EHM The Blue Line

Shindigs wrote: I don't entirely know how it works in hockey. But going by FM your players will actually have holiday in the off-season. There is such a thing as over-training. If you never ever give your players time off there will be both physical and mental backlash due to being constantly overworked. I'd argue that keeping your players on any schedule with team scheduled training every single day for off-season (which is the norm in EHM) would probably not be a good thing. I only played up to junior level, but we only had like 3 training days a week in the summers.

In hockey players train on voluntary basis in the summer, which I suppose means the gym, jogging, perhaps they get access to an ice hall, that kind of stuff, but it's still like "half a holiday", not full throttle. I guess some players may have a short holiday too, where they train even less, but they do need to keep in shape, of course. Also playing so many exhibition games would tire the players out in real hockey. Many European players see the pace of games in North America very tiring. In the NHL you play 82 regular season games, plus the playoffs, if your team makes it. To add exhibition games all summer long to that, would be suicide. I was under the impression that is somehow implemented into EHM, in a way that over using your stars in the regular season might make them perform at a lower level in the playoffs. But that is just an impression, no testing done, as far as I know. And maybe exhibition games were overlooked? I've not heard anyone else using this trick, and did not come to think of it myself, so it might be exhibition games don't count like regular season games. And now that I'm aware of it, still wont' use it.

When I tried playing in the lowest tier of british hockey (word from the wise, don't) the fact that their season is full length but only has like 20-30 games in it means that your players never actually get match fit. So I figured I'd add in exhibition games in the longer gaps between games. The end result was my players playing well in the exhibition games, and losing all the "real" games. It's like your players lose focus on what is important if you play them in too many pointless games. They don't know when to give it all, and when not to. So from my very limited testing it seems that adding exhib games mid season, at least, has a very detrimental effect on your team. Honestly just about everything about off-season is broken in this game though. Like players being immune to injuries for starters, so I see no reason why exhib spam in the summers wouldn't be too. Just like how in late august you get some kind of buff put on your players that make them go to 100 con every single day no matter what they had the previous day. Means you can play a pre-season game every single day from about mid august to early september if you feel like it. And since goalies take about 30 games to get fully match sharp at worst, you might need to. My pre-season is normally 15 games to a 52 game season, which seems silly much. But if you bought a goalie who used to be backup in a higher league club, you will need to give him more games than make any sense whatsoever to actually have him fit for the season.

Also on the topic of over-using stars making them worse in the playoffs. I'm not necessarily seeing the same thing. What I am seeing is that the last game of each season, your players will lose about double the normal condition (to make your stars have no con for the first playoff game #artificialdifficulty) and you will generally have 2-3 players get injured right as the playoff starts, normally your 3 highest scorers, maybe your first goalie too. Your goalie(s) will also let in 1 GAA more than in the normal season regardless of consisitency, determination and their "big games" hidden attribute. Cause as we all know the AI is bad at setting up the opposing teams, so it just blatantly and obviously cheats to make the playoffs "hard" instead. At least that's how every single SHL relegation playout has worked over about 10 seasons in 3 different saves for me. The pattern is pretty hard to miss. I understand making good AI is hard, so you kinda have to make it cheat to maintain a challenge. But I'd like them to have the tact to make it cheat in slightly less obvious ways, cause when you lose to stuff like that there isn't really much you can do as a manager to fix it, which is infuriating /end rant

I on the other hand have not seen that. I get some injuries, but I wouldn't say it's always my top scorers. Maybe one here, sometimes, but not in an obvious way. Maybe you're over using them? And my goalies often perform better in the playoffs. They may be pretty inconsistent in the regular season, and then light up in the playoffs, to the point that I decide to offer an extension. Perhaps you've not found the right goalie with high "Important Games" attribute?

This is specific to the HockeyAllsvenskan -> SHL promotion playoffs. Never seen it anywhere else either. But there it happens like clockwork every year. Some years you get lucky and only get 1 injury. But the goalies always just lets in goals left right and center, it's always that dumbass onetimer off the faceoff that is almost exclusively scored by the AI too. It's pretty much just a really cheap way of "increasing difficulty", but alas we're getting off topic. I could talk for days on how "good" the actual match engine is, there is a reason I focus on the player development side of the game.

Well, if it's specifically for the SHL promotion playoffs, I doubt it's an intentional thing. More like a bug -- assuming you are correct. If it were a real means of increasing difficulty, they wouldn't have aimed it at only one specific event in the world of hockey.

Shindigs wrote:The reason I started posting my findings here is so we can start having multiple players with different styles add in their 2 cents until we find out as much as we can about how training works......But if no one shares their findings we never will truly understand it.....

FYI - I've done a lot of related testing, some examples/links below...

EHM:EA Attribute testing (from July 2015) http://www.ehmtheblueline.com/forums/vi ... 10&t=15852
EHM:EA Attribute testing - Staff http://www.ehmtheblueline.com/forums/vi ... 10&t=14812
EHM:EA Attribute Testing - Major Junior http://www.ehmtheblueline.com/forums/vi ... 10&t=14811
EHM07 Height, weight and CA monitoring (12 years of data) http://www.ehmtheblueline.com/forums/vi ... 110&t=8146

Tasku wrote:Well, if it's specifically for the SHL promotion playoffs, I doubt it's an intentional thing. More like a bug -- assuming you are correct. If it were a real means of increasing difficulty, they wouldn't have aimed it at only one specific event in the world of hockey.

I assume it would be the same for all leagues that have the same means of being promoted/demoted from one league to another. I honestly don't know which other ones there are in EHM with the same system. And none of them pique my interest enough to play them. And I don't want to play multiple seasons in some league I don't care about just to find out. It comes across as pretty darn rigged though, even the teams who randomly have 2-3x the budget of the other HockeyAllsvenskan teams have a real problem trying to get up into SHL with half their roster being players like Jonathan Bernier, Alex Tanguay, Brenden Morrow, etc. Which is absolutely ridiculous.

nino33 wrote:

Shindigs wrote:The reason I started posting my findings here is so we can start having multiple players with different styles add in their 2 cents until we find out as much as we can about how training works......But if no one shares their findings we never will truly understand it.....

FYI - I've done a lot of related testing, some examples/links below...

EHM:EA Attribute testing (from July 2015) http://www.ehmtheblueline.com/forums/vi ... 10&t=15852
EHM:EA Attribute testing - Staff http://www.ehmtheblueline.com/forums/vi ... 10&t=14812
EHM:EA Attribute Testing - Major Junior http://www.ehmtheblueline.com/forums/vi ... 10&t=14811
EHM07 Height, weight and CA monitoring (12 years of data) http://www.ehmtheblueline.com/forums/vi ... 110&t=8146

I appreciate that you put a lot of work into that. But since it was done well before the game left early access (if I'm not mistaken) we don't really know for certain how much of it still applies. Or do we? I am just for trying to get as much up to date info as possible in one place. I started doing all the stat tracking for my own benefit only, but I figure I might as well share. And it seems to have gotten the discussion going again. Which is a nice side effect. For example the gain in skating seems very low compared to what I've been seeing. Unless your slowest players had 14 Spe when you started I can't see your max being +6. I've had players get that in 1-3 years. Not 13 years. To be fair I tend to gravitate towards getting very fast players (even with hidden attributes on when I scout) so that might heavily skew my numbers. To me the 20 cap on Spe and to a lesser degree Acc is an actual issue. Cause when players hit it at like 19, they randomly stop improving for the rest of their career due to an arbitrary maximum, which is just silly.

