Jump to content
Sports Interactive Community
Sign in to follow this  
tran

A few criticisms of prospect development

Recommended Posts

1) Going back to the original NHL EHM, too many good prospects have really bad physical ratings.

Look around the NHL. There aren't many top flight players with 6 speed and or 8 stamina.

If a player is at a threshold of -7 or better (or 120+) there should be a minimum level of physical ability.

2) Players take too long to develop. Also, top 5 and even #1 picks are virtually useless in the NHL. This can't be by design, can it?

The speed at which players develop needs to be tweaked so they develop a little faster.

3) Prospects are too specialized. In the DB, players like Crosby or Malkin or even lesser players like Komarov... they are well rounded players.

The prospects this game develop all look like Mark Ribeiro (all offense, no defense) or Johnny Oduya (all defense, no offense).

Good players should be more likely to have all-around ratings... We see to many top players with 18 shooting and 18 deflections and then like 8 passing and 8 checking.

That's broken, if you ask me.

It really takes the fun out of the draft.

And in online leagues, it makes the teams that have great young players in 2016 great for a decade to come.

Share this post


Link to post
Share on other sites
1) Going back to the original NHL EHM, too many good prospects have really bad physical ratings.

Look around the NHL. There aren't many top flight players with 6 speed and or 8 stamina.

If a player is at a threshold of -7 or better (or 120+) there should be a minimum level of physical ability.

3) Prospects are too specialized. In the DB, players like Crosby or Malkin or even lesser players like Komarov... they are well rounded players.

The prospects this game develop all look like Mark Ribeiro (all offense, no defense) or Johnny Oduya (all defense, no offense).

Good players should be more likely to have all-around ratings... We see to many top players with 18 shooting and 18 deflections and then like 8 passing and 8 checking.

That's broken, if you ask me.

I've got three "regen tests" going right now, two are in 2025 (heading towards 2035/2045) but I haven't compiled the data yet

When I do I'll look for examples of support

Also, the NHL average CA in TBL 8.2 is 138 and a 4th liner/fringe player is in the high 120s, so a 120+ or -7 player is a fringe player who "might, at best" be an average NHL player (they'll never be a top 6 F/top 4 D)

In addition to the regen testing I've been doing extensive testing on startup Player Roles and Attributes (using TBL 8.2 and EHM 1.1) and at 6 different CA levels, and I don't see what you're describing (testing is all posted publicly in the TBL Data Editing Forum).

Also included in the TBL Data Editing Forum for public viewing/knowledge (with Riz's permission!) is a recently posted list of all the Player Roles and their Key/Essential/Non-Essential/Irrelevant/Regular Attributes (it should help significantly in understanding the new Player Role sytem)

I agree that "good players should be more likely to have all-around ratings" and they do compared to the less than good players, but many/most good players are not true two-way players and should not (I think EHM is much improved and more realistic in this area, and the previous versions of EHM produced "superplayers" that had high ratings in almost all/all Attributes for virtually every player that had a high CA - this was IMO extremely unrealistic and it's much better now)

2) Players take too long to develop. Also, top 5 and even #1 picks are virtually useless in the NHL. This can't be by design, can it?

The speed at which players develop needs to be tweaked so they develop a little faster.

What time frame is "to long"? Because EHM05 and EHM07 had ridiculously unrealistic development, with players regularly developing almost immediately, and I think the development now is way better/more realistic

Development is a very different in EHM now as opposed to previous versions of EHM, with more Attributes developing (all of them now it seems!) and more affecting development, and so strategies for the development/use of young players are not at all the same as they were with previous versions

It really takes the fun out of the draft.

And in online leagues, it makes the teams that have great young players in 2016 great for a decade to come.

I do not "know" this at all, but I suspect changes in the game that have occurred in 2016 (versions 1.1 and 1.2) have not been done in such a way that all the players/prospects in an existing league are "recreated" but rather future players/prospects are created with the improvements updates provide

I don't think players starting new games now and in recent months have been experiencing the issues you're referring to (it's not being reported, it was last year).....when the game starts up it creates the prospects as young as 9, so I suspect if your online league started in late 2015 IRL you'll need to be around 10 years (ingame time) into your game before you'll really see the fixes that have gone on this year

EHM now is much better/more realistic than EHM05/EHM07 and perhaps some players used to the older games may have difficulty adjusting [you have to develop a team over time, and actually develop players (with proper training/tactics for their Player Role) and old systems that worked fine when "superplayers" were the norm no longer work at all & the "trading GM style" that was so popular in EHM05/EHM07 will not really work anymore with the increased focus on development/tactics, which again is I believe significantly more realistic)

Share this post


Link to post
Share on other sites

Just because EHM 05 was "unrealistic" doesn't mean the new Eastside is realistic.

It needs to be realistic. Every year 18 year olds make the jump. With EHM, you often have to wait til a top 5 pick is 21 before they're as good as a third liner.

