EHM Realism Patch

[bobmcgoo]

This patch removes the high-scoring nature of most - as far as we know, all - leagues in EHM. Players no longer go on 25-game point streaks, average ratings are much more realistically variable, teams score much less frequently, game scores are more random etc.

Please make sure if you do run any of these patches to use "full detail" for all leagues you want realism for; the patches don't work otherwise - see Archi's guide in the first download link below on how to do this.

A screen shot (click to enlarge if necessary):
http://imageshack.us/f/641/emhscreenshots.jpg/
The lines demonstrate how much more accurate simulation results are with the patch than without it. It only covers NHL scoring but our resident expert has confirmed it also significantly improves European league realism.

Note: I wanted realistically lower powerplay opportunities per game, but to do that I had to reduce aggression (there's no alternative) which in turn reduces fighting. So really what's happening in each game is now more realistic except for the lower number of fights.

All credit for the success of this patch goes to Archi and his mass editing tool, without which this project could never have happened. His tool is available as part of the download link below.

DOWNLOAD THE ATTRIBUTE EDITOR AND NHL SCORING REALISM PATCH HERE: viewtopic.php?f=88&t=9088

----------------------------------------------------------------------------------
A slightly improved version of the patch can be downloaded here: http://www.mediafire.com/?2270leyxlj26l9j. This patch merely stops the game thinking everyone is perfect in offensive situations (scouting reports would always call a player offensive), although simulation results are identical. This patch is recommended if you are starting a new game, however it is not needed if you have already started a game with the first patch.

----------------------------------------------------------------------------------
Also a second version of the patch was created to adjust for the huge advantage the game gives to heavier players. Some players get more hits, (a lot) more points and higher average ratings simply because they are extremely heavy - see the top post here for explanation http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=8909&start=88.

As well as including the scoring realism improvements, this patch also equalises every forward's and every defenseman's weight which erases this problem. This is the recommend patch version for highest realism; you can download it from http://www.mediafire.com/?ie5attw4vmwl77v.

[nino33]

Fantastic work! Thanks so much for your efforts!
I've been working on a 1974 db for a year and a half now, and I'm just a few weeks away from doing some serious playtesting, and I too would like a more realistic game...I'm very interested in what you've done (and how) as I'd like to do the same kinda thing with the 1974 db.

I haven't tested this database any further than six years; while I think it's unlikely the stats will suddenly go unrealistic again

Maybe I'm misunderstanding what you've done, but I've been thinking that Attribute adjustments would NOT be made by the computer when players were regened, so I would think the game/stats WILL "go unrealistic" again (which for me might work out, as I'll be able to "recreate" the high scoring, offensive minded 1980s)

I wanted realistically lower powerplay opportunities per game, but to do that I had to reduce aggression (there's no alternative)

Have you tried messing around with Referee Attributes at all? I'll be trying this (though I'll also likely try messing about with Aggression too, as I don't like the unrealistic level of fighting/penalty minutes)


Regards

[markiemark66]

This is an interesting project. I have wondered of ways to accomplish the same thing. I would help to test it.

[archibalduk]

I haven't tested this database any further than six years; while I think it's unlikely the stats will suddenly go unrealistic again

Maybe I'm misunderstanding what you've done, but I've been thinking that Attribute adjustments would NOT be made by the computer when players were regened, so I would think the game/stats WILL "go unrealistic" again (which for me might work out, as I'll be able to "recreate" the high scoring, offensive minded 1980s)

I think you're probably right. However, there are some attributes that regens pick up from the original player (IIRC some mental and physical attributes). Having said that, there is obviously no guarantee that those regens will end up in the NHL and so I expect it is inevitable that the NHL will go back to the norm again.

3) The database I have at the moment has changed every player in the game, so while the NHL is now realistic I imagine that other leagues won't be. But I think if there's enough demand for it me and Archi may be able to develop a way to only change NHL players.

This should certainly be possible. I won't have the time to add this to the tool for a while however.