Any idea what might have caused that really weird outlier in average stickhandling attribute for Wingers in 2011 in your Major Junior test? It really sticks out like a sore thumb. A general observation otherwise seems to be that regens have suspiciously low stickhandling, even the finesse ones. On the flipside it does tend to grow a fair bit. But it's generally way lower than their deking, and since both are off. skill and fairly closely tied it would make sense for them to be closer. Trying to find a 10+ deking 16 y/o is not too hard. Finding a 10+ Stickhandling is something I myself have done only once, and that was with a 14 y/o offensive defenseman phenom with 14 passing and 11 stickhandling and 264 (if memory seves me right) total attributes at 14, which is just insane.

As far as I know there's promotion / relegation only in Finland and Sweden, and only in Sweden you can go all the way to the top (unless the rules haven't been updated). That's why it seems a bit strange they would add an intentional mechanic only for these two countries.

Shindigs wrote: Any idea what might have caused that really weird outlier in average stickhandling attribute for Wingers in 2011 in your Major Junior test? It really sticks out like a sore thumb. A general observation otherwise seems to be that regens have suspiciously low stickhandling, even the finesse ones. On the flipside it does tend to grow a fair bit. But it's generally way lower than their deking, and since both are off. skill and fairly closely tied it would make sense for them to be closer. Trying to find a 10+ deking 16 y/o is not too hard. Finding a 10+ Stickhandling is something I myself have done only once, and that was with a 14 y/o offensive defenseman phenom with 14 passing and 11 stickhandling and 264 (if memory seves me right) total attributes at 14, which is just insane.

I think the same. I think that is probably the effect of stickhandling being under-represented in the "important" ratings for the new player roles system, but it's just my opinion and I didn't have any testing for it.

Tasku wrote:As far as I know there's promotion / relegation only in Finland and Sweden, and only in Sweden you can go all the way to the top (unless the rules haven't been updated). That's why it seems a bit strange they would add an intentional mechanic only for these two countries.

Yeah, it just acts in a very weird manner. In all my seasons they have all played out in the exact same way, regardless of if I finished 1st or 2nd. I play the other team (all but one time It's been AIK) in the "champion series" which is 1st place vs 2nd place. The winner of this best of 3 series is the actual HockeyAllsvenskan champion. I've lost this every single time, regardless of how much I've demolished AIK in the season, my form or any other things I can think about. After this I will win the Best of 3 series against the winner of the playoffs (3rd-8th place all play each other once, highest points gets to face the loser of the champion-series) I've won it 2-1 every single time. In this game the match engine plays like it normally does. In the championship game it does not, the amount of faceoff goals for the opposition goes through the roof every single year. It's quite demoralizing since there is literally nothing you can do to stop those goals. Then the winner of the championship serier, and the winner of the other match will face the bottom and 2nd from bottom SHL teams. over 10 seasons (20 series) I've seen Leksand win it once to go up (22Mkr total salary, that is more than a fair few SHL sides) Södertälje win it once (They had 18Mkr salary in the third tier of Sweden, which is 6 times as much as the 2nd biggest) every other time the HockeyAllsvenskan side has lost 4-1 in the series. Like at this point I don't even care about the playoffs cause what you do as the manager seems to have little to no effect on the outcome. You just have to pray the opposing side doesn't get 2-3 free faceoff goals per game and that what remains of your top 6 can be in hot form to outscore what faceoffs goals do come your way. But I mean having watched as many games in full in match engine as I have I've sort of come to terms with the fact that the AI almost always score canned goals that always look the same and where all your players are doing their best to either aid them in scoring or just staying out of the way. It sucks, and it's why I don't watch games in EHM anymore cause it's like watching iBP play CS:GO

Oh and while I'm on the match engine. The player AI doesn't recognize auto-icing. It literally breaks the game some matches, your players have more icings than passes and constantly dumps the puck from 1 inch before the red line while under no pressure. And if you put the dumping setting on "very rarely" they actually do it more often, which just blows my mind.

The upside to it being very very unlikely to go up from HockeyAllsvenskan regardless of how good your team is, is that the league is the perfect skill level for developing 16-20 y/o's and by the time they become good enough that you'd play them in an SHL side, <insert NHL-team> will just swoop in and take them from you anyways. So I just sit pretty in Swe-2 and farm talents, which is what I like doing anyways. So there's that.

Edit: In an attempt to actually make this post slightly on topic, does anyone have good insight into what it takes to make a goalie grow well? I've had one who just went insane, but that was just cause I put him in my J-18 squad where he gained like 50 attribute points in 1½ season and got drafted by the sharks following season. In training they almost all seem to just either randomly lose a bit here and there, gain a little once in a while or simply be stagnant for fairly long periods of time. And since you only really have 1-2 "real" goalies at a time getting good stats on it is proving a bit of a pain. So far the schedule I've seen a big (over 10 in a 4 month window) increase with is this one:

Con: Int
Ska: Med
Tac: Lig
Off: Lig
Def: Med
Goa: Int

But on the flipside I've seen next to no gain from it most of the time, or at least very slow gain. To be fair goalies get into their prime very late compared to skaters, and they have less attributes so they get smaller attribute totals than skaters. Both those together means they should have fairly slow gains, but it makes it so hard to tell if my schedule is working or not. Add in that goalie scouting is a mess, at least in Swe-2. Your scouts will call any goalie who isn't abysmal (and some who are) 5 star. So 5 star means nothing. Their reports will often lie "Has a quick glove hand" -> actually has like 7 glove skill. So half the time you don't even know if your goalie is supposed to get good or not. The same goes for their comparisons. "Like Erik Hanses" <- He's had 92% or higher SV% every season so far. But they will say "Like Erik Hanses" for players with anything from 2-5 star potential, which is just a blatant lie. And this is coming from 10 scouts with generally solid mentals and judging ability/potential in the 12-19 range. Most being in the 15-17 range with some outliers. So I'd love it if someone could enlighten me on the goalie front, it's the only place where I just buy older players with a good history in similar skill or higher leagues, rather than try and farm my own young goalies (I tried twice, SJ Sharks stole both of them within a year...)

Edit: Updated my TL;DR post a bit. Ran a full summer of nothing but Tech2, it was absolutely awful. None of my young players grew at all, but the older players didn't decline as much as they have previously. Next summer I'll be doing Phys on my special snowflakes who randomly grow their technicals on Phys, Tech on everyone else who isn't old and Tech2 for maintaining the veterans. Will see if that works out as well as it reasonably should.

Shindigs wrote:I appreciate that you put a lot of work into that. But since it was done well before the game left early access (if I'm not mistaken) we don't really know for certain how much of it still applies. Or do we?

I think I have a pretty good idea myself, yes

Shindigs wrote:I started doing all the stat tracking for my own benefit only, but I figure I might as well share. And it seems to have gotten the discussion going again. Which is a nice side effect.

I really appreciate your sharing of testing data/results!
FYI - the public discussion may have started up again, the discussion on the Researchers Forum & amongst the primary researchers has been ongoing

Shindigs wrote:For example the gain in skating seems very low compared to what I've been seeing. Unless your slowest players had 14 Spe when you started I can't see your max being +6. I've had players get that in 1-3 years. Not 13 years. To be fair I tend to gravitate towards getting very fast players (even with hidden attributes on when I scout) so that might heavily skew my numbers.

Yeah.....you're using such a small sample size, and you're looking at "NHL players" while I looked at much larger numbers (often whole leagues/the entire database) + you're not repeating the testing.....I almost always do a minimum of 3 tests (to ensure findings are consistent/reproducible)

Shindigs wrote:Cause when players hit it at like 19, they randomly stop improving for the rest of their career due to an arbitrary maximum, which is just silly.....since both are off. skill and fairly closely tied it would make sense for them to be closer

These statements are not necessarily completely correct (are you aware of the Player Role system? there are different levels of Attribute importance/relevance depending on Player Role + there are guidelines based on CA).
Just because the explanation on something isn't publically available doesn't mean it's "arbitrary" (the vast majority of players not only don't need to know such details, my experience has been they don't care.....and as I think you know posts "explaining things" can easily get very lengthy! HaHa)

Shindigs wrote:Any idea what might have caused that really weird outlier in average stickhandling attribute for Wingers in 2011 in your Major Junior test? It really sticks out like a sore thumb.