Considering the age that Gretzky, Lemieux, Yzerman, Fedorov, etc. etc had their career highs, the new EHM is not realistic at all.

Many superstars peak at 20-23 years old.

That's an age at which many Eastside top rated prospects are only beginning to be good enough to play in the NHL

Share this post


Link to post
Share on other sites
Just because EHM 05 was "unrealistic" doesn't mean the new Eastside is realistic.

It needs to be realistic. Every year 18 year olds make the jump. With EHM, you often have to wait til a top 5 pick is 21 before they're as good as a third liner.

Considering the age that Gretzky, Lemieux, Yzerman, Fedorov, etc. etc had their career highs, the new EHM is not realistic at all.

Many superstars peak at 20-23 years old.

That's an age at which many Eastside top rated prospects are only beginning to be good enough to play in the NHL

Like I said, I'll try to look for support/evidence of your concerns

I've got the data, it'll be really easy to see if what you're saying is actually happening

Is your only experience based on online play? If so, then that's why your "problem" is occurring IMO (I honestly think the things you're talking about were a significant issue last year but aren't now, so a game started last year is going to have issues that have been fixed for months); in the online league I was in I was completely rebuffed when I tried to explain such things, and when I quit it was easier doing so in large part because I knew I wouldn't want to continue playing with the database/issues they were playing with anyways

And given the game hints say that forwards usually peak in the 27-33 age, you're asking the game to be something it says up front it's not

And what makes you think many NHL superstars peak at 20-23?

I googled "what age do NHL players peak at?" https://www.google.ca/#q=what+age+to+nhl+players+peak+at%3F and the answer that came up at the top (from actual research) for "most" was 29, and it referenced 25 year old Kane and 26 year old Toews (superstars) as being in the prime age for NHL forwards.....the UBC article discussion didn't even include a reference to players age 20-23

More interesting research showed up in the google search (like the nerdhockey article, the arcticicehockey article and the sbnation article) and they don't support your view either

Share this post


Link to post
Share on other sites

Very few freshly drafted 18yr olds are ready for the NHL. Most are sent back to juniors and really aren't ready for the NHL until 19 and even 20 is more realistic. Very few NHL teams rush their prospects into the NHL these days and it can do more damage than good if they are. A player has to be a freakish generational talent to be ready to jump into the NHL right away and these players only come around once in a blue moon. And even those players are never frontline players for their teams. I think EHM is pretty good with respects to prospect placement and development. In my opinion it's the best it's ever been. I've been playing EHM for over 10 years and I don't think the sim has ever been as good as it is now.

Share this post


Link to post
Share on other sites

Anywhere from 2 to 5 kids play every year.

And they are usually top 5 picks.

Connor McDavid at 18 was better than 90 percent of the players in the NHL.

EHM doesn't create Connor McDavids and Austin Matthews.

I hope you'll just drop your preconceived ideas and pay attention to what I'm saying and encourage Riz and co to fix an element of this game that needs to be fixed

Share this post


Link to post
Share on other sites
Anywhere from 2 to 5 kids play every year.

And they are usually top 5 picks.

Connor McDavid at 18 was better than 90 percent of the players in the NHL.

EHM doesn't create Connor McDavids and Austin Matthews.

I hope you'll just drop your preconceived ideas and pay attention to what I'm saying and encourage Riz and co to fix an element of this game that needs to be fixed

For myself, I've got lots on the go and have decided given your argumentative responses/lack of appreciation for actual IRL data and testing EHM (and the many hours it takes to compile testing results) I'm not going to even bother looking into the issue.....

Share this post


Link to post
Share on other sites
Anywhere from 2 to 5 kids play every year.

And they are usually top 5 picks.

Connor McDavid at 18 was better than 90 percent of the players in the NHL.

EHM doesn't create Connor McDavids and Austin Matthews.

I hope you'll just drop your preconceived ideas and pay attention to what I'm saying and encourage Riz and co to fix an element of this game that needs to be fixed

Your attitude is terrible and your opinion about McDavid being better than 90% of the league is statistically wrong.

Share this post


Link to post
Share on other sites

IMO it's likely that EHM needs to improve in it's ability to generate players (the handful at best per year this applies to) that can play in the NHL at 18-19 but views like "Many superstars peak at 20-23 years old" & "Connor McDavid at 18 was better than 90 percent of the players in the NHL" don't fit with reality (and they don't fit within the common/known EHM editing practices either)

There's lots for volunteers to look into/do regarding EHM, and I've been so much happier lately that I can pursue my own interests (that are still very helpful for the community), and don't feel the guilt/obligation (my issues!) that I did before & without feeling the same level of stress when I feel the poster is rude/arrogant/ignorant (I'm not going to force anyone to accept my help! HaHa now I just move on to something else/there's lots to do)

Share this post


Link to post
Share on other sites

Nothing is ever perfect. But I have been playing this sim for literally over 10 years and it's never been better. I have no issues with the prospect system in the current state. I rather enjoy managing my prospects after draft. Bringing them through the system until they become important pieces is quite satisfying. I think it would be extremely boring and not very realistic if drafted players were immediately ready for duty. Yes, there may be X amount of players ready for the NHL at draft, but those players are definitely the exception and not the norm.

