline.com/forums/vi ... researcher
If things aren't set up right/best by a human being in the first place, the game does not "create" well by itself or "repair" things well either IMO (especially in regards to non-playable leagues!) & I think it mostly just essentially repeats what's there (while a dynamic/intelligent database would be awesome, it's a Wish List item IMO)



P.S. I think the NHL and the main "feeder leagues" probably run very well to extremely well (as those are what almost everyone plays) and then it gets less-and-less.....for the NCAA I suspect the standard is "does it produce enough guys for the NHL" and given we haven't actually looked at the quality NCAA leagues I'd say we don't even know there's a problem!

I get that you care enough bout the NCAA to watch it/read about it/discuss it.....care enough to research it? :-D http://www.ehmtheblueline.com/forums/vi ... r&start=25

It's a non-playable league that's not thoroughly researched and doesn't have a significant "demand" that it be made playable.....I suspect that Riz (and others like myself) would LOVE for things to be better/maybe the way you describe, but we can honestly see our list of "things to do" already and know we probably won't be adding to it (I suspect the NCAA runs "good enough" for now, as a non-playable league.....there certainly aren't a host of complaints about it)

I talked to one researcher who didn't want to research "their country" until Riz made it playable first
I thought the chances of Riz adding to his list of things to do to make it playable (given there was no demand for it, like the NCAA) were far less likely if the league's not researched


I just don't understand your starting your position with "these aren't the right/representative leagues" and then start making comparisons/extrapolations based on them
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

nino33 wrote:a dynamic/intelligent database would be awesome
Especially for a retro database! Especially for the 1974 database!

The thing about a retro database is the talent available is going to increase significantly over time, as you're actually adding talent ever year (in addition to any regens!).....so staring in 1974, I wonder if/when it becomes "to noticeable/problematic" to be enjoyable?

I'm working on the 1974 database today.....hopefully later this year I'll be able to have an updated/new EHM version of the 1974 database and find out the answer

Though I've thought "at the least/for now" with the EHM Assistant one could remove top fake players/regens (by lowering their PA when they're first created)
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Primis wrote: You don't have a bunch of 7 and 8 year old Americans in the starting DB to 10 years down the road populate NCAA's do you?
7 and 8 year olds no, but a starting database does have players start at age 9 and 10 (but you don't see them in game until they're 14) :-)

So a researcher could put some young Americans in there, and ensure their Preference is College and maybe include some top NCAA teams as Favourite Teams :-D


Now I've got Bowie going through my head :-) "Young American, young American, she wants the young American"
User avatar
Manimal
TBL Admin Team
Posts: 6344
Joined: Thu Apr 24, 2008 4:01 am
Custom Rank: EHM Rosters Man
Favourite Team: Djurgårdens IF
Location: Karlstad, Sweden

Re: EHM1 Regen Testing (2016)

Post by Manimal »

nino33 wrote:
Primis wrote: You don't have a bunch of 7 and 8 year old Americans in the starting DB to 10 years down the road populate NCAA's do you?
7 and 8 year olds no, but a starting database does have players start at age 9 and 10 (but you don't see them in game until they're 14) :-)

So a researcher could put some young Americans in there, and ensure their Preference is College and maybe include some top NCAA teams as Favourite Teams :-D

I thought this test was made with the modern-db?
That one does not have 9 or 10 year olds?

Primis was not referring to what you have put into the db. He was commenting on what he (and I) thought were game produced (modern db in 2025).

You might not have meant to compare things but everyone who reads the numbers will compare them to eachother.
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Manimal wrote:I thought this test was made with the modern-db?
That one does not have 9 or 10 year olds?
All databases have such players, unless you select "Disable All Extra Players" (as of version 1.2)
I believe they were put in to improve the draft quality between versions 1.0 and 1.1
The March testing showed 1.1 generated WAY to many stars/superstars (page 1 of this thread)


Manimal wrote:Primis was not referring to what you have put into the db. He was commenting on what he (and I) thought were game produced (modern db in 2025).
The whole concept behind the "underdatabase" is the game adds players/staff and they're unrealistic/problematic/nowhere near as good as what a human researcher provides.....the blank database testing shows the game seems to lack the ability to "create a quality database on it's own" - this thread has some comments about starting with a blank database with EHM07 http://www.ehmtheblueline.com/forums/vi ... lit=+blank including references to the low CA/PA players generated

I guess I thought it was more known/accepted that the game seemingly doesn't really "produce by itself from scratch" very well/needs the initial human inputted guidelines for best results. I wonder if this is in part (large part?) to Riz making changes after the initial release of the new EHM so the game didn't change so many of the database values (because initially the game was changing many researcher edits!).....maybe this is why there's seemingly more of an in general repeating of what's initially inputted/less game overwriting of actually inputted values.....and so "researcher issues/errors" are essentially repeated when players regen

I did some blank database testing last year/start of this year (but if I posted the results they must be in the Researchers Forum).....IIRC last year the blank database testing confirmed "the 10/10 bug" (that the public never did seem to realize existed) where thousands and thousands of players had an Offensive/Defensive Role ratings of 10 & then the blank database testing in January was to confirm the problems with low Offensive/Defensive Role totals (the game's all added up to close to 20 while researcher inputs were commonly in the mid/high 20s and even mid/high 30s for stars/superstars - this is when Riz went with the guidelines/suggestions I provided from looking at what we'd done in the database)

These are just a few examples...I'm sure there's more
Maybe the game's gotten better, but as it takes so much time to do such testing/compiling/comparing I haven't looked at it closely lately (and didn't intend to.....I'm promoting human researcher input if/whenever possible!)