My first thought is it's maybe/probably a typo

Shindigs wrote:and 264 (if memory seves me right) total attributes at 14, which is just insane.

Don't forget the hidden Attributes can significantly affect a player's development/performance!
Attributes like Injury Proneness, Natural Fitness, Consistency, Decisions, Important Matches, Pass Tendency, Ambition, Pressure and more have a huge impact on the player's performance.....in particular Consistency has a HUGE impact on what a player does with the ingame visible talent they have (but so do other hidden Attributes!)


A final thought.....I see you referencing the 2D game and making comments as if it's an exact representation of what's happening & since I've known of EHM I've never seen anyone say that (including Riz/SI) but I have seen it said many, many times here at TBL that it's not (but not by Riz/SI) - the point usually made is you'll see your D give the puck away onscreen (in 2D) but the player doesn't receive a "giveaway" statistically (so the idea is the simulation is correct, the 2D view is a less than perfect representation of the simulation...the 2D is not the actual exact simulation playing out before your eyes, but rather a representation of the simulation).

My first thought is it's maybe/probably a typo

Good old Occam's Razor then, haha:)

These statements are not necessarily completely correct (are you aware of the Player Role system? there are different levels of Attribute importance/relevance depending on Player Role + there are guidelines based on CA).
Just because the explanation on something isn't publically available doesn't mean it's "arbitrary" (the vast majority of players not only don't need to know such details, my experience has been they don't care.....and as I think you know posts "explaining things" can easily get very lengthy! HaHa)

I'd love to know more about it. Does this mean that an attributes shown as 20 can actually still improve, just not in a visible way to the player? Or are you more referring to the fact that some players (generally finesse forwards) will get a unreal amount of acc/spe gain, whereas a stay at home def has a equally sick gain to Bal/Sta/Str, but they very rarely have the ability to gain a lot of the opposite (bal/str for finesse and acc/spe for S@H def). I've also seen several players change roles depending on how I've trained them. Flashy Def forwards-> Power Forwards. I imagine it working as a few different very specific roles with certain requisites to "unlock" them. For example a Flashy Fwd might "unlock" Flashy Def Fwd and then later Power Fwd. Based on how good and balanced his attributes get. But that's just how I imagine it working based on what I've seen in my saves. I've also seen Power Fwd's regress to Flashy Def Fwd. A post somewhere on these forums had made me think there was no way for the projected role to change like that until I noticed it happening quite a lot in my saves.

FYI - the public discussion may have started up again, the discussion on the Researchers Forum & amongst the primary researchers has been ongoing

And as a part of the huddled masses it's nice to get to see more of it. Knowledge is power and all that good jazz!

Yeah.....you're using such a small sample size, and you're looking at "NHL players" while I looked at much larger numbers (often whole leagues/the entire database) + you're not repeating the testing.....I almost always do a minimum of 3 tests (to ensure findings are consistent/reproducible)

I'm actually looking at Swe-2 players almost exclusively. But I do agree that we have two polar opposite approaches. I am very much looking into the micro management aspect whereas you focus on the macro side of things. Your analysis on which attributes grow/don't grow and the fact that you use some save editor or whatnot to actually see the hidden attributes gives a much deeper insight in the nuts and bolts of the attributes themselves. My interest lies more in the "how do I get player X to grow to his max potential in a reasonably small amount of time". While we're on the subject of nuts and bolts; how early is it possible for a defenseman to have his peak? I've got this one player, who isn't even a newgen named "Jacob Moverare" who has started declining at 23. Which just seems really early to me.

Don't forget the hidden Attributes can significantly affect a player's development/performance!
Attributes like Injury Proneness, Natural Fitness, Consistency, Decisions, Important Matches, Pass Tendency, Ambition, Pressure and more have a huge impact on the player's performance.....in particular Consistency has a HUGE impact on what a player does with the ingame visible talent they have (but so do other hidden Attributes!)

I am well aware, I have no way to actually watch them myself. But I've had players with 225 total attributes pull 50 point seasons, and players with over 300 struggle to ever get over 20. But I try and read as much of that as I can from the scout reports and past history (60+ points in ECHL/AHL for several seasons on the bounce? That points to high consistency and good hidden attributes), especially now that I don't watch player attributes at all until I have to record them for training stats. I have certain things I'm looking for, which tends to lead to me getting players who look very much alike. Jarkko Nyberg from my big post is like the poster-child of the player type I end up getting more often than not. A funny thing about injury proneness though, If you really really make sure to never drop your players below 80% con if you can help it. There seems to be next to no injuries ever, to the point that I actually ignore that part of the scout report because at most it will result in a single 10 day injury a year in my experience, but this is a field where your type of macro testing would be much better at giving a big picture of how that all works.

A final thought.....I see you referencing the 2D game and making comments as if it's an exact representation of what's happening & since I've known of EHM I've never seen anyone say that (including Riz/SI) but I have seen it said many, many times here at TBL that it's not (but not by Riz/SI) - the point usually made is you'll see your D give the puck away onscreen (in 2D) but the player doesn't receive a "giveaway" statistically (so the idea is the simulation is correct, the 2D view is a less than perfect representation of the simulation...the 2D is not the actual exact simulation playing out before your eyes, but rather a representation of the simulation).

I'm aware. Doesn't make the 2D engine any better though. The part about players icing the puck like their lives depend on it seems indicative of something being very wrong with the simulation though. It's also very odd that the goals I as the player score have a fairly great diversity. Whereas the goals I concede are next to always one of 4 or so. Even if there is no shenanigans at play it just makes it look like it. But again, the post game stats are so lacking (no time in off zone, no time in neut. zone, no time in def zone, no time with puck, etc.) that you kind of have to watch it to get an idea of how your system is/isn't working. And when the only representation you can see isn't actually showing you what is happening it makes it a bit of a headache to try and fix whatever is wrong with the setup. But now I just have my coach deal with that, I just check the post game ratings and base my roster decisions around that. The nice sideeffect being a complete detachment from players who "feel" like they are doing great things, but don't make it show on paper.

And on the note of it getting a bit verbose when you try and explain things as deep as most game mechanics...I can relate. I should note for those who somehow haven't noticed yet that I am a bit prone to hyperbole, so when I start flinging superlatives just see it for what it is, me being overly emotive with my use of the english language.

P.S. I just have to mention the most hilarious simulation/2D engine bug I have ever had. The opposing team tried to swap goalie, but the new goalie somehow ended up taking the Center's spot in the UI and the old goalie was still in the crease. So the swapped in goalie kept making figure-8's around the net trying to push his way into the crease, but the other goalie was just standing there blocking it. This continued for about 30 minutes on max replay speed before I gave up and did ctrl+alt+del. Even turning off the highlights wouldn't make the game progress beyond it:S
Here it is in its full glory: https://youtu.be/9vYI7hd5Om8

Shindigs wrote:Does this mean that an attributes shown as 20 can actually still improve, just not in a visible way to the player?

AFAIK there is nothing like an Attribute value "above 20" BUT.....if two players each have a 20 in the same single Attribute but one has a significantly higher CA (overall ability) I don't think these players are "the same" even though they both have a 20 in the Attribute; this might be because they are "the same" but in the "related Attributes" there are significant differences

Shindigs wrote:Does this mean that an attributes shown as 20 can actually still improve, just not in a visible way to the player? Or are you more referring to the fact that some players (generally finesse forwards) will get a unreal amount of acc/spe gain, whereas a stay at home def has a equally sick gain to Bal/Sta/Str, but they very rarely have the ability to gain a lot of the opposite (bal/str for finesse and acc/spe for S@H def). I've also seen several players change roles depending on how I've trained them. Flashy Def forwards-> Power Forwards. I imagine it working as a few different very specific roles with certain requisites to "unlock" them. For example a Flashy Fwd might "unlock" Flashy Def Fwd and then later Power Fwd. Based on how good and balanced his attributes get. But that's just how I imagine it working based on what I've seen in my saves. I've also seen Power Fwd's regress to Flashy Def Fwd. A post somewhere on these forums had made me think there was no way for the projected role to change like that until I noticed it happening quite a lot in my saves.