Share this post


Link to post
Share on other sites

That's nice that you are happy with the game. I am happy for you:)

But for fans who want realism, there should be franchise players in every draft.

It's unacceptable to have so many top 5 picks are are not even close to being ready to being impact players.

Please fine tune the prospect development so that the players who are world class tend to develop faster and earlier.

Share this post


Link to post
Share on other sites
Like I said, I'll try to look for support/evidence of your concerns

I've got the data, it'll be really easy to see if what you're saying is actually happening

Is your only experience based on online play? If so, then that's why your "problem" is occurring IMO (I honestly think the things you're talking about were a significant issue last year but aren't now, so a game started last year is going to have issues that have been fixed for months); in the online league I was in I was completely rebuffed when I tried to explain such things, and when I quit it was easier doing so in large part because I knew I wouldn't want to continue playing with the database/issues they were playing with anyways

And given the game hints say that forwards usually peak in the 27-33 age, you're asking the game to be something it says up front it's not

And what makes you think many NHL superstars peak at 20-23?

I googled "what age do NHL players peak at?" https://www.google.ca/#q=what+age+to+nhl+players+peak+at%3F and the answer that came up at the top (from actual research) for "most" was 29, and it referenced 25 year old Kane and 26 year old Toews (superstars) as being in the prime age for NHL forwards.....the UBC article discussion didn't even include a reference to players age 20-23

More interesting research showed up in the google search (like the nerdhockey article, the arcticicehockey article and the sbnation article) and they don't support your view either

Stamkos best season - 21

Ovechkin best season - 21

Gretzky best season - 20 (92 goals) 24 (215 points)

Fedorov best season - 23

Yzerman best season - 23

M Lemeiux best season -22

Crosby best season - 19 or 22

Kane best season - 27

Malkin best season -- 22 or 25

Corey Perry - 25

H. Sedin - 29

Thornton - 26

Forsberg - 22

Sakic - 31

Messier - 29

Trottier -- 22

Laflleur - 25 or 26

Bobby Clarke - 26

Bobby Orr- 26

Look how many superstars peaked at 23 or earlier?

There also appears to be a disconnect with Riz's system and with the players in TBL databases.,

If you notice, the TBL players in the game and about to join the game are great compared to the players produced by the game.

The point is, the players like Gretzky, Lemieux, Crosby, Ovechkin, Yzerman, etc... They are rare.

But they exist.

And this game doesn't create them in a way that resembles the real world.

If you don't see the problem with that, I don't know what to say.

Throughout history the NHL has had 18, 19 and 20 year olds dominate.

Share this post


Link to post
Share on other sites
If you don't see the problem with that, I don't know what to say.
Actually after rescinding my offer to help because of your negativity/rudeness, I still tried to help by saying "IMO it's likely that EHM needs to improve in it's ability to generate players (the handful at best per year this applies to) that can play in the NHL at 18-19." Did you actually read anything I wrote? If you did, did you not understand it?

I don't accept the inflated opinion you hold of your understanding of hockey and EHM (especially EHM), and much of what you say is wrong/hyperbole and negative/attacking, and that's the problem for me.....that's why I said "when I feel the poster is rude/arrogant/ignorant I'm not going to force anyone to accept my help!"

P.S. 11 of your 19 examples (over almost 50 years!) don't even support your views! It's like you don't even try to be correct/make sense (IMO you're your own worst enemy, both with your negativity and with your factual errors/hyperbole).....and my most recent testing has got me thinking the problem IS the database

Share this post


Link to post
Share on other sites
Actually after rescinding my offer to help because of your negativity/rudeness, I still tried to help by saying "IMO it's likely that EHM needs to improve in it's ability to generate players (the handful at best per year this applies to) that can play in the NHL at 18-19." Did you actually read anything I wrote? If you did, did you not understand it?

I don't accept the inflated opinion you hold of your understanding of hockey and EHM (especially EHM), and much of what you say is wrong/hyperbole and negative/attacking, and that's the problem for me.....that's why I said "when I feel the poster is rude/arrogant/ignorant I'm not going to force anyone to accept my help!"

P.S. 11 of your 19 examples (over almost 50 years!) don't even support your views! It's like you don't even try to be correct/make sense (IMO you're your own worst enemy, both with your negativity and with your factual errors/hyperbole).....and my most recent testing has got me thinking the problem IS the database

Why do you resort to insults?

Since you've done all the testing, please show me all the 19-22 year old Hart Trophy candidates (EHM created players) in your sims.

I would love to see your results and compare those results to Reality.