Manimal wrote:You might not have meant to compare things but everyone who reads the numbers will compare them to eachother.
I understand that, and would totally accept it if the reasoning/comaprison didn't start with "these NCAA leagues aren't the good ones/they don't represent the NCAA"


P.S. I don't recall what the suggested NCAA ranges were in the Researchers Forum, but I'm pretty sure the 07 standard was "same as Major Junior" so the low NCAA ratings are a researcher issue IMO (and the game repeats the issue moreso than corrects/fixes it) - for example, Hockey East in TBL 8.2 has 114 rated players and a CA average of 49.4 with a CA range of 27-93 & top PAs are -14 players (five), -7 players (four) and -13 players (twenty-five)
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

nino33 wrote:The whole concept behind the "underdatabase" is the game adds players/staff and they're unrealistic/problematic/nowhere near as good as what a human researcher provides
nino33 wrote:I don't recall what the suggested NCAA ranges were in the Researchers Forum, but I'm pretty sure the 07 standard was "same as Major Junior" so the low NCAA ratings are a researcher issue IMO
But there's not enough researchers! not enough time to do the research! and non-playable leagues are going to naturally get less attention (especially when there's nowhere near enough researchers to cover the playable leagues)

The volunteers really are doing the best they can :-)
User avatar
Primis
Freeware Moderator
Posts: 1698
Joined: Tue Aug 03, 2010 7:46 pm
Location: Michigan, USA

Re: EHM1 Regen Testing (2016)

Post by Primis »

I could WALL O TEXT several times here with lots of quotes and everything but instead I'm going to keep it down and concise as possible because this thread has already gone a-waaaaaaaaay from its intent IMO, and I think some things are being seen as criticism that have nothing to do with criticism, and we're kinda' on the verge of the whole thing collapsing and people tossing their arms up in frustration.
nino33 wrote:I guess I thought it was more known/accepted that the game seemingly doesn't really "produce by itself from scratch" very well/needs the initial human inputted guidelines for best results.
No.

At no point do I ever recall that being something that was a "given" or common knowledge, and IIRC you yourself posted a few posts back that you weren't even sure about that either. So now honestly i'm just more confused than anything.

What I am aware of was/thought was that EHM used to do "regens" based heavily on retiring players, and that it does not do so now. My understanding was that EHM has moved on more toward simply generating its own players based independent of what was already there retiring individual player-wise. My understanding was/is that it *tries* to generate certain ratios of players, but otherwise is more random and truly "generated" than it was previously. It also explained why some players are generated weirdly, and my understanding was that as player role gets further implemented that it would slowly do a better job generating specific roles at specific ratios as well. But it's early on.

And to be honest, that's part of what I thought this particular testing was setting out to prove/disprove, is how it was using the starting DB to generate players.


RELATED: to be honest, if we think it DOES base off what is already present in the starting DB heavily, then... isn't that going to be a mess to test? Because that's going to require to also testing entirely different DB's as well, no, to really get an idea of things? I'm not sure what would serve as a vastly different starting DB except a couple of the retro ones...


Please let's not turn this into Researchers vs The World though? That's going to end very poorly, very quickly, all-around, and going to discourage people all around.

We're all trying to do the same thing here, right? There are no sides or attack here.
User avatar
Manimal
TBL Admin Team
Posts: 6344
Joined: Thu Apr 24, 2008 4:01 am
Custom Rank: EHM Rosters Man
Favourite Team: Djurgårdens IF
Location: Karlstad, Sweden

Re: EHM1 Regen Testing (2016)

Post by Manimal »

nino33 wrote:But there's not enough researchers! not enough time to do the research! and non-playable leagues are going to naturally get less attention (especially when there's nowhere near enough researchers to cover the playable leagues)

The volunteers really are doing the best they can :-)
Yes, they are
We even have a new researcher looking at the NCAA.
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

nino33 wrote:These are just a few examples...I'm sure there's more
think there's additional examples of concerns about how the database does things throughout the Player Role tests for forwards http://www.ehmtheblueline.com/forums/vi ... 10&t=16776 and defensemen http://www.ehmtheblueline.com/forums/vi ... 10&t=16775

My many comments in red note a number of significant issues/weirdness IMO of how the databases handles things
Also, sometimes the differences between the 70-99 CA group and the 100-139 CA group and the 140-159 CA groups don't "progress as expected" and/or have significant differences in how Key/Essential/Non-Essential/Irrelevant Attributes are handled in relation to the "average Attribute value for CA" (the connection/relation between the two doesn't always follow/connect as expected).....as I recall in its seemed to often things always made the most sense the higher the CA (as the higher the CA, the more likely it was a human researcher inputting an actual value)

Also distribution maybe seems off in some cases (should we really have far more Physical Playmaking Forwards than the combined totals of Regular/Finesse Playmakers? Are there really more Power Forwards in hockey as there Playmaking Forwards? seems to me this is closer to my childhood hockey than current/modern hockey

Are there really more than twice as many Physical Defensive Defensemen as there are Regular/Finesse Defensive Defensemen combined? in an era where more and more hitting's being taken out of the game? I wonder if this affects PIM totals...

Are there really less than 650 Playmaker Defensemen worldwide? and less than 600 Pointman Defensemen worldwide? Compared to 7700 Defensive Defensemen (plus another 1100 Rugged Defensemen) and 3200 Standard Defensemen (pretty much the worst of the D it seems, not good at particular good at anything)

And of these over 12,000 defensemen referenced above only 22 have a 160+ (Good NHL) CA and only 61 have a 130-159 CA
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Primis wrote:because this thread has already gone a-waaaaaaaaay from its intent IMO
I started the thread/am providing the data for discussion
I was/am OK with the discussion (I think it's needed)
It can be hard to discuss things sometimes I agree, but as mentioned I'm into "working through them" as much as possible

Primis wrote:
nino33 wrote:I guess I thought it was more known/accepted that the game seemingly doesn't really "produce by itself from scratch" very well/needs the initial human inputted guidelines for best results.
No.