Please feel free to distribute the tool with your csv if you like. Just my two cents, but I think actually that you might be better off distributing the tool and csv rather than a modified database. This will allow others to apply the changes to whichever database they like and also allows them to try tweaking it further.

A couple of other thoughts with regards to what may or may not happen once the tool/changes are just applied to NHL contracted players:

1) I wonder what effect it will have on the AHL and ECHL affiliates. These teams will have a mix between modified NHL players and unmodified AHL/ECHL contracted players.

2) I wonder what will happen to the NHL in general. Will the game regard the modified players as less valuable and therefore will sign players from Europe, etc? Will the modified players end up being dropped from NHL teams?

As I've said before, the graphs are extremely impressive.

[bobmcgoo]

I have to say I was worried about the effects of regens too, but firstly, I've simmed one game up to 2019 from 2011 and the stats are still very stable, very similar to previous years. I know that's not far but I'm probably going another 10 years or so, we'll see what happens. Secondly all changes made to this database are "acceptable" in the saved-game editor; i.e. you can continue to edit regens in exactly the same way as the players at the start of the database were originally edited. And Archi, I haven't touched player CA or potential, or reputations, so with regards to your second point I would guess the game will not see a player as "less valuable", but I might be wrong. However your first point is definitely possible; I'm not sure what could be done if it affected the AHL and ECHL realism...

I think releasing the tool instead of my database is a great idea, but of course it's your call not mine, I didn't write it. The most I could do is write the user guide for it!

@Nino No I didn't think to adjust officials, though it's a very interesting idea, thank you. I will try it at some point.

[nino33]

I think if there's enough demand for it me and Archi may be able to develop a way to only change NHL players.

This should certainly be possible. I won't have the time to add this to the tool for a while however.

Because the NHL draws players from everywhere (worldwide), I'm not sure how well just editing "NHL Players" would work....the players NHL teams brought in after start-up would not be edited, and this could cause problems...

all changes made to this database are "acceptable" in the saved-game editor; i.e. you can continue to edit regens in exactly the same way as the players at the start of the database were originally edited

How would one figure out each year which players were regens?

[markiemark66]

how about editing all players? is that too hard, is the process automated?

[Manimal]

This sounds very interesting.
I would like to know what changes you have made to the players. Was it only the aggression?

[nino33]

how about editing all players? is that too hard, is the process automated?

From what I understand he has edited all players (that's why he's not sure how the changes will affect other leagues); the process is to use an application/program that Archi wrote that does edit players in mass


I'm wondering...are the only tests done using a "quick sim"? - because my understanding is just simming produces noticeably different results/stats than actually playing/watching the games (and so for example in the Challenges participants MUST actually play/watch their games...just simming is not allowed)

[archibalduk]

And Archi, I haven't touched player CA or potential, or reputations, so with regards to your second point I would guess the game will not see a player as "less valuable", but I might be wrong.

Sorry, I didn't mean with regards to CA/PA/reputation. I meant that if players are now scoring less goals, etc, will the AI GMs therefore think that these players are now less valuable? Will they therefore sign some of their draftees (e.g. some of the older Europeans) who wouldn't have been subject to your tweaks (that is of course once the tool looks solely at adjusting NHL contracted players)?

From what I understand he has edited all players (that's why he's not sure how the changes will affect other leagues); the process is to use an application/program that Archi wrote that does edit players in mass

That's absolutely correct. It goes through every player in the database and makes the changes. There are various filters that can be applied if you only want to edit certain players. Also, you can set different settings for goalies, defencemen and forwards.

I think releasing the tool instead of my database is a great idea, but of course it's your call not mine, I didn't write it. The most I could do is write the user guide for it!

Yes, please feel free to upload the tool with your csv. You might also want to copy and paste into a text file parts of my PMs which explain what the tool does and how it works:

Extract the ZIP somewhere and then place your player.dat file in the same location. Then open the CSV to change the settings. Here is a brief explanation of the settings:

1) Adjustment type - Set this to either Value or Percentage depending on the adjustment you want to make (i.e. do you want to increase/decrease by a set value or by a percentage).