We're talking about different things.....

What you're referring to is ingame descriptions. What I'm talking about is the new "Player Role" system created from EHM1 (it didn't exist in EHM07).
The ingame descriptions are NOT necessarily congruent with the new Player Role! But my understanding is they will be as of the next update!

When EHM:EA was released last year some thought it was "essentially the same game as EHM07" and others (like myself) noted that there were significant changes "under the hood" - the Player Role system is one of these changes. In my post last August 30th I listed them off http://www.ehmtheblueline.com/forums/vi ... st#p198379

The individual Attributes are divided into different categories based on importance/relevance, and there are differences depending on the Player Role. In most cases determining "what Attributes connect to which Player Roles?" is "common (hockey) sense; IMO the most significant ones that might not be thought of, because they're not visible ingame, are Offensive/Defensive Role

FYI the TBL wiki page explaining all the Attributes http://www.ehmtheblueline.com/wiki/inde ... s:_Players
I'm not sure how accurate section 11 still is (I suspect OK to pretty good).....I don't think section 12 (regens) or section 15 (player roles) should be relied upon. Section 14 (calculating Attributes/CA) has some wisdom, but can't be relied upon for details as #1 the Attributes that can/do develop is different #2 the new Player Roles affect development #3 development has also been tweaked in regards to things like player age and more

There is a PDF "Player Role chart" that comes with the EHM Updater http://www.ehmtheblueline.com/forums/vi ... 10&t=11154

From a "testers perspective" I highly recommend the EHM Assistant (for use with saved games, spits out the player/bio data & ALL Attributes into a spreadsheet!) http://www.ehmtheblueline.com/forums/vi ... &start=400

Shindigs wrote:

FYI - the public discussion may have started up again, the discussion on the Researchers Forum & amongst the primary researchers has been ongoing

And as a part of the huddled masses it's nice to get to see more of it. Knowledge is power and all that good jazz!

I totally agree!
If you ever get the desire work on rosters, send me a PM with what you might be interested in doing! Whether modern/current TBL roster work, or work on a retro roster project, we can always use more thoughtful, data driven minds!

Shindigs wrote:I'm actually looking at Swe-2 players almost exclusively. But I do agree that we have two polar opposite approaches. I am very much looking into the micro management aspect whereas you focus on the macro side of things. Your analysis on which attributes grow/don't grow and the fact that you use some save editor or whatnot to actually see the hidden attributes gives a much deeper insight in the nuts and bolts of the attributes themselves. My interest lies more in the "how do I get player X to grow to his max potential in a reasonably small amount of time".

Both are significant pieces of the puzzle! And there's much knowledge that can be shared/applied too!

Shindigs wrote:While we're on the subject of nuts and bolts; how early is it possible for a defenseman to have his peak? I've got this one player, who isn't even a newgen named "Jacob Moverare" who has started declining at 23. Which just seems really early to me.

Hmmmmm.....does Moverare continue to decline?
Back when I did my EHM07 testing (3 tests, 125 players age 13-24 at startup, each test 10+ years, CA/PA and all Attributes tracked yearly) I was surprised to see that CA could actually drop sometimes! And then go back up! So a "bad/off year" that can happen IRL, where a player regresses, is likely possible ingame too

Shindigs wrote:Doesn't make the 2D engine any better though.

I don't get to playing much with all the editing/testing I do, but I'm a fan of the 2D myself (so I know there are issues! HaHa) - I would love to see more time/attention spent on improving the 2D myself (so that it becomes a more accurate representation of what's happening)

Shindigs wrote:And on the note of it getting a bit verbose when you try and explain things as deep as most game mechanics...I can relate. I should note for those who somehow haven't noticed yet that I am a bit prone to hyperbole, so when I start flinging superlatives just see it for what it is, me being overly emotive with my use of the english language.

Good to know! HaHa I'm more "cold & clinical"

I don't mean to lack "bedside manner" but it seems I do sometimes; I love data/testing and I love details and I love understanding things.....I discovered EHM 6 years ago, created a 1974 DB back in 2012 http://www.ehmtheblueline.com/forums/vi ... 116&t=9915 (and along with others I'm currently working on an improved/soon to be released 1998 DB); I've been helping with the TBL Rosters since 2011


Regarding your testing regarding Practice/Development....keep in mind Coaches have hidden Attributes too (both Coaching Atts and Mental Atts), and they may/likely play a part too (I've always wanted to spend more time looking into "Coaching/GM/Scouting Attributes" but never had the needed time...though I have done some work in this area recently, while working on the 1998 DB)

Was the "always tired players" topic covered already?

I don't get to playing much with all the editing/testing I do, but I'm a fan of the 2D myself (so I know there are issues! HaHa) - I would love to see more time/attention spent on improving the 2D myself (so that it becomes a more accurate representation of what's happening)

It should be noted the 2D engine (and supposedly the simulation) works a lot better in hybrid icing leagues. There are some quite major oversights with how the game deals with the redline in auto-icing leagues. I understand the game was developed for the NHL first and foremost, and it's no doubt what the majority of the player base is playing (that power fantasy). But last I checked more playable leagues use auto-icing than hybrid icing. Also at least in the 2D engine it looks a lot like the system that is used for offside in FM has been repurposed for Icings in EHM, which means the refs call a lot of icings that aren't icings (at least in the 2D) which is really annoying to watch when the AI scores a very large portion of their goals either straight from a face-off or just after it. Another thing I noticed when watching games in full (despite what that did to my sanity) is that if you have any offensive zone system that isn't "positional" the pointman will just pass to where the wingers would be if you were playing with "positional" when I started using that offensive system the random passes to no one that always lead to turnovers suddenly always went to one of my wingers. Stuff like that really rubs me the wrong way, since it's a technical issue forcing me to use a really boring tactical setup simply to play around the flaws in the match engine. It was after I noticed that and changed my tactical setup accordingly that I started seeing 80%+ win rate seasons. So some of the 2D bugs are actually simulation bugs, but knowing which when the 2D is such a loose representation is much harder than it needs to be. For example all your defenders always play horribly in the 2D, which lead to me always changing out my entire defensive lineup from year to year in a rage-induced frenzy when I still watched the 2D. When really the 2D engine just makes all defenders look incompetent always, which causes incorrect conclusions from a manager side. It's essentially the game lying to you about what is happening, and that is never a good thing in a tactical game (or in any game really).

Good to know! HaHa I'm more "cold & clinical"

I don't mean to lack "bedside manner" but it seems I do sometimes; I love data/testing and I love details and I love understanding things.....I discovered EHM 6 years ago, created a 1974 DB back in 2012 viewtopic.php?f=116&t=9915 (and along with others I'm currently working on an improved/soon to be released 1998 DB); I've been helping with the TBL Rosters since 2011

I figured fast enough, my first gut reaction was that you were a bit widdled at me. But tone often goes missing in forum posts, and based on the quality of your answers I quickly realized it wasn't the case. So no harm done:) I'm the same about needing to understand things, so I can relate. I come from a quite different background and only just got EHM on the sale early February. But I've done a lot of min/maxing in other games so the fact that so much was still unknown (at least to the public) in EHM struck me as odd.