At no point do I ever recall that being something that was a "given" or common knowledge, and IIRC you yourself posted a few posts back that you weren't even sure about that either. So now honestly i'm just more confused than anything.
Sorry for any confusion/misunderstanding
When I wrote that I was responding directly to Manimal (and thinking of CJ/Alessandro/Named too)

To everyone/anyone reading these posts/what I post.....please keep in mind there's going to be LOTS that's common knowledge to me and a few others and not to the general public (yet).....and while I'm happy to try and bring things "out into the light" more, there's still some things I can't say, because I always consider anything Riz tells me that's not public knowledge as something I shouldn't share without his permission - I recently shared the Key/Essential/Non-Essential/Irrelevant Attributes for all Player Roles after getting the OK to do so from Riz (critical information IMO)

Over time my posting publicly should just help IMO
Anyone interested can gain a better understanding, hopefully enjoy the game more and give better feedback when they do


Primis wrote:What I am aware of was/thought was that EHM used to do "regens" based heavily on retiring players, and that it does not do so now. My understanding was that EHM has moved on more toward simply generating its own players based independent of what was already there retiring individual player-wise. My understanding was/is that it *tries* to generate certain ratios of players, but otherwise is more random and truly "generated" than it was previously. It also explained why some players are generated weirdly, and my understanding was that as player role gets further implemented that it would slowly do a better job generating specific roles at specific ratios as well. But it's early on.
Some misconceptions here.....
  • Riz has said that they're NOT newgens, they're a combination of newgens and regens (I think the evidence seems to indicate they're essentially regens that aren't as easy to "find" because more information can be randomly changed; I think making "regen hunting" not work was all that was needed myself)
  • the database is not dynamic, it doesn't "read the talent level" and produce players accordingly (that's why I said already "a dynamic/intelligent database would be awesome" - but it's a Wish List item); all evidence I've seen (and I've been saying this for a year) says that regens are essentially the same as they were (really it's only PA that matters, to ensure the database "talent pool" doesn't get to high/low)
  • regarding "players generated weirdly" - you didn't notice this problem in EHM05/EHM07? don't you think it's better than it was in EHM05/EHM07? but still an issue, right? It's precisely because the game itself doesn't have all the AI to better create players that you get such weirdness! weirdness a good human researcher would never include (my whole concept behind the underdatabase is to not allow such weirdness to occur/try to fix such weirdness as much as possible before release of a database)
  • I've never heard anything referenced to Player Role ratios until I mentioned it
Primis wrote:And to be honest, that's part of what I thought this particular testing was setting out to prove/disprove, is how it was using the starting DB to generate players.

RELATED: to be honest, if we think it DOES base off what is already present in the starting DB heavily, then... isn't that going to be a mess to test? Because that's going to require to also testing entirely different DB's as well, no, to really get an idea of things? I'm not sure what would serve as a vastly different starting DB except a couple of the retro ones...
I started doing this/seeing this 5 years ago when working on the 1974 database
A lot of what I was told couldn't be done with Attributes actually could be done (and BobMcGoo proved that some more with his Realism Patch)
So for me what you're just discovering/finding confusing I've seen for many years (I did blank database testing with EHM07 and the new EHM, and testing with the 1974 database and the 1998 database and multiple modern/TBL databases.....so what you're suggesting has already ben done to, literally years ago)

You think there's lots of "walls of text" here? - there's even more in the Researchers Forum! HaHa and some of my PMs HaHa
All everything I've brought out from the Researchers Forum recently and posted publicly I did this year, in 2016 (most just in the last few months)
I've been doing such testing/compiling the data/sharing the results for 5-6 years now (it's just it was shared in the Researchers Forum, or directly to couple/few researchers or just Riz, and now it's public.....so keep in mind that many new issues to you/others are likely not new to me)


Primis wrote:Please let's not turn this into Researchers vs The World though? That's going to end very poorly, very quickly, all-around, and going to discourage people all around.
I'm honestly 100% not trying to! Not at all!
And I'm sure others would like you to say Nino vs The World HaHa


I went public/dropped out of the Researcher/Veteran Forums so I could speak just for myself
No one that's uncomfortable has to deal with it/me, and those that see value can
And if things don't work out and I quit all things EHM and move on I'm not directly hurting anyone/letting them down

Primis wrote:We're all trying to do the same thing here, right? There are no sides or attack here.
OK, I'm going to tell the truth.....
From an emotional point of view, I absolutely want to think what you're saying is true, but from a practical point of view it's not IMO

We are not ALL doing the same things here
A handful of us are doing almost all the research, and have been for YEARS
I don't see anybody but me doing any real testing

Can you point me to similar testing done by anybody regarding Player Roles or regens or anything EHM related? Because I don't think you can, so we're not all doing the same thing (I honestly see it as some have no interest in reading/understanding what they need to first, and think their opinion is just as valid as someone who's put in the effort, and they argue points in such a way that it's not proof/evidence of anything to me)

I think I've posted more testing data/information on EHM then all others combined
I simply don't see anyone else doing what I'm doing (and honestly, I think I wish there was as much or more than anybody!)

Do you realize I worked on Major Junior in 2012, 2013, 2014 and 2015 and never once wanted to?
I only did it to help out Manimal/the Community (I've never played a full season with a modern DB in six years!)