2) Alter zero settings - If a player has an attribute as zero, do you want to include these in the adjustments or leave it as zero. Y = adjust zero attributes. N = don't adjust zero attributes.

3) Allow zero settings - If the adjustment results in the player's attribute becoming lower than 1, do you want it to be set to 1 or 0? Y = set to 0. N = set to 1.

4) CA Min & Max - Determine the range of players to alter. If you want to alter all players, set it to 0 - 200. If you want to alter the attributes of only those with a CA of 100 or above then enter 100 - 200 etc.

5) The G, D and F settings determine the changes you want to make to each attribute. Enter a negative number if you want to decrease a value. If you set the Adjustment Type as Value then obviously it will just add/remove the number you enter from the player's attributes. If you are using Percentage, then the number you enter will be treated as a percentage e.g. if you enter -50 it will decrease the attribute by 50% whereas entering 50 will increase by 50%.

Once you have finalised your settings, save the CSV and then load the EXE file. It should only take a short moment to apply. You will see that the EXE loads a terminal window with information displayed. This window shows you how many records were written (it should be the same number as those read) and whether there were any errors.

The EXE creates a new file called new_player.dat which contains the modified data. Rename this to player.dat and put it back in your database folder.

I have given the application some testing and it appears to work okay. However I would recommend that you do a quick before and after comparision in the Pre Game Editor just to make sure that it has made the changes you expected.

Let me know how you get on.

P.s. If you're using Windows Vista or 7 and you run the application from within C:\Program Files\..\ then you'll need to use the Run as Admin option.

I've added a PA filter so you can now run your own combination of CA and PA filters simultaneously (e.g. you could apply to just those with a CA or 0-100 and a PA of 90-200 etc).

One thing to keep in mind is that PAs for some players are between -1 and -15. Therefore you might want to apply several different changes to the same player.dat file. This can easily be done by renaming the new_player.dat file to player.dat, making the relevant adjustments to the csv file and then reloading the exe.

You'll see there is a new section in the csv file for the PA min and max filter. I haven't tested this at all and so you might want to check in the Pre Game Editor that the changes are working as you expected. It's pretty easy to add new filters/attributes to the code and so I wouldn't expect there to be any issues (unless, of course, I made a typing mistake).

[bobmcgoo]

All sims were full detail for the NHL, I set myself as a GM every time but I always went on holiday. I haven't actually played a game where I've actively managed a team, although this may affect the stats of course.

@Manimal - I will take this opportunity to say thanks for all the amazing effort you put into these rosters, this project wouldn't have happened without you. However, no, I changed more than aggression, the exact changes I made will be on the csv file which operates Archi's tool. If anyone is interested in the explanations for each change, and if I can remember why I did what I did, I will be happy to answer.

@nino - like you and Archi say I think it would be difficult to accurately say what defines an NHL player. No solution would be perfect, but then the database I have isn't close to perfect; at best it's a good compromise. And I figured out regens purely by recongising a player or not. If you wanted a rule I would go by birthdate in the game, that should tell you who's real and who's generated.

@Archi - yes I see, but I think that the game goes a lot on reputation; even when the backup goalie is clearly outperforming the starting goalie, the game will continue to pick the starting goalie more often than not. But I agree, I think that just trying to edit "NHL" players is always going to be flawed, I can't think of a good definition, especially with the movement between the NHL and AHL. To me it depends if the user wants slightly more realism in terms of NHL stats or keeping the game as familiar as they have played it before as to whether they use your tool or not.

[bobmcgoo]

http://www.mediafire.com/?2766bwg1m3bnnab

This is the link for Archi's database editor. I've also included my edited database ("Manimal4edited") for those who, for whatever reason, don't wish to use the editor. However, Manimal4edited has not been comprehensively tested. Although I've ran three separate sims, one into season 2021-22, and stats remain very consistent, I cannot guarantee the same will happen every time until more testing has been done (notably, as nino33 pointed out, actively managing a team instead of being on vacation all the time may change the stats entirely, as may regens when they start to become common).