Show me the way this game creates a Sidney Crosby or Mario Lemieux or Connor McDavid.

If you want to argue that a majority of NHLers aren't ready for the NHL at 18 or 19, you won't get an argument for me.

But what I'm telling you is many of the very ELITE players are ready.

You may have noticed that all the players I listed were Hart Trophy or Hall of Fame players.

I appreciate that the drafts are more random. But until you show me how EHM creates a 19 year old 100-point man, I'm simply going to continue to believe that EHM development doesn't allow for it.

Because I've started 10-15 games now and never seen it.

The only players who do so are the ones created by TBL.

And even in those cases (-10 Joe Veleno, for example) the development is very slow.

Share this post


Link to post
Share on other sites
Why do you resort to insults?
Nothing I said was meant to be an insult, I really believe what I said (emotionless facts).....but I'm done with you/this

Share this post


Link to post
Share on other sites

There is nothing rude/arrogant about criticizing a development model that doesn't mimic reality.

I have asked you to show me the example of a first or second year player capable of scoring 100 points or 50 goals.

I have asked you to show me the examples of 18, 19 and 20 year olds who were able to do things like Stamkos, McDavid and Crosby.

I have asked you, because I don't believe this game produces these talents at this age. And I believe that is unrealistic.

Share this post


Link to post
Share on other sites
But for fans who want realism, there should be franchise players in every draft.

It's unacceptable to have so many top 5 picks are are not even close to being ready to being impact players.

Ok I'm going to somewhat disagree with you here. There are franchise/Elite players in every draft, but the number of them that are able to make their respective NHL teams and be put in a position to succeed are few. Attribute wise EHM could be creating players that can jump right in after the draft, but i'm wondering if the Coaching Attributes/GM Attributes/AI are not allowing said rookies to flourish in roles that they'd score in, like the Human GMs are capable of.

I have asked you to show me the examples of 18, 19 and 20 year olds who were able to do things like Stamkos, McDavid and Crosby.

In 1998-99 DB Save that I have there are a few 18-21 year olds that have come into the league and done VERY good(Two defensemen on the level of Ekblad, for example). The FWD shown followed almost the same path as Stamkos. Good 1st year, excellent 2nd year(Granted not 50 Goals, but 50 goals is rare in EHM. Between 1998-2017 I only saw 6 50+ goal scorers, team scoring is on par with the past few seasons 284-322 Goals scored)

Now in this era I do believe we will see 21-24 year olds be the dominating forces. We're already seeing it not only in the NHL but in the other Professional leagues.

wkTglzW.jpg

mb9drdv.png

I have asked you to show me the example of a first or second year player capable of scoring 100 points or 50 goals

In all honesty in this day and age with hockey. I don't think we will ever see this again. OV and Crosby did it right after a Lockout with huge rule changes that teams needed to adjust to and they were also Generational Players. But the others that did this were during the free-offensive style of the 80s-early 90s. We'll NEVER see those numbers again.

I looked up the players that have scored 50+ POINTS as 18-19 year olds. Here's the list with (draft years) in the last 20 or so years

Sidney Crosby(2005)(As an 18 and 19 year old)

Jeff Skinner(18 and 19 year old)(2010)

Nathan MacKinnon(2013)

Jaromir Jagr(1995)

Ryan Nugent-Hopkins(2011)

Steven Stamkos(2008)

Patrick Kane(2007)

Marian Gaborik(2000)

Tyler Seguin(2010)

Matt Duchene(2009)

Vincent Lecavalier(1998)

Anze Kopitar(2005)

Rick Nash(2002)

Jack Eichel(2015)

Jonathan Toews(2006)

John Tavares(2009)

Gabriel Landeskog(2011)

Jarome Iginla(1990)

Connor McDavid(2015)(He would have hit 50 points)

And this game doesn't create them in a way that resembles the real world.

They are once/twice in a decade if that. Can't say if I've seen/noticed a generational players being generated(though I did have a 192 PA player created in the 1998-99 db

Share this post


Link to post
Share on other sites

A helpful, logical/reasoned post Rittflyers! :thup:

Can't say if I've seen/noticed a generational players being generated
I have done testing on PA distribution/regen testing three times this year, in January (after 1.0) and in March (after 1.1) and now this month (looking at 1.2); I run multiple tests going out decades each time

I most definitely see generational players being produced, and have posted the data at TBL (and am actually compiling the "Test 3" data right now) http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=16787

For those that might be interested, there's a number of other threads I've done full of testing results in TBL's Data Editing Forum http://www.ehmtheblueline.com/forums/viewforum.php?f=110

Share this post


Link to post
Share on other sites
On 7/23/2016 at 23:19, Rittflyers said:

 

"Ok I'm going to somewhat disagree with you here. There are franchise/Elite players in every draft, but the number of them that are able to make their respective NHL teams and be put in a position to succeed are few."

They are few. Sure. Exactly.

They are few.

But they are real.