So to be honest, no, it often doesn't always feel like we're all pulling in the same direction

I spent many hours responding to and setting up potential researchers in the first half of this year, and not one person produced anything.....almost all volunteers produce nothing, ever.....it's been pretty much the same very small group of people doing everything for years

I'm all for trying to explain things to people, or help people understand/solve a problem, but I'm not thinking we're all on the same side because I don't find I'm getting much back myself and I get uncalled for "flack" (and I think it's disrespect, because I've clearly put the time and effort in to know what I'm talking about as far as I'm concerned).....my TBL/EHM experience started to "fall apart" quite awhile ago over this issue, when someone was sure they were right, I showed actual evidence (testing data) that they couldn't be, and I still had to endure them thinking they were "right" (I think now it'd be easier for me not to care/not be so affected if it happened now, because I'm just a public member, and not actively leading anything or officially involved in anything - that's what I did in two SI threads recently...I did my best and got "unnecessary flack" IMO and then I said "I'm done with this" and I moved on; not always easy for me so I feel good about it)

I don't accept that my THOUSANDS of hours of EHM testing/reading/studying/editing means nothing and "I'm just like everybody else"
Even avoiding posting answers more and more, I still answer more general questions and get thanked far more often than I have conflict! HaHa
And those that I do still talk to are the ones actually doing the work on EHM, so I think there's value in what I'm doing
User avatar
Alessandro
Olympic Gold
Posts: 2865
Joined: Sat Aug 27, 2005 1:54 pm
Custom Rank: TBL Rosters Man
Favourite Team: Team Russia
WHL Team: Calgary Flames

Re: EHM1 Regen Testing (2016)

Post by Alessandro »

Manimal wrote:The things we could look at from a database point is league and team reputations. However, comparing the KHL and SHL averages shows it might not matter. Strange to see the KHL averages so low.
It is not too surprising that the KHL averages is low after 10 years. KHL teams are nowhere as active as needed in the game, and there are a few leagues who spend an unrealistic amount of money (DEL especially, the SHL and Allsvenskan a little bit). In my games I see too many high Russian talent playing in neither the NHL and the KHL and this is not realistic (Prokhorkin in Germany, Gusev in Switzerland, Nalimov in Sweden, etc.)
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Alessandro wrote:
Manimal wrote:The things we could look at from a database point is league and team reputations. However, comparing the KHL and SHL averages shows it might not matter. Strange to see the KHL averages so low.
It is not too surprising that the KHL averages is low after 10 years. KHL teams are nowhere as active as needed in the game, and there are a few leagues who spend an unrealistic amount of money (DEL especially, the SHL and Allsvenskan a little bit). In my games I see too many high Russian talent playing in neither the NHL and the KHL and this is not realistic (Prokhorkin in Germany, Gusev in Switzerland, Nalimov in Sweden, etc.)
From the Test 1 2025 data spreadsheet I sorted by "First Nation" for Russia and then by CA - here's a look at all Russians with a CA of 100-200 and where they're playing in 2025