Regarding your testing regarding Practice/Development....keep in mind Coaches have hidden Attributes too (both Coaching Atts and Mental Atts), and they may/likely play a part too (I've always wanted to spend more time looking into "Coaching/GM/Scouting Attributes" but never had the needed time...though I have done some work in this area recently, while working on the 1998 DB)

I'm aware. Something I've been meaning to ask though; I haven't tracked coach attributes, but do they grow too? Or are they stationary? All my numbers (in the second big post and beyond) are done with the exact same coaching staff (from the screenshot at the end), bar head coach. Cause my coaches have had "shock retirements" out of the blue 4 out of the 5 last seasons. One did it 3 weeks after signing a 4 year contract, which is just awful:S

AFAIK there is nothing like an Attribute value "above 20" BUT.....if two players each have a 20 in the same single Attribute but one has a significantly higher CA (overall ability) I don't think these players are "the same" even though they both have a 20 in the Attribute; this might be because they are "the same" but in the "related Attributes" there are significant differences

I see, my comment wasn't relating so much to the player not growing as a whole anymore. But if player X has gained 1 or more spe every single seasons from age 16 to about 19-20 when they will normally cap out, it seems logical that it would keep growing. Naturally the player would eventually hit his own highest possible skating speed. My issue is that if he reliably gained in it for 4 years, showing no sign of slowing down, then suddenly just stopped because the game doesn't have spe above 20. Even though logically that particular players personal maximum would be higher than 20. Part of the issue here seems to be that attribute gain isn't on a logarithmic scale. going from 1->2 skating and 19->20 takes just as much (as far as I can tell from my stats) which just isn't realistic. The amount of training needed to go from 19->20 should be years in the making. Not 4 months. If we were to think of it in regular gaming terms, it's standard for each new level to take considerably more experience points than the previous. A player going from level 1-10 in an MMORPG takes less than an hour, 90-100 takes between a day and a week. If this was how it worked in EHM it would make it very hard to ever hit attribute caps for players, which makes sense. But it can't possibly work that way based on what I'm seeing, which strikes me as very odd.

We're talking about different things.....

What you're referring to is ingame descriptions. What I'm talking about is the new "Player Role" system created from EHM1 (it didn't exist in EHM07).
The ingame descriptions are NOT necessarily congruent with the new Player Role! But my understanding is they will be as of the next update!

That is interesting, but it does sort of fit with my theory, I may just have worded it poorly. Say that Player X is "Player Role" w_finesse. When he's a kid with "meh" attributes he may have the ingame description "quick/flashy/skilled off fwd" as he improves this will eventually change to something like "goal-scoring(if he's w_sniper_finesse)/off playmaking fwd(if he's w_playmaker_finesse)". It's the game giving you some kind of feedback on how far long in his progressing within his role the player is. It's a bit vague still, no doubt. But it does give some hint to what "Player Role" he is, and how well he is progressing within it. This presumably means that many roles start with the same in-game description. But as the player improves he will branch into a more specific descriptor. If he is simply in the "w_finesse" role maybe he keeps being described as "quick/flashy/skilled off fwd" no matter how good he gets? Whereas the others eventually become goal-scoring, playmaking, etc.

I totally agree!
If you ever get the desire work on rosters, send me a PM with what you might be interested in doing! Whether modern/current TBL roster work, or work on a retro roster project, we can always use more thoughtful, data driven minds!

I appreciate the offer, but my actual real world hockey knowledge is pretty lacking for stuff like that. I used to play myself until I sustained a shoulder and knee injury in my teens. Since then I've been on a 10 year hockey hiatus. I hadn't watched a single game or kept up at all until I bought EHM in feb. and started keeping up with NHL happenings and the constant economical drama that is HockeyAllsvenskan. The team I manage in HockeyAllsvenskan in EHM used to be our nemesis team back when I played, we almost always faced them in the "state" finals, our top player from back then is actually currently playing in HockeyAllsvenskan too which makes it so much more relatable and fun.

Hmmmmm.....does Moverare continue to decline?
Back when I did my EHM07 testing (3 tests, 125 players age 13-24 at startup, each test 10+ years, CA/PA and all Attributes tracked yearly) I was surprised to see that CA could actually drop sometimes! And then go back up! So a "bad/off year" that can happen IRL, where a player regresses, is likely possible ingame too

So far he's declined for 3 out of the last 5 periods, his first one came at the end of a season where we made the playoffs and he had an above 7 average rating as a D-man with around 0.5PPG from defence. So he had a good year, but a -13(!) loss at the end of season. The summer before he also showed a small loss of -4, but he was on the phys schedule and it was during summer. So I didn't put to much stock in it since he gained those 4 back on Tech in the early season. The summer after on Tech2 he was static and this early season on Tech he showed a -7 loss, having a 7.07 rating, 19 points in 28 games and being 3rd in the league on assists. So he's been having career seasons and losing attributes like it's going out of style at the same time. It's really weird.

Historically speaking before this decline he always had bad summers (generally small loss or static, one decent gain of 4) but he did really well in the seasons (7,10,8,7,10 Trainable attribute gains) from the age of 17 to 22.

Edit: He's been losing the attributes all over the place too. From mentals that aren't directly affected by training, and from all the subsets of training. He's even had some of the bigger losses in the things that are on intense (-3 to off skill in 4 months while on Int @23y/o) so I can't even see any way a change of schedule would fix it by focusing more on getting back what he's lost. Cause he's literally losing everything, and I've only seen that happen over this long a period of time in players who are "getting old" but who gets old at 23!? It sucks even more because his consistency coupled with his ability to lead meant that I was going to make him our next Captain, once the current one hits that points where his "leg loss to absurd wage demand"-ration hits the negatives. Sadly it seems like Moverare might decline even faster than said 31y/o though. Which is a bummer.

Was the "always tired players" topic covered already?

Not as far as I've seen. I'd love to hear more.

Shindigs wrote: Not as far as I've seen. I'd love to hear more.

I have some players which are constantly at about 80% fitness, and thus I have them most of the time to the "resting" schedule. I observed it both in the NHL and in Denmark, where I'm playing now. Mostly is players from the top six which play more, but it doesn't seem too related to TOI as other players have similar TOI and aren't that tired.

Alessandro wrote:

Shindigs wrote: Not as far as I've seen. I'd love to hear more.

I have some players which are constantly at about 80% fitness, and thus I have them most of the time to the "resting" schedule. I observed it both in the NHL and in Denmark, where I'm playing now. Mostly is players from the top six which play more, but it doesn't seem too related to TOI as other players have similar TOI and aren't that tired.

Oh, THAT. Yeah, I've mostly had the issue with Defensive Defensemen. If your player are on a slightly too high stress training schedule for too long (this is all a hypothesis) they will eventually get long term exhaustion that means they will essentially freeze at a certain level of fitness until you rest them long enough to regain 100% Condition. I had every single Defensive D-man frozen at 94-96% maximum for an entire season. It seemed to have little to no impact on their performance since D-men use much less Condition in this game than Forwards. When I put them on rest during an international break they very slowly started to gain back condition above their 94-96 "caps". This was using a 2xInt, 3xMed, 1xLig schedule and having them all play 20+ minutes most, if not all, games.

While we're on the topic of Condition. Powerplay uses up waaaaay too much of it. Playing the same amount of time on "very low" tempo in power play will use up more Con than that amount of 5v5 played at normal/high. This means that if you have a very fast team (like I tend to) and you force a lot of hooking and slashing penlties you will have games with upwards of 20 minutes of PP, this will absolutely demolish the Con of all players involved, making it very hard to have them ready for the next game in full. If this happens several games in a row you can end up in a Condition hole so deep that you will need to rotate your squad for the rest of the month. So one of the best strategies for the AI to use against the player is to take a tonne of penalties on purpose to bone the team long-term.