McDavid, Eichel. Hanifin. Ekblad. Pastrnak. McKinnon. Seth Jones. Elias Lindholm. Monahan. 

These are in just the last 3 years.

The idea that Crosbys and Ovechkin-type rookie campaigns are a thing of the past? Not even close to true.

 

It truly boggles the mind that anyone believes it's realistic to have so many top 5 picks be NOWHERE near ready.

Edited by tran

Share this post


Link to post
Share on other sites
22 hours ago, tran said:

The idea that Crosbys and Ovechkin-type rookie campaigns are a thing of the past? Not even close to true.

 

It truly boggles the mind that anyone believes it's realistic to have so many top 5 picks be NOWHERE near ready.

I agree with you, there is something wrong with the way prospects are rated for the draft. The game is producing players of that quality but for some reason not recognizing them as top 5 picks and in some cases not highly rating them at all. I think CA/PA is broken or if WAD then something behind the reasoning used is faulty. I don't know how much time you have for scouting in online games but really the only way around this is to extensively scout youth players and then ignore the scout recommendations and use your own judgement when it comes to evaluating skill and potential, once I started doing that I've had a few kids have close to point per game seasons and none of them were rated highly by any of my scouts. I've got a 19 year old Russian kid on my team right now in the current season of my current save who has 23 goals and 51 points from 26 games played. I drafted him with the 29th overall pick. My scouts all told me he was a 2 star CA/ 1 star PA kid who wouldn't crack the lineup.

Share this post


Link to post
Share on other sites
51 minutes ago, Didaro said:

I agree with you, there is something wrong with the way prospects are rated for the draft

I think this is a different issue, as tran is saying in EHM not enough top 5 players (draftees) are "ready" to immediately make an actual impact in the NHL

 

I (and I'm sure many/most/all others!) think you're right Didaro, Scouting needs improvement! I myself think it needs significant improvement! Ideally everything regarding EHM would be fixed/improved upon by the next update, but alas that's just not reality, and things are prioritized.....and of course people have different views on what should be prioritized! HaHa

I myself like to do extensive testing and obtain real "proof" that results in a complete/thorough understanding of an issue before pushing for change/improvement (for myself I want to have solutions I'm pretty sure of, as opposed to just complaining)

IMO improving Scouting (because it affects far more of a user's gameplay experience) should be a high priority and is IMO far more important than the issue of getting a couple/few players to be more "NHL ready" each year. I also think that for the database researchers a lot has been learned about how the game works since the last database release (last fall, when EHM was still in early access!).....perhaps we'll see improvement in top picks being more NHL ready when the next TBL database is released (based on my years of history with EHM editing and with the TBL rosters I think it's very reasonable to think so)

 

 

tran - FYI Until the last 24 hours I had never used an "ignore list" on any site, but as part of my own learning of how not to be so bothered by some postings/posters (that I consider to be far more unhelpful than helpful), I've put you on my ignore list as I'm just not interested in reading your hyperbole and rudeness (in this thread I even offered to help and you carried on with the same hyperbole/rudeness!)

 

Share this post


Link to post
Share on other sites
1 hour ago, Nino-33 said:

IIMO improving Scouting (because it affects far more of a user's gameplay experience) should be a high priority and is IMO far more important than the issue of getting a couple/few players to be more "NHL ready" each year. I also think that for the database researchers a lot has been learned about how the game works since the last database release (last fall, when EHM was still in early access!).....perhaps we'll see improvement in top picks being more NHL ready when the next TBL database is released (based on my years of history with EHM editing and with the TBL rosters I think it's very reasonable to think so)

 

 

 

 

I know this sounds nit-picky but I don't agree that scouting is broken, my scouts do exactly what they're supposed to do which is identify players based upon their current ability rating and for young players potential ability. The reason so many recommendations are wrong is because CA/PA is broken. If that is fixed then scouting will work fine without any revamping and the teenage players in the game who are NHL ready will be properly evaluated. Tran is wrong when he says they're not in the game, they're generated every year, you just have to look for them without regard to CA/PA ratings.

Share this post


Link to post
Share on other sites
2 hours ago, Didaro said:

I know this sounds nit-picky but I don't agree that scouting is broken, my scouts do exactly what they're supposed to do which is identify players based upon their current ability rating and for young players potential ability. The reason so many recommendations are wrong is because CA/PA is broken. If that is fixed then scouting will work fine without any revamping and the teenage players in the game who are NHL ready will be properly evaluated. Tran is wrong when he says they're not in the game, they're generated every year, you just have to look for them without regard to CA/PA ratings.

Not nit-picky at all to me!

How do you know scouts are identifying players based on their CA/PA? I thought there was many posts saying that this is not true (for example, the players drafted are not at all picked in order of PA and bad scouting reports on high PA players are not uncommon)

 

What makes you think CA/PA is "broken"? What exactly does that mean?