Player Name - League - CA - PA
Yakov Trenin NHL 190 190
Vladimir Tarasenko NHL 174 174
Ivan Barbashev NHL 173 173
Evgeny Kuznetsov NHL 171 171
Grigori Denisenko Kontinental Hockey League 165 165
Artemi Knyazev NHL 159 174
Andrei Vasilevskiy NHL 159 159
Nikolai Goldobin NHL 158 158
Dennis Yan NHL 158 158
Evgeny Svechnikov NHL 157 157
Mikhail Grigorenko NHL 156 156
Artemi Panarin NHL 154 158
Nail Yakupov NHL 152 152
Grigory Petrov NHL 151 168
Andrey Lukin NHL 151 154
Sergei Bobrovsky NHL 150 170
Nikita Zaitsev Kontinental Hockey League 149 149
Matvei Zaseda NHL 147 157
Stanislav Sinyugin NHL 146 155
Dmitry Sinitsyn NHL 146 146
Nikolai Kovalenko NHL 145 160
Semyon Varlamov --- 145 164
German Naumov AHL 145 156
Kirill Nizhnikov Kontinental Hockey League 145 145
Evgeni Malkin NHL 144 180
Nikita Kucherov NHL 144 146
Ilya Nazarevich Kontinental Hockey League 144 144
Alexander Ovechkin NHL 143 186
Sergei Koroteyev Kontinental Hockey League 143 159
Nikolai Prokhorkin Kontinental Hockey League 143 149
Denis Golubev Kontinental Hockey League 142 155
Nikita Zadorov Kontinental Hockey League 142 142
Viktor Antipin Kontinental Hockey League 141 141
Dmitry Makarov AHL 140 152
Ilya Sorokin Kontinental Hockey League 138 138
Vladislav Sukhachyov NHL 136 158
Danila Galenyuk NHL 136 149
Anton Shenfeld Kontinental Hockey League 136 140
Valeri Nichushkin NHL 136 136
Vladislav Namestnikov NHL 135 140
Alexander Zhabreyev Kontinental Hockey League 132 153
Mark Skutar Kontinental Hockey League 132 144
Andrei Svechnikov NHL 131 172
Denis Alexandrov Kontinental Hockey League 131 156
Boris Perevozchikov NHL 130 152
Vladimir Plotnikov Kontinental Hockey League 130 151
Mark Razumnyak Kontinental Hockey League 130 148
Pavel Buchnevich NHL 130 145
Dmitry Kulikov NHL 130 144
Mark Fomich NHL 130 144
Pavel Dorofeyev NHL 129 142
Dmitri Sokolov NHL 128 155
Ivan Provorov NHL 128 153
Danil Bashkirov NHL 128 151
Denis Guryanov NHL 128 146
Evgeny Andrusovich Kontinental Hockey League 128 146
Maxim Bobrov Kontinental Hockey League 128 146
Vladimir Fyodorov AHL 128 144
Nikolai Starchenko Kontinental Hockey League 128 144
Leonid Petrov AHL 128 141
Valentin Zykov Kontinental Hockey League 128 142
Maxim Chudinov Kontinental Hockey League 128 141
Ivan Karpov Kontinental Hockey League 127 139
Alexander Barabanov Kontinental Hockey League 126 145
Alfred Kuzmin NHL 126 132
Artyom Alyayev Kontinental Hockey League 126 132
Roman Churekov AHL 126 127
Mikhail Bitsadze Kontinental Hockey League 126 127
Valery Kuzmin Kontinental Hockey League 126 126
Yegor Yakovlev Kontinental Hockey League 125 131
Givi Vinogradov Kontinental Hockey League 124 145
Alexander Dergachyov NHL 124 135
Ignat Shakhvorostov NHL 124 130
Nikita Pivtsakin Kontinental Hockey League 124 133
Vasily Demchenko Kontinental Hockey League 124 131
Yegor Zaplatnikov Kontinental Hockey League 124 130
Alexei Kraskovsky NHL 124 126
Nikita Gusev Kontinental Hockey League 122 145
Alexander Burmistrov NHL 122 142
Fyodor Malykhin Kontinental Hockey League 122 141
Nikolai Glukhov SHL 122 133
Danil Kaskov Vysshaya Liga 122 129
Kirill Slepets Kontinental Hockey League 122 137
Ilya Samsonov Kontinental Hockey League 122 136
Alexander Khovanov AHL 122 130
Zakhar Arzamastsev Kontinental Hockey League 122 131
Dmitri Yudin Kontinental Hockey League 122 131
Nikolai Vladimirov Kontinental Hockey League 122 130
Yaroslav Dyblenko Kontinental Hockey League 122 129
Ivan Sentyurin NHL 122 125
Mikhail Kochetkov Kontinental Hockey League 122 123
Vladislav Kara Liiga 122 123
Dmitri Rodionychev Kontinental Hockey League 121 121
Ivan Romanov Kontinental Hockey League 121 121
Artem Anisimov Kontinental Hockey League 120 148
Nikita Filatov Kontinental Hockey League 120 141
Stepan Khripunov AHL 120 133
Nikolai Skladnichenko Kontinental Hockey League 120 135
Vyacheslav Dubrovsky Vysshaya Liga 120 124
Alexander Alexeyev Vysshaya Liga 120 122
Yegor Korshkov Kontinental Hockey League 120 133
Andrei Kuzmenko Kontinental Hockey League 120 130
Ilya Avramenko Kontinental Hockey League 120 120
Vladislav Gavrikov Kontinental Hockey League 120 120
Stanislav Kandzyuba Kontinental Hockey League 120 120
Damir Sharipzyanov Kontinental Hockey League 120 120
Alexander Galkin AHL 119 132
Sergei Alexeyev AHL 118 138
Oleg Skriabin Kontinental Hockey League 118 136
Evgeny Grigorenko Kontinental Hockey League 118 129
Anton Alexeyev Kontinental Hockey League 118 125
Evgeny Poltorak Kontinental Hockey League 117 155
Mark Pavlikov Kontinental Hockey League 117 128
Artyom Zemchyonok Kontinental Hockey League 117 126
Valentin Tarasenko NHL 116 156
Rustem Kuklev NHL 116 150
Vitaly Kozhevnikov AHL 116 132
Kirill Golubev Kontinental Hockey League 116 132
Ivan Afanasyev Kontinental Hockey League 116 127
Anton Belov Kontinental Hockey League 115 149
Alexei Murygin Kontinental Hockey League 115 141
Andrei Zubarev Vysshaya Liga 115 150
Vyacheslav Psarev Kontinental Hockey League 115 130
Igor Myasishchev Kontinental Hockey League 115 129
Pavel Poryadin Kontinental Hockey League 115 122
Semyon Afonasyevsky Kontinental Hockey League 115 120
Dmitry Kugryshev Kontinental Hockey League 114 146
Rushan Rafikov Kontinental Hockey League 114 137
Vladimir Bryukvin Kontinental Hockey League 114 136
Sergei Shibayev AHL 114 132
Maxim Kucherov Kontinental Hockey League 114 132
Alexander Kadeikin Kontinental Hockey League 114 131
Ivan Vereshchagin Kontinental Hockey League 114 131
Albert Yarullin Kontinental Hockey League 114 131
Dmitri Yushkevich Kontinental Hockey League 114 130
Anton Krasotkin Vysshaya Liga 114 127
Ivan Rotenberg ALIH 114 132
Igor Ozhiganov Kontinental Hockey League 114 129
Sergei Kuptsov Kontinental Hockey League 114 128
Aleksandar Mikulovich Kontinental Hockey League 114 128
Egor Popov Kontinental Hockey League 114 128
Sergei Denisov Kontinental Hockey League 114 126
Mikhail Plotnikov Kontinental Hockey League 114 126
Maxim Uskov Kontinental Hockey League 114 126
Nikita Kamalov Kontinental Hockey League 114 125
Artur Lauta Kontinental Hockey League 114 124
Alexander Polunin Kontinental Hockey League 114 123
Igor Shestyorkin Kontinental Hockey League 114 123
Igor Zenchikov Kontinental Hockey League 114 123
Damir Zhafyarov Kontinental Hockey League 114 123
Dmitri Volkov Kontinental Hockey League 114 120
Nikita Glukhov Kontinental Hockey League 113 123
Andrei Vasilyev Kontinental Hockey League 112 138
Vladimir Tkachyov Kontinental Hockey League 112 135
Vyacheslav Osnovin Kontinental Hockey League 112 127
Alexander Delnov Kontinental Hockey League 112 122
Alexander Janzen Kontinental Hockey League 111 129
Sergei Tereshchenko Kontinental Hockey League 111 126
Pavel Poluektov Kontinental Hockey League 111 121
Vitaly Yefimov Kontinental Hockey League 110 158
Pavel Belousov AHL 110 138
Andrei Piskunov Kontinental Hockey League 110 130
Alexei Statsenko Vysshaya Liga 110 128
Igor Smolin Vysshaya Liga 110 121
Nikita Suchkov Kontinental Hockey League 110 127
Rizvan Mansurov NHL 108 139
Dmitry Lugin Kontinental Hockey League 108 132
Valery Korinevsky Vysshaya Liga 108 134
Artur Kayumov Vysshaya Liga 108 127
Ilya Zinoviev Vysshaya Liga 108 125
Dmitri Zhukenov Kontinental Hockey League 107 135
Ruzal Polyanski Kontinental Hockey League 107 134
Sergei Shmelyov Kontinental Hockey League 107 131
Artyom Zhelezkov Kontinental Hockey League 107 129
Mikhail Sergachev Vysshaya Liga 107 137
Anton Slepyshev Kontinental Hockey League 107 127
Ivan Yemets Kontinental Hockey League 107 125
Andrey Makarov Kontinental Hockey League 106 136
Alexander Khokhlachev Kontinental Hockey League 106 133
Artyom Blazhiyevsky Kontinental Hockey League 106 131
Ivan Fishchenko Kontinental Hockey League 106 131
Nikolai Shefer NHL 106 128
Nikita Nesterov Kontinental Hockey League 106 129
Evgeny Kobyakov Kontinental Hockey League 106 128
Vadim Kudako Kontinental Hockey League 106 128
Nikita Belov MHL2 106 149
Vyacheslav Leshchenko Kontinental Hockey League 106 127
Dmitri Shulenin Vysshaya Liga 106 126
Marsel Ibragimov Kontinental Hockey League 106 121
Denis Cherepukha Kontinental Hockey League 106 120
Nikita Kvartalnov Kontinental Hockey League 106 120
Yegor Nazarov Kontinental Hockey League 106 120
Dmitry Orlov Kontinental Hockey League 105 138
Anton Burdasov Kontinental Hockey League 105 131
Klim Kostin Kontinental Hockey League 105 125
Dmitry Ermakov Vysshaya Liga 105 126
Stanislav Galiev Vysshaya Liga 105 127
Alexei Nazarkin AHL 105 120
Andrey Pedan Kontinental Hockey League 104 133
Rafael Bikmullin Kontinental Hockey League 104 130
Artyom Sergeyev Kontinental Hockey League 104 128
Vasili Khabarov Kontinental Hockey League 104 127
Andrei Bannikov Kontinental Hockey League 104 125
Sergei Tolchinsky Kontinental Hockey League 104 125
Vyacheslav Litvinov Vysshaya Liga 104 121
Marcel Zykov Vysshaya Liga 104 131
Igor Boldin Vysshaya Liga 104 133
Artyom Prokhorov Vysshaya Liga 104 127
Yury Avramenko Kontinental Hockey League 102 126
Maxim Bain Kontinental Hockey League 102 121
Yevgeni Yaroslavlev Kontinental Hockey League 101 129
Mark Topol NHL 101 126
Maxim Alyapkin Kontinental Hockey League 101 120
Pavel Ivanov Kontinental Hockey League 101 118
Alexander Tarasov Kontinental Hockey League 100 138
Yury Sokolov Kontinental Hockey League 100 134
Andrei Zimin Kontinental Hockey League 100 132
Pavel Kozyrev Kontinental Hockey League 100 125
Artemi Yannurov AHL 100 122
Alexander Plotnikov Kontinental Hockey League 100 121
Yury Alexandrov Kontinental Hockey League 100 121
Pyotr Boblenkov MHL 100 131
Gleb Bondaruk Pervaya Liga 100 120
Denis Kostin Vysshaya Liga 100 112
Danil Vasilyev Russia Junior 100 126
Tigran Vinogradov AHL 100 118
Bogdan Fyodorov Kontinental Hockey League 100 116
Grigori Zharkov Kontinental Hockey League 100 115
Daniil Ilyin Kontinental Hockey League 100 114
Mikhail Meshcheryakov Kontinental Hockey League 100 113
Alexander Sharov Kontinental Hockey League 100 111
Kirill Shchukin Kontinental Hockey League 100 111
Dmitry Denisenko Kontinental Hockey League 100 111
Vladislav Gromov Kontinental Hockey League 100 108
Vsevold Khapko Kontinental Hockey League 100 106