The above does NOT work the other way around, as in you can't take penalties on purpose to put the AI in a Con hole. They cheat on con, if you don't believe me try playing in EIHL. Due to the squad rules there, they need to have one line with british players. But (no offense british people) british players are so darn bad they just put them all in the 4th line and then play on "Use only 3 lines" forward settings. This means their top line will drop to around 50 Con every single game. You play on average 2-3 games a week between the EIHL and BCC all season. A player resting will regain 10-12 con per day (70-84 a week) but playing 2-3 games at -50 Con a game per week costs you 100-150 Con a week. Which means that from the 2nd week of the season all your top 6 players would be stuck at low Con and get injured non-stop. This simply put doesn't happen, which means the AI cheats with Con (probably part of why playoffs are so hard as well) because if they didn't, they would have no players available.

Shindigs wrote:

Alessandro wrote:

Shindigs wrote: Not as far as I've seen. I'd love to hear more.

I have some players which are constantly at about 80% fitness, and thus I have them most of the time to the "resting" schedule. I observed it both in the NHL and in Denmark, where I'm playing now. Mostly is players from the top six which play more, but it doesn't seem too related to TOI as other players have similar TOI and aren't that tired.

Oh, THAT. Yeah, I've mostly had the issue with Defensive Defensemen. If your player are on a slightly too high stress training schedule for too long (this is all a hypothesis) they will eventually get long term exhaustion that means they will essentially freeze at a certain level of fitness until you rest them long enough to regain 100% Condition. I had every single Defensive D-man frozen at 94-96% maximum for an entire season. It seemed to have little to no impact on their performance since D-men use much less Condition in this game than Forwards. When I put them on rest during an international break they very slowly started to gain back condition above their 94-96 "caps". This was using a 2xInt, 3xMed, 1xLig schedule and having them all play 20+ minutes most, if not all, games.

I tried easing up the training schedules. I had 4x med 2x intens, now I have 3x med, 2x int and 1 lite. In January I'll have steady 2 games a week I'll let you know if things change.

I tried easing up the training schedules. I had 4x med 2x intens, now I have 3x med, 2x int and 1 lite. In January I'll have steady 2 games a week I'll let you know if things change.

Looking forward to seeing the results. But unless you actually let them rest back to 100% at some point it probably won't help. I've had that issue where my top 6 get to like 95-98% con between games, but never to 100. Eventually they will start regaining con slower and lose more in games, forcing lower tempo settings in the tactics screen. As soon as you let them get back to 100% once it seems to "reset" this mechanic, making them gain and lose at the same pace as they originally did.

Here is a screenshot with some new numbers from after my last season, which I actually managed to win (even the champions playoff) due to my goalie going full hero mode after having been out injured the first two games (BO5, not BO3 I misremembered before) he kept two clean sheets and let us come back to a 3-2 series win. Against the SHL side we lost 4-1....where have I seen that before? Oh yeah, literally every season ever, that's where!

The numbers are all the average gain/loss across all players in that age range over a 4 month period. Yes, this does mean that the average annual gain of players under 20 on Tech is 15.

Shindigs wrote:

Alessandro wrote:

Shindigs wrote: Not as far as I've seen. I'd love to hear more.

I have some players which are constantly at about 80% fitness, and thus I have them most of the time to the "resting" schedule. I observed it both in the NHL and in Denmark, where I'm playing now. Mostly is players from the top six which play more, but it doesn't seem too related to TOI as other players have similar TOI and aren't that tired.

Oh, THAT. Yeah, I've mostly had the issue with Defensive Defensemen. If your player are on a slightly too high stress training schedule for too long (this is all a hypothesis) they will eventually get long term exhaustion that means they will essentially freeze at a certain level of fitness until you rest them long enough to regain 100% Condition. I had every single Defensive D-man frozen at 94-96% maximum for an entire season. It seemed to have little to no impact on their performance since D-men use much less Condition in this game than Forwards. When I put them on rest during an international break they very slowly started to gain back condition above their 94-96 "caps". This was using a 2xInt, 3xMed, 1xLig schedule and having them all play 20+ minutes most, if not all, games.

While we're on the topic of Condition. Powerplay uses up waaaaay too much of it. Playing the same amount of time on "very low" tempo in power play will use up more Con than that amount of 5v5 played at normal/high. This means that if you have a very fast team (like I tend to) and you force a lot of hooking and slashing penlties you will have games with upwards of 20 minutes of PP, this will absolutely demolish the Con of all players involved, making it very hard to have them ready for the next game in full. If this happens several games in a row you can end up in a Condition hole so deep that you will need to rotate your squad for the rest of the month. So one of the best strategies for the AI to use against the player is to take a tonne of penalties on purpose to bone the team long-term.

The above does NOT work the other way around, as in you can't take penalties on purpose to put the AI in a Con hole. They cheat on con, if you don't believe me try playing in EIHL. Due to the squad rules there, they need to have one line with british players. But (no offense british people) british players are so darn bad they just put them all in the 4th line and then play on "Use only 3 lines" forward settings. This means their top line will drop to around 50 Con every single game. You play on average 2-3 games a week between the EIHL and BCC all season. A player resting will regain 10-12 con per day (70-84 a week) but playing 2-3 games at -50 Con a game per week costs you 100-150 Con a week. Which means that from the 2nd week of the season all your top 6 players would be stuck at low Con and get injured non-stop. This simply put doesn't happen, which means the AI cheats with Con (probably part of why playoffs are so hard as well) because if they didn't, they would have no players available.

Howdy Shindigs. Been following the debate on here closely. Pleased to see you've had some experience in the British leagues! I play most of my games 'there' albeit in the league below (EPIHL) with my hometown team.

I've noticed that running certain schedules on here (3 med/3 Ints, 3 med/2 Ints/1 light etc) and similarly found that condition would plummet.

Part of that I feel might be that while the above training regimes might work at the high levels of the game, with professional players, they could be ill equipped to use to train Semi-pro/Amateur players at this level (though you're right the EIHL is a pro-league). I've certainly looked at just lightly adjusting the standard schedules and using them at this level, purely to try and keep the condition of the players up between games.

Also; I've not checked the data recently, but might be that players at the British level have lower Natural Fitness than some of the NHL-Calibre players people have developed the 'old' schedules for, or else they might not have a 'hard' stat encoded for Nat Fitn, so the game is generating one based on the CA of the player.

Just my thoughts anyway. Would be nice if we could get some hard-and-fast rules down about training and what effects what.

The British teams I've tried out so far are Dundee Stars (in the EIHL) which worked just fine with the 2x Int, 2-3x Med 1-2x Lig setup. I was using all 4 lines though, cause I stole all the acceptable british newgens and put em on my 3rd/4th line. The other team is Slough Jets, and let's be real, the game isn't balanced around playing with players who have attributes that low. No one (including goalies) can ever really do anything with certainty. When a "godtier" player has 6 in his technicals you just end up with everything being so random. Tactics just don't work like they should and it all comes down to which goalie randomly actually saves some shots that particular night. And with games sometimes being 3 weeks apart, you never ever get match sharp. Making it even more random:/

I haven't gone back there since I started getting more into the training stuff due to just how random everything comes across there. But generally speaking, if you're having trouble with players getting too tired on a match to match basis, just record how much they gain back per day on your chosen schedule, then have them on rest and see how much they gain there, apply changes to Conditioning and Skating training downwards until you reach a level where you get like 2 less con back per day than on resting, and from that point on just make sure to set the trouble players on very low tempo in the PP and then preferably normal or higher in 5v5 if you can help it, and make sure to avoid playing any one player on both PP, BP and line 1 or 2. Honestly managing condition has much more to do with tactical settings than with training most the time.

Other useful things to do is to put the forwards on line 1 in PP2 and vice versa, and set PP in tactics to "equal" this will limit the amount of negative effect PP has on your con, without really having any big effect on your PP (I still normally have 15-20% PP return doing this, using a 50/30/20 normal/PP/BP training split). For reference the players I'm testing on are of a similar or slightly lower quality to EIHL players, the bigger teams like Steelers have almost the exact same wage budgets as all the non-bugged Swe-2 sides. As long as your players don't have Red level (i think that is the default low attrib colour?) stamina they should be fine. My players run the range of like 7 stam to 17 stam, and they do mostly fine.