 

Actually there is a big difference already in how some players are rated, and changes are ongoing as researcher understanding increases.....the NHL CA average is about 138 in the TBL Rosters, and Major Junior players used to be given much lower CAs in EHM07 because they almost immediately/immediately progressed to NHL caliber in EHM07 - but now development is different (slower/more realistic)  the Top draftees are given a CA that's much higher than they ever were with EHM07 (McDavid's CA in TBL 8.2 is 142 and Eichel's is 138, by far the highest starting CA for top draftees ever.......looking at the top 5 picks from TBL 7.0 their CAs were 96-96-98-99-90 & looking at the top 5 picks from TBL 6.1 their CAs were 118-115-88-121-95)

And it's really important to remember that researchers were still learning LOTS about how the game's changed & an enormous amount more has been learned since the ratings being discussed were released! The 2015-16 TBL rosters were "start of season" rosters and the CA/PA and Attributes haven't been edited since last November! So I'm quite sure you'll see improvement in the 2016-17 rosters due to researchers having gained more knowledge/understanding since the game came out

I'm not sure what you're basing your thoughts on regarding CA/PA - I've been actively editing EHM since 2010, and in addition to thousands of hours of testing (and compiling results in excel spreadsheets) I created the 1974 DB and co-created the 1998 DB & I actually did all of Major Junior (WHL/OHL/QMJHL) from 2012-2015 for the TBL rosters, so I'm quite familiar with the CA/PA ranges for young players, and the CA range for top young draftees has without doubt changed/is changing

 

Tran was talking about player PERFORMANCE (how many points they scored/whether they could win the league MVP) and not how they were evaluated.....seems to me you were the first one to mention evaluation in any way

 

And finally, I don't understand when you say "they're generated every year, you just have to look for them without regard to CA/PA ratings" - CA/PA ratings are NOT visible ingame, and people that use the EHM Assistant to see such ratings when "playing" the game are cheating (and by the way, if you did look "by PA" you most definitely would see them, as my extensive testing has shown http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=16787)

 

 

Edited by Nino-33

Share this post


Link to post
Share on other sites
16 hours ago, Nino-33 said:

Not nit-picky at all to me!

How do you know scouts are identifying players based on their CA/PA? I thought there was many posts saying that this is not true (for example, the players drafted are not at all picked in order of PA and bad scouting reports on high PA players are not uncommon)

 

What makes you think CA/PA is "broken"? What exactly does that mean?

 

Actually there is a big difference already in how some players are rated, and changes are ongoing as researcher understanding increases.....the NHL CA average is about 138 in the TBL Rosters, and Major Junior players used to be given much lower CAs in EHM07 because they almost immediately/immediately progressed to NHL caliber in EHM07 - but now development is different (slower/more realistic)  the Top draftees are given a CA that's much higher than they ever were with EHM07 (McDavid's CA in TBL 8.2 is 142 and Eichel's is 138, by far the highest starting CA for top draftees ever.......looking at the top 5 picks from TBL 7.0 their CAs were 96-96-98-99-90 & looking at the top 5 picks from TBL 6.1 their CAs were 118-115-88-121-95)

And it's really important to remember that researchers were still learning LOTS about how the game's changed & an enormous amount more has been learned since the ratings being discussed were released! The 2015-16 TBL rosters were "start of season" rosters and the CA/PA and Attributes haven't been edited since last November! So I'm quite sure you'll see improvement in the 2016-17 rosters due to researchers having gained more knowledge/understanding since the game came out

I'm not sure what you're basing your thoughts on regarding CA/PA - I've been actively editing EHM since 2010, and in addition to thousands of hours of testing (and compiling results in excel spreadsheets) I created the 1974 DB and co-created the 1998 DB & I actually did all of Major Junior (WHL/OHL/QMJHL) from 2012-2015 for the TBL rosters, so I'm quite familiar with the CA/PA ranges for young players, and the CA range for top young draftees has without doubt changed/is changing

 

Tran was talking about player PERFORMANCE (how many points they scored/whether they could win the league MVP) and not how they were evaluated.....seems to me you were the first one to mention evaluation in any way

 

And finally, I don't understand when you say "they're generated every year, you just have to look for them without regard to CA/PA ratings" - CA/PA ratings are NOT visible ingame, and people that use the EHM Assistant to see such ratings when "playing" the game are cheating (and by the way, if you did look "by PA" you most definitely would see them, as my extensive testing has shown http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=16787)

 

 

What else other than CA/PA can a scout use to evaluate a player? In order to make the game interesting and replayable there have be some modifiers in place but the recommendations have to be based on CA and PA. Game performance is clearly not used in scout recommendations for veteran players and can't be used for prospects playing in non-playable leagues because in their case there aren't any games. That leaves CA and also PA for young players.

Broken was perhaps the wrong word to use about CA/PA. A better way to put it might be to say that the method used to arrive at a CA rating for a player usually works well but sometimes allows for unusual results which skew CA upward for a player who is clearly not at that level. 