Out of 728 players in the KHL in 2025 (including those listed above)
  • 72 have a CA of 18-50 (9.9%)
  • 207 have a CA of 51-80 (28.4%)
  • 228 have a CA of 81-100 (31.3%)
  • 221 have a CA of 101-165 (30.4%)



The data from KHL data from TBL 8.2 shows only 531 players "Playing For" the KHL (28 teams x 23 players = 644).
To make the TBL 8.2 data more comparable with the 2025 Test 1 data I used "Contracted To" the KHL and got 857 players

Of the 857 players (CA average 86.4)
  • 109 have a CA of 12-50 (12.7%)
  • 271 have a CA of 51-80 (31.6%)
  • 153 have a CA of 81-100 (17.9%)
  • 322 have a CA of 101-168 (37.6%)
User avatar
Asher413
Challenge Moderator
Posts: 355
Joined: Wed May 20, 2009 2:28 am
Favourite Team: Pittsburgh Penguins

Re: EHM1 Regen Testing (2016)

Post by Asher413 »

Thanks for the clarification on the "Newgens", I was also still under the impression that they were brand new players based on the current players, not a varation of regen.
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Asher413 wrote:Thanks for the clarification on the "Newgens", I was also still under the impression that they were brand new players based on the current players, not a varation of regen.
Hmmmm....to me a regen IS a brand new player based on the current/old player
The difference now I think is they're not "recognizable" anymore - when the new EHM came out some people said they'd found the old style "copies" (like Chara) but no screenshots were ever posted

If you have anything at all "based on" the previous player it's a regen I think (even if it was just PA and Position, those two and some combination of some of the remaining 50+ Attributes would be a regen IMO).