Oh yeah, and another point on condition management, put your forward usage on "normal" and turn off line matching (unless you have a really really specific reason not to). In my experience all line matching does is mess up your condition management, since it will only match the fwds, not the d-men. So you can have your worst d-men on against their 1st line, making the point moot. I prefer to just have enough faith in my team's ability that they should have to try and match my lines, not the other way around. If you put your forwards on "overload" for more than like 10-15 minutes or so per game it tends to just wreck their con for way longer than it's worth. I only really use overload if I'm trying to pile on the pressure after a goal to break their morale, or to try and get back from being the odd goal down in the 3rd.

Another upside to coaching in England is that the Canadians love you for some reason, this makes it so you can get in Canadian coaches of much higher quality than the Swedish/Finnish coaches you get in Sweden. This actually makes it so your training can be a lot more potent in the British leagues, unless you're in the lowest ones where you can only get player/coach coaches who are just so bland and underwhelming. But such is life in amateur hockey.

Shindigs wrote:

nino33 wrote:AFAIK there is nothing like an Attribute value "above 20" BUT.....if two players each have a 20 in the same single Attribute but one has a significantly higher CA (overall ability) I don't think these players are "the same" even though they both have a 20 in the Attribute; this might be because they are "the same" but in the "related Attributes" there are significant differences

I see, my comment wasn't relating so much to the player not growing as a whole anymore. But if player X has gained 1 or more spe every single seasons from age 16 to about 19-20 when they will normally cap out, it seems logical that it would keep growing. Naturally the player would eventually hit his own highest possible skating speed. My issue is that if he reliably gained in it for 4 years, showing no sign of slowing down, then suddenly just stopped because the game doesn't have spe above 20. Even though logically that particular players personal maximum would be higher than 20. Part of the issue here seems to be that attribute gain isn't on a logarithmic scale. going from 1->2 skating and 19->20 takes just as much (as far as I can tell from my stats) which just isn't realistic. The amount of training needed to go from 19->20 should be years in the making. Not 4 months. If we were to think of it in regular gaming terms, it's standard for each new level to take considerably more experience points than the previous. A player going from level 1-10 in an MMORPG takes less than an hour, 90-100 takes between a day and a week. If this was how it worked in EHM it would make it very hard to ever hit attribute caps for players, which makes sense. But it can't possibly work that way based on what I'm seeing, which strikes me as very odd.

AFAIK there is no "Attribute capping" in EHM1 - but I think there should be!
IMO no player with a CA below a 170 should ever have a "technical" Attribute value in the 18-20 range

I have no idea about the coding details (like logarithmic scale).
I think in hockey terms moreso than gaming terms...I do think the player development in EHM takes a lot into account, and different Attributes develop for different reasons (not just Training).....skating used to be more static in EHM07, and now it develops more and I suspect connects more to age. And when I think of the fastest players I remember it doesn't occur to me that a Pavel Bure took longer to get from "almost as fast as possible" to "as fast as possible" - maybe (and this would be best, and I think it might be) what is actually happening is "in general with development" is different factors are given different weightings at different times and so development is not always the same (with hypothetical "same player/team/coaches/training" testing you'd get somewhat differing results); those "different times" could connect to age, a player playing in a league well below/above his ability, contract, coaching, practice/training, icetime, etc


In EHM07 for sure there were many "superplayers" (players who had extremely high/top values in virtually every Attribute).
The Player Roles are supposed to create a better (more realistic) variety of player, and help reduce/eliminate the superplayers.

Offensive Role and Defensive Role are very important too IMO (in terms of player performance, I'm not sure there's any affect in Practice).
Each hidden 1-20 rating indicates a willingness/interest in playing offense/defense & an understanding of offensive/defensive play & also connects somewhat to ability to play offense/defense - so for example two players may have the same Player Role (say w_sniper) and have the same CA and very similar Attributes but one will have an Offensive Role of 19 and the other 15.....I'd think the one with an Offensive Role of 19 will likely be a better sniper

Unless the coaching settings are limiting his opportunity to be a sniper! Lots to consider! HaHa It's why I love EHM

Shindigs wrote:

nino33 wrote:We're talking about different things.....

What you're referring to is ingame descriptions. What I'm talking about is the new "Player Role" system created from EHM1 (it didn't exist in EHM07).
The ingame descriptions are NOT necessarily congruent with the new Player Role! But my understanding is they will be as of the next update!

That is interesting, but it does sort of fit with my theory, I may just have worded it poorly. Say that Player X is "Player Role" w_finesse. When he's a kid with "meh" attributes he may have the ingame description "quick/flashy/skilled off fwd" as he improves this will eventually change to something like "goal-scoring(if he's w_sniper_finesse)/off playmaking fwd(if he's w_playmaker_finesse)". It's the game giving you some kind of feedback on how far long in his progressing within his role the player is. It's a bit vague still, no doubt. But it does give some hint to what "Player Role" he is, and how well he is progressing within it. This presumably means that many roles start with the same in-game description. But as the player improves he will branch into a more specific descriptor. If he is simply in the "w_finesse" role maybe he keeps being described as "quick/flashy/skilled off fwd" no matter how good he gets? Whereas the others eventually become goal-scoring, playmaking, etc.

I don't think the comments you're referring to connect to the new Player Role system...my understanding is they will next update...I think what you're seeing refers to Attributes (for example, "flashy" = Flair Attribute) and also "common hockey sense"

Regarding the idea that "many roles start with the same in-game description. But as the player improves he will branch into a more specific descriptor. If he is simply in the "w_finesse" role maybe he keeps being described as "quick/flashy/skilled off fwd" no matter how good he gets? Whereas the others eventually become goal-scoring, playmaking, etc." - No, the Player Role does not change (a w_finesse will always be a w_finesse).....but if the CA is high enough a player can become more "well rounded" than their Player Role

Regarding the idea that "many roles start with the same in-game description. But as the player improves he will branch into a more specific descriptor. If he is simply in the "w_finesse" role maybe he keeps being described as "quick/flashy/skilled off fwd" no matter how good he gets? Whereas the others eventually become goal-scoring, playmaking, etc." - No, the Player Role does not change (a w_finesse will always be a w_finesse).....but if the CA is high enough a player can become more "well rounded" than their Player Role

That was what I meant too actually, just worded it a bit poorly again. a w_sniper is always a w_sniper, but a bad/young/underdeveloped sniper might have the current in-game descriptor "flashy/quick/skilled Off Fwd" I've had several who did indeed have that descriptor, but as they improved through playing and training they got their current role in the position tab changed to "goal-scoring forward". My intrerpretation of what is happening there is that due to said player being of the w_sniper role, as he...actually she in the case I'm referring to...improved the attributes towards what she has an affinity to due to her role, she changed to the goal-scoring forward descriptor. If she was actually a simple w_finesse she would have remained a Quick Off Forward no matter how good she became. I've had several other finesse player with much, much better visible goal-scoring technicals than her never pick up that description. So maybe there is some connection, but not one good enough. Will be cool to see how much easier hidden attribute scouting becomes after next update.

AFAIK there is no "Attribute capping" in EHM1 - but I think there should be!
IMO no player with a CA below a 170 should ever have a "technical" Attribute value in the 18-20 range

I have no idea about the coding details (like logarithmic scale).
I think in hockey terms moreso than gaming terms...I do think the player development in EHM takes a lot into account, and different Attributes develop for different reasons (not just Training).....skating used to be more static in EHM07, and now it develops more and I suspect connects more to age. And when I think of the fastest players I remember it doesn't occur to me that a Pavel Bure took longer to get from "almost as fast as possible" to "as fast as possible" - maybe (and this would be best, and I think it might be) what is actually happening is "in general with development" is different factors are given different weightings at different times and so development is not always the same (with hypothetical "same player/team/coaches/training" testing you'd get somewhat differing results); those "different times" could connect to age, a player playing in a league well below/above his ability, contract, coaching, practice/training, icetime, etc


In EHM07 for sure there were many "superplayers" (players who had extremely high/top values in virtually every Attribute).
The Player Roles are supposed to create a better (more realistic) variety of player, and help reduce/eliminate the superplayers.