Tran was talking about performance for much of the thread but the first thing Tran mentioned in his first post was physical attributes. He then talked about the other attributes being in his opinion unbalanced and that causing top prospects to not be NHL ready. His comments on performance were based on his observation that top prospects do not have the attributes necessary to step in and play well in the NHL right away.

CA/PA numerical values aren't visible in game but the star ratings given for current and potential ability in scout reports are based on them, aren't they? I don't use an editor, I like a little bit of mystery when playing but while I don't know the exact values I can certainly guess the range those values are in using the star ratings in the reports and the league the player is being rated for. 

Here's an example of what I'm talking about. This is an extreme example but not even close to the only one and it perfectly illustrates the problems I'm seeing, in fact this is the reason why I decided to ignore scout recommendations. These recommendations were so absurd that I still remember them. A few saves ago while GM'ing an NHL team I had the #4 overall pick in the 2028 entry draft. My head scout, my Asst. GM, and my head coach all recommended the same player. The player was an 18 year old playmaking center who had spent the last two years with an OHL team. He was rated 3 star current ability and 5 star future ability and all of them said he could step in and contribute immediately and would become a key player. All of this players physical attributes were in the red; acceleration and speed of 4, stamina was 4, agility was 5, strength was 3. Teamwork was 1, work rate was 2. Flair was 3. Passing was 7 but every other skill attribute was in the red. His creativity was 20.

This player also had some youth stats generated and they were excellent but then he hit the OHL , which was playable, and didn't do anything for 2 years. He had 8 points in his second season and an avg. rating of 5.8. Like I said, this is an extreme example but this general pattern repeats itself in game. This player not only had no business being a top prospect he had no business being an NHL prospect at all, in fact I'd say he had no business even sniffing a major junior roster but his out of whack creativity attribute seems to have given him an ability rating he didn't deserve and that's the problem, one obviously out of whack attribute shouldn't turn a scrub into a top prospect.

Share this post


Link to post
Share on other sites
3 hours ago, Didaro said:

What else other than CA/PA can a scout use to evaluate a player? In order to make the game interesting and replayable there have be some modifiers in place but the recommendations have to be based on CA and PA. Game performance is clearly not used in scout recommendations for veteran players and can't be used for prospects playing in non-playable leagues because in their case there aren't any games. That leaves CA and also PA for young players.

Sounds interesting, but I have no idea if you're right; I believe there are other factors that could be considered (to be honest I suspect you are wrong as I don't think it's as simple as just looking at CA/PA)

 

 

 

3 hours ago, Didaro said:

Tran was talking about performance for much of the thread but the first thing Tran mentioned in his first post was physical attributes. He then talked about the other attributes being in his opinion unbalanced and that causing top prospects to not be NHL ready. His comments on performance were based on his observation that top prospects do not have the attributes necessary to step in and play well in the NHL right away

I don't believe what Tran was referring to was at all the norm regarding Attributes (that was hyperbole) based on the testing I've done http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=14804 

What he expressed I thought was connected to the Player Role system http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=16777 (which I thought he either didn't know about and/or didn't understand); I'm not saying things are working perfectly with the Player Role system! But I am saying that I thought the initial posts exhibited a lack of understanding and drew incorrect conclusions

I offered to help, but I'm not interested in the hyperbole/rudeness and when it continued I chose not to (#1 I've got LOTS on the go regarding EHM and #2 and maybe most importantly, I think I've already covered the concerns (see below regarding Player Role testing) and #3 I lost interest in helping/looking at the specific issue of top draftees further when the response was repeatedly so negative)

I know for a fact there are "issues" as I've pointed them out (along with testing results to support my views, questions to clarify and suggestions for fixes) for Forwards http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=16776 and for defensemen http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=16775 and for goaltenders http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=16774

 

 

 

3 hours ago, Didaro said:

CA/PA numerical values aren't visible in game but the star ratings given for current and potential ability in scout reports are based on them, aren't they?

No, I don't think they're based on CA/PA, I think they're based on your scout's recommendation/view (and as mentioned, I don't think that's a view that simply looks at just CA/PA); if what you're saying is true than all similarly rated Scouts would agree and Scouts would never change their views, but this is not the case. I myself think scouting "needs improvement" but I like data based proof before I draw conclusions and make suggestions on fixes, and I've not done any real testing on scouting.....but I do know that Scouts can have the same top Judging Ability/Potential Attributes and significantly different CAs!