The multiple tests I've run out to 2035 and 2045 don't show the fluctuations in the database that you'd expect with a truly newgens system, and I'm not aware of any editor that shows anything like the kinds of settings you would need, and I've seen/heard no posts or comments (private or public) anywhere that indicates the talent pool is "dynamic" because the game reassesses database needs yearly and creates players accordingly

As far as I can tell, retro databases don't create less talented regens because the required/quality players are already in the database for each draft year (meaning the game doesn't seem to notice the talent pool already in the database and the game doesn't say I'll lower the PAs of these recreated retired NHL players to "below NHL standard" because there's already decades of real players in the database - like the 1998 database wouldn't need the game to create a potential NHLer for the first 18 years of play & the 1974 database wouldn't need the game to create a potential NHLer for the first 42 years!)

You don't see countries have a fluctuating talent pool as far as I can tell (which you'd think you would sometimes if they were truly newgens)

I think all that's PA/Position is likely all that's "needed" to keep the talent pool/player position distribution pretty close to what was started with (so the game remains playable years into the future)



Further to this regen/newgens discussion
- here's the new EHM Regen Thread from TBL http://www.ehmtheblueline.com/forums/vi ... =3&t=15843
- here's a 2008 thread regarding FM regens/newgens http://community.sigames.com/showthread ... -and-regen
- here's a 2011 French article on FM newgens (I used google translate) http://www.fmeurope.net/forum/viewtopic.php?t=4592

What I found most interesting is the French article.....the 2008 FM article seems to be talking more about the kind of hybrid regen/newgen system we apparently have now in the new EHM) while the 2011 French article gives a much better description IMO of what a "newgen" system would require...for example, it says "So with FM2011 we saw appear any new newGens profile was modelled according to the nation: the English are often physically strong but limited technique then the Spaniards will be smaller but much more technical. It means depending on the different types of real players for each position, grounds are identified and are the basis for the creation of models of newgens. The original proportions of the positions are also kept in the creation of the newgens, there will be more DC that libero for example, and for a same position certain models are more frequent than others, for example less goalkeeper libero to standard goalkeeper.
Certain physical characteristics such as size relate to the jump for example, attribute, which is the maximum height that can reach a player. In other words, a player who is great, will have a relatively large jump attribute, which influences the choice of the model of newgens as a result and are taken into account when choosing the model of newgens to apply.
It is also the geographical variations, technically, once positions "sorted", a secondary sort for countries with active leagues there. For example, there will be a Brazilian model "BT-pivot – Russian" and "BT-pivot". For inactive leagues, there are models more global which are not national but continental, if the only League active is france, there will be an "African BT-pivot" and "BT South American hub" model.

This novelty was also hiding the appearance of regions. Thus, as in reality, each region has specific nature (mainly of languages or dialects but also names) which had repercussions on the names in order to have more varieties and less of inconsistencies."

I think (I don't know 100% for sure) that we have the two things I highlighted in blue, but not the two highlighted in red
The second highlighted in red seems like the Player Roles that have been added - it seems to be described above as if the game uses the Attributes to determine the "model" of player (I'm assuming model of player/Player Role are similar in meaning).....I think in EHM the game chooses the Player Role first, and then fits the Attributes accordingly (Essential/Non-Essential/etc)


The idea of "modeled according to the Nation" and "geographical variations" sounds so awesome!
Especially with retro databases/older hockey, where such differences were WAY more pronounced


I think football/soccer would be very similar/the same way as hockey if for decades virtually every top player worldwide only played in one League (like the hockey and NHL)......but because top players play in multiple Leagues in different countries it's been much less of a homogenizing effect

As I recently told Alessandro, "Nowadays the hockey is almost all the same (the rules/ice surface make a bigger difference than the players, players are almost interchangeable with little difference anymore from country to country).....Ovechkin (the best current Russian to many) IMO plays a North American game (like Bobby Hull, Mike Bossy, etc), and IMO he's nowhere near as good as talented as players like Maltsev, Kharlamov, Makarov (or even as good as the "guys with size" like Mikhailov and Yakushev) - if you've never watched the Russian games from the 1970s and 1980s you should (that's when there actually was real "Russian hockey" IMO)"


So are they regens or newgens?
I'd think a "much improved newgen system" or a "hybrid newgen/regen system" would be the most accurate response...it seems to me very likely that PA (maybe slight variations? maybe not) and Position are kept for sure.....I'm not sure about Player Role or any other Attributes. Ultimately, at this point if "regen hunting" no longer works than I don't think I could tell the difference myself between a regen system & a regen/newgen hybrid system & a truly newgen system!
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

EHM1 Regen Testing (2016)

Post by nino33 »

Here's the Pro League data from Test 2 (all possible players/staff added) at 2025
.