Offensive Role and Defensive Role are very important too IMO (in terms of player performance, I'm not sure there's any affect in Practice).
Each hidden 1-20 rating indicates a willingness/interest in playing offense/defense & an understanding of offensive/defensive play & also connects somewhat to ability to play offense/defense - so for example two players may have the same Player Role (say w_sniper) and have the same CA and very similar Attributes but one will have an Offensive Role of 19 and the other 15.....I'd think the one with an Offensive Role of 19 will likely be a better sniper

Unless the coaching settings are limiting his opportunity to be a sniper! Lots to consider! HaHa It's why I love EHM

That's the thing though. Someone like Pavel Bure became the way a "20" in speed was judged in the 90s. Really what a 20 means would need to change at all times depending on who is the best in world at skating/shooting/knitting at that point in time. But that would make things so confusing, your players attributes would fluctuate up and down on a weekly basis and you'd get so confused trying to make sense of it. The other option is for a 20 to be so insanely hard to reach that maybe one player per generation actually can reach it. If not, the 20 cap will actually limit player progression. And for someone like Bure, I'm not super familiar with his early career, but presumably most if not all his speed was gained before he reached the NHL over lord knows how many years of training, not to mention his incredible affinity for skating fast. That is a rarity, you don't get that many players who are super standout at one thing, like Kessel for "weak foot" wristers, Bure for speed, etc.

Whereas in EHM I currently have 4 or 5 players who are 23 or younger with either 20/20 or 19/20 Acc/Spe, they are a dime a dozen. You can see it in a league like HockeyAllsvenskan as it progresses. With the starting TBL roster one of the fasters players in the league has like 13/14. Two years later with EHM training, newgens and regens you have the average top 6 player rocking more than that, and several players up against the 20 cap. That does mean something isn't working quite right. It's too easy to become "the best in the world" at physicals specifically. Technicals are a lot more reasonable, you see very few players go up against the 20s there, it's essentially Ovi, Crosby and Toews around when they turn 30. They have so many 17+ attributes it's disgusting at that point in my saves. But since they are real players they rarely actually have their Physicals hit the 20s, it happens, but not as much as it does with virtual players and young "unknown" real players, like Moverare for example. He's a real player but he's like 16 at the start of the game, so a lot of unknowns. For me, before the weird decline, he had like 20 str, 17 stam, 17 bal. Which is a lot for a player like that, if you look at the newgen defenders most of them have that much or more in my team. I actually think all my virtual defensive D-men have 20 str at this point, and the oldest one is 23 and has had 20 str for years now. 20 should mean you are the best anyone could every possibly become at that aspect, not one of 3 young D-men on a Swe-2 team who are all that strong. If you look at FM16 most insane world-beater players have like 18 at most in just about everything except determination and flair, they seem to come in 20s quite often. Seeing someone with 20 in anything else is actually quite rare, even at the top top flight of football. In EHM you see it in 20y/os who aren't even drafted to the NHL or CHL.

Also there is naturally always a random element to the training, or anything really. Just about any game that involves simulation or multiple outcome stuff will use some sort of behind the scenes virtual dice rolling, would be awfully boring if it didn't. That's why you need to so many iterations to actually be able to say anything with certainty. Where there are so many unknown possible outside influences on your players' progression, the only way to get somewhat reliable data is to just keep on recording until the RNG evens out. Which naturally your large scale tests get around to doing much faster. Which leads to something I was wondering, were you controlling all teams putting them on the "Malhotra" schedules, or was it the AI doing the training? Because there seems to be some question as to if the AI cheats when it does the training. And since we know the AI does cheat in other facets of the game (condition managment, at the very least), it wouldn't be totally out of the realm of possibilities that it did.

I don't think the comments you're referring to connect to the new Player Role system...my understanding is they will next update...I think what you're seeing refers to Attributes (for example, "flashy" = Flair Attribute) and also "common hockey sense"

Flashy doesn't actually mean high Flair, it should, but it doesn't. That's what I thought it meant until I saw a lot of players with "meh" Flair have the descriptor. Flashy = mentals are the strongest part of his game, Quick = Physicals are the strongest part of his game, Skilled = Technicals are the strongest part of his game. In a lot of the players only described as "flashy/quick/skilled fwd" you can barely even tell which they should be, cause they tend to be jack of all trades with like 8-12 across the board. It also begs the question why the defenders don't get those. Why are there no "Flashy Defensive Defensemen"? Is it somehow less important to know what kind of attributes defensemen favor compared to forwards? Seems really strange to me, and also makes it a bit harder to judge D-men at a glance with hidden attribute scouting.

On the note of CA and PA, could you elaborate a bit more into exactly how that works. I read the wiki post about them. But certain things were less than thoroughly explained. My understanding is that they can be in multiple brackets of PA, some with huge variance which leads to those players you have 2 scouts saying 1 star, 2 saying 5 star etc. Then you have the "sure things" who have a fairly low variance, but they can come in different PA's like 200 was Gretzky mode if I recall correctly? How does CA/PA translate into actual total attributes, since that is what I use as a measuring stick, it would be nice to know how the two relate. The way the wiki post was written the roll of Off/Def seemed to decide how much attributes they have in the "offensive" or "defensive" attributes, explaning the players with 16+ acc/spe and like 5 sta/str. But then there is also the Player Role Off/Def roll? These are two different things right, one only affects how much visible attributes the player has in the two fields, and the other describes how willing they are to use said attributes on the ice? It gets a bit confusing at times.

Edit: I was just flicking through my player bio's and I think I found the bug that might be the cause of Moverare's decline.

As previously mentioned when you finish 1st or 2nd you play in a champions series BO5 in HockeyAllsvenskan, this does NOT count as "the playoffs". If you win or lose this game you either play an SHL side in the SHL Playouts, this counts as part of the SHL, not HockeyAllvenskan, thus also does NOT count as "the playoffs". Or another BO3/5 my memory is failing me against the winner of what IS counted as "the playoffs", thus this series does NOT count as "the playoffs" either. But since you won the league you count as having "made the playoffs". As a result every single one of my players has the bio entry "he will be upset that he didn't get to represent his team in this years playoffs" or something to that effect. Add in that Moverare had Key player status, is "Very ambitious" and is one of the assisting captains, and there has to be some pretty big backlash from him "getting cut" for the playoffs. Even though there literally was no way for me to play in "the playoffs" that is only played in by 3rd-8th place finish teams. Due to huge goalie issues I have been in those 3rd-8th for all but two seasons in this save. And the first one lines up right when Moverare had his huge -13 decline at end of season. Talk about a way to have one of your players ruined.

You guys should have a 3-way meeting with Riz
I eased up the practices a bit (as wrote earlier) and the conditioning problem is less of a problem, but still a problem. I don't get why some of my players are so off in conditioning, now, I may imagine that they are lazy or whatever (it's not the case probably, but I'll return to this later), but I think that if this was the case, the game should tell you. A coach can understand who practices and who doesn't have conditioning etc. in one minute of practice IRL, therefore the game should tell you. But this is not the case because I had this problem with Ovechkin, Kuznetsov and Backstrom in Capitals and they have excellent mental ratings.
This problem is costing me the regular season title lol. Now, my first line center has 5 of stamina (and it shows) (and Lindros him for next season), but I had the same problem with Ovechkin who has 19.