If you do use it, I hope you enjoy this database. If you do have any bug reports or suggestions, please do contact me or Archi. Otherwise, happy editing!

[nino33]

I figured out regens purely by recongising a player or not. If you wanted a rule I would go by birthdate in the game, that should tell you who's real and who's generated

I haven't used the SavedGame Editor a lot...can one sort by "Birth Year"? Or does one have to go through pretty much every team in the database when looking for regens?
For myself and the 1974 db I have players as young as 6, and so I'd have a bit more sorting between in-game and regens (though I do have lists of every player I've put in the database)

It's valuable work you've done for those that edit...thanks again for sharing!

[squadz]

Just want to thank you for this. Been using your database since you released it, still some games that are blowouts, but overall I'm extremely happy with how much better scoring is. Definitely saved my interest in this game, the high scoring was getting extremely annoying. No more crazy points per game.

Thank you again!

[archibalduk]

Just want to thank you for this. Been using your database since you released it, still some games that are blowouts, but overall I'm extremely happy with how much better scoring is. Definitely saved my interest in this game, the high scoring was getting extremely annoying. No more crazy points per game.

Thank you again!

How are you finding the players' individual performances? Are they still what you'd expect from real life?

I haven't had the chance to try out Alex's database yet, but I'm interested to hear other users' views.

[naash]

Thanks a lot! This seems to work really well with Updated Unfaking database too.

[bobmcgoo]

Just want to thank you for this. Been using your database since you released it, still some games that are blowouts, but overall I'm extremely happy with how much better scoring is. Definitely saved my interest in this game, the high scoring was getting extremely annoying. No more crazy points per game.

Thank you again!

I would say if there are some blowouts, that's realistic. Thanks for the praise, glad it's working well =)

How are you finding the players' individual performances? Are they still what you'd expect from real life?

I think forwards might be slightly too inconsistent (news items like X has five goals in his last five games are a lot rarer) but if consistency is made higher, scoring is too high... *shrug*. And you still get the whole issue of bigger guys playing better than small guys. Otherwise it's working fine for me.

[archibalduk]

Thanks; that's reassuring to know.

I'm pretty certain I'll be using your change file when I start up my next career game (when Manimal v4.1 is released). I think reduced consistency is worth it if it results in more realistic scores. Perhaps in some ways the reduced consistency compensates for how EHM favours forwards so much over defencemen... I really think having a change file rather than a database is the best way to go - users then have the choice whether to play with the traditional style or whether to go with the modified style (particularly as the changes are more aimed at the NHL rather than other leagues). I'll get the Attribute Editor and your change file added to the Downloads Database.

[squadz]

Just want to thank you for this. Been using your database since you released it, still some games that are blowouts, but overall I'm extremely happy with how much better scoring is. Definitely saved my interest in this game, the high scoring was getting extremely annoying. No more crazy points per game.

Thank you again!

How are you finding the players' individual performances? Are they still what you'd expect from real life?

I haven't had the chance to try out Alex's database yet, but I'm interested to hear other users' views.

Pretty much yes, besides to ovechkin, who is blowing out the league. I'll post standings when I can.

[archibalduk]

Pretty much yes, besides to ovechkin, who is blowing out the league. I'll post standings when I can.

That would be really interesting - thanks!

I was thinking of just vacation simming through a season to see how the player performances pan out over the first season. If I do then I'll also post some screenshots.

[nino33]

To those posting screenshots, if you can please post season ending "Referee Stats"

Thanks

[archibalduk]

I've uploaded the patch and bobmcgoo's patch to the Downloads Database: viewtopic.php?f=88&t=9088

[squadz]

So the season hasn't ended yet, about 46 games played. Here's the stats so far:

http://www.mediafire.com/?rb1sycqvqfpqcud

Above is the print file, let me know if need other stats.

[Alessandro]

Did anyone try this with non-NHL leagues?

[archibalduk]

No, I don't think anybody has mentioned doing that. The patch is designed for the NHL and I guess is really a band-aid fix, but it would be useful to know how all of the other leagues fare with it.