CA/PA averages at 2025 by League (as of August 29, 2025)
The left hand side is all players Contracted To (so includes signed minor pros/juniors and is much higher than the number that "play" in the League)
The right hand side is the number of teams x 23 players, so it is reasonably representative of those actually playing in the League)
LEAGUE Players CA average CA range PA average PA range ---------- Players (23 per team) CA average CA range PA average PA range
NHL 844 136.5 52-193 152.4 81-199 ---------- 690 144.8 118-193 158.2 120-199
KHL 680 84.6 8-176 113.1 10-178 ---------- 644 87.5 43-176 116.0 60-178
AHL 554 104.2 51-151 129.8 80-180 ---------- 527 106.1 74-151 131.4 100-180
SHL 374 93.5 27-146 120.2 55-178 ---------- 322 100.9 64-146 126.6 84-178
HockeyAllsvenskan 430 77.8 18-126 105.5 29-168 ---------- 322 88.6 64-126 115.8 80-168
Liiga 406 92.5 22-159 118.1 45-166 ---------- 345 100.3 67-159 125.1 81-166
NLA 327 71.7 8-154 100.0 27-172 ---------- 276 78.8 43-154 106.3 60-172
Czech Extraliga 410 74.7 1-141 103.2 16-166 ---------- 322 84.9 50-141 113.4 63-166
DEL 386 74.6 4-166 101.8 12-175 ---------- 322 84.8 40-166 112.9 60-175
NOTE - fewer teams are shown in these Test 2 results as I'm tired but want to get what I have done posted & also the "add all" option is far less important to me and is really for reference only (I'd never play that way myself)




Here's the Test 1 data for reference...
nino33 wrote:CA/PA averages at 2025 by League (as of August 29, 2025)
The left hand side is all players Contracted To (so includes signed minor pros/juniors and is much higher than the number that "play" in the League)
The right hand side is the number of teams x 23 players, so it is reasonably representative of those actually playing in the League)
LEAGUE Total PlayersCA averageCA rangePA average PA range--Players (23 per team) CA average CA range PA average PA range
NHL 825 135.5 50-195 147.1 63-195 -- 690 143.6 112-195 152.8 120-195
KHL 728 87.9 18-165 110.4 35-175 -- 644 93.9 55-165 115.4 61-175
AHL 527 98.7 47-160 116.7 61-178 -- 527 98.7 47-160 116.7 61-178
SHL 405 94.9 15-156 114.2 32-158 -- 322 106.1 74-156 123.6 85-158
HockeyAllsvenskan 430 77.6 3-130 97.7 17-152 -- 322 88.8 55-130 108.5 60-152
Liiga 440 86.7 23-157 108.4 41-170 -- 345 98.1 60-157 119.0 80-170
NLA 330 72.3 4-150 94.0 7-156 -- 276 79.5 47-150 100.5 60-156
Czech Extraliga 408 74.3 4-130 96.2 13-157 -- 322 85.4 47-140 107.2 60-157
DEL 376 79.0 3-163 100.6 11-163 -- 322 88.2 42-163 110.2 60-163
Erste Bank Eishockey Liga 366 69.0 1-140 93.0 10-148 -- 276 82.8 50-140 107.6 62-148
Slovakian Liga 281 71.0 9-120 93.8 30-146 -- 207 82.2 55-120 104.7 65-146
GET Ligaen 246 68.9 4-128 93.5 10-156 -- 230 72.2 30-128 97.1 45-156
Metal Ligaen 284 57.4 3-125 80.8 10-144 -- 230 67.2 30-125 92.0 42-144
EIHL 317 53.2 1-128 71.5 10-158 -- 230 70.0 19-128 91.3 20-158
EPL 286 38.9 1-110 57.3 10-130 -- 230 46.4 17-110 66.5 20-130
ECHL 437 81.1 35-128 101.1 44-137 -- 437 81.1 35-128 101.1 44-137
LNAH 190 67.0 33-124 92.4 43-139 -- 172 69.7 49-124 94.5 64-139

NOTE - PA range for "Players (23 per team)" is the PA average/range of those players listed in the CA range
NOTE - AHL/ECHL/LNAH data is the same on "both sides" as they don't have an average of 23 players per team
Last edited by nino33 on Sun Apr 21, 2019 4:36 pm, edited 2 times in total.
User avatar
Alessandro
Olympic Gold
Posts: 2865
Joined: Sat Aug 27, 2005 1:54 pm
Custom Rank: TBL Rosters Man
Favourite Team: Team Russia
WHL Team: Calgary Flames

Re: EHM1 Regen Testing (2016)

Post by Alessandro »

So Nino you see there are many things that make zero sense. Sergachev playing in the VHL lol. Kaskov in SHL and Kara in the Liiga makes zero sense as well...
Your tests are great, but somehow also the quality tests should be considered, not only the quantity ones.
Keep it coming. As you said, I think the 2035 one should be more important and I can't wait for the results of that.
User avatar
Primis
Freeware Moderator
Posts: 1698
Joined: Tue Aug 03, 2010 7:46 pm
Location: Michigan, USA

Re: EHM1 Regen Testing (2016)

Post by Primis »

My thoughts (mainly on that first test at the bottom, because I must have missed that the first time?):

Why on earth is anyone with a CA of 124 in the LNAH? Or a CA of 128 in the ECHL for that matter? That to me is messed-up.

I'm assuming the low bottom-end for say the KHL is due to more to other things like possibly lack of research (CA's of 8 and 13?)... but I can't really see the upper-end CA guys in those leagues (ECHL and LNAH) unless they're really, really old and quickly on the decline or something else very weird.

A lot of the euro leagues it could make more sense (formerly elite NHL player returning home to end career, etc), but we obviously don't normally do that sort of thing in the minor North American leagues. So this means while some NHL team has a 50 CA guy on roster, and an AHL team has a 47 CA guy, there's a 128 CA guy sitting in the ECHL doing... what?

Important question here now: do the players exported with the Assisstant include the fake, grayed-out players the game itself creates for one game as fill-ins? Could those be affecting the averages on the low end?
User avatar
nino33
Mr. Goalie
Posts: 6088
Joined: Sat Aug 07, 2010 3:37 am
Custom Rank: Retro Rosters Specialist
Favourite Team: 1970s hockey

Re: EHM1 Regen Testing (2016)

Post by nino33 »

Alessandro wrote:Your tests are great, but somehow also the quality tests should be considered, not only the quantity ones.
I don't understand....what's a "quality test" and what's a "quantity test" and how are they different/what is considered in each?
Post Reply