I've always been under the impression/understanding that the stars ratings can be wrong from mildly to severely (like scouts in real life), and I thought other factors influenced things including the Scouts Attributes (including Adaptability/Determination) & how many times the player was scouted/how the player actually performed in the game(s) that were scouted 

I don't think the game does a good job of creating good/realistic Staff (the unrealistic Attribute distribution is actually far more evident among Staff than it is among Players IMO), and this summer I've become a strong advocate of an idea I've had called the underdatabase to attempt to help improve things http://www.ehmtheblueline.com/forums/viewforum.php?f=110

 

 

 

What you're referring to with your example I've seen too, but I didn't at all draw the same conclusions as you did.....the issues you mention I think connects to needed further tweaking of the Player Role system & the issues regarding players simply not good enough for the League they're in I've also seen in my testing on regens when looking at the data "by League" http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=16787&p=214831#p214831 (but neither issue connects directly to CA or PA)

Edited by Nino-33
fixed typos/syntax errors

Share this post


Link to post
Share on other sites
On 7/31/2016 at 14:41, Didaro said:

 

He was rated 3 star current ability and 5 star future ability and all of them said he could step in and contribute immediately and would become a key player. All of this players physical attributes were in the red; acceleration and speed of 4, stamina was 4, agility was 5, strength was 3. Teamwork was 1, work rate was 2. Flair was 3. Passing was 7 but every other skill attribute was in the red. His creativity was 20.

This player also had some youth stats generated and they were excellent but then he hit the OHL , which was playable, and didn't do anything for 2 years. He had 8 points in his second season and an avg. rating of 5.8. Like I said, this is an extreme example but this general pattern repeats itself in game. This player not only had no business being a top prospect he had no business being an NHL prospect at all, in fact I'd say he had no business even sniffing a major junior roster but his out of whack creativity attribute seems to have given him an ability rating he didn't deserve and that's the problem, one obviously out of whack attribute shouldn't turn a scrub into a top prospect.

The biggest "problem" (IMHO) is that the scouting staff only reports on the maximum level they think any player can reach but they don't say anything about the likelihood of the player ever reaching that level. That is why we get "...thinks he may become a key player" for those that are incapable of lacing their skates by themselves as well as for the Crosby types out there.

There is a lot of information to glean from the scouting reports by paying close attention to how things are said and what's not said but the star system is not reliable at all. I'd love if the scouts were to tell us their confidence level of their predictions, like; "...believes he has a good chance to become a key player" , "...thinks he might become a key player" , "...I'll eat my hat if he doesn't become a key player" , "...is confident he will become a key player" etc. etc.

Share this post


Link to post
Share on other sites
5 hours ago, IvanIvanovich said:

The biggest "problem" (IMHO) is that the scouting staff only reports on the maximum level they think any player can reach but they don't say anything about the likelihood of the player ever reaching that level. 

Wow. you're awesome Ivan. You lectured me on the scouting reports and tried to teach me about the game but the only problem is you don't know what you're talking about. I said the kid was a #4 overall...that means it wasn't a few scouts. he was rated. You can go away now.

Share this post


Link to post
Share on other sites
8 hours ago, Didaro said:

Wow. you're awesome Ivan. You lectured me on the scouting reports and tried to teach me about the game but the only problem is you don't know what you're talking about. I said the kid was a #4 overall...that means it wasn't a few scouts. he was rated. You can go away now.

This is a public forum and not your living room. You don't have to like my posts, or even read them, but whether I post or not is a decision that you are not in any way involved in.

Share this post


Link to post
Share on other sites

It's clear that fans of the game are too defensive to accept criticism of the game.

Meanwhile, there's a major hole in the way the game functions.

Whether Riz and company want to believe it or not, there is a long history of hype and fame for #1 overall picks (and other high picks) who step right into the NHL and make an impact.

That is missing from this game.

Share this post


Link to post
Share on other sites

Its great how people are still surprised when others fail to pay attention to their tone on the internet, and then get all self-righteous. I think both sides of the principal argument have been guilty of tonal insensitivity and then become terribly upset at getting the same in return. Why get so worked up when it is very difficult to communicate tone in a text based vacuum. Benefit of the doubt and all that.

On the issue in question, I think the latest development model is pretty good, but is perhaps a little too unpredictable. From my tests I've seen players with similar PA and CA develop at wildly different rates despite similar hidden ratings and both being left in junior. Also, much as in real life, some players take the NHL like ducks to water and others fail hard...the problem is a lack of adequate information as to why this may happen.

The scouting reports aren't too bad if you look at the content, but perhaps there needs to be more information about character and thus likelihood of reaching potential and ability to step up to a higher level. Ideally, longer term this could maybe be via some pre-draft combine kind of model, where a GM could choose certain prospects to find out more about, but be very limited in number, It would certainly add a level of extra interest to the draft.

Picking up on the opening post, I think there are some issues about insufficient drafts with players with CA levels even in the Mcdavid/Eichel ball-park in the figures given above. Of course PA could still let these players bust, but it would be great to have a few more 'next one'  style massive consensus picks that are just well ahead of the pack. Maybe not at that level, but real life has more 1st or 2nd picks that are well ahead of their peers than the game does. Perhaps this is an issue with development in junior in the game, as I'd agree that there are enough players with the correct PA?

It would also be nice to find away to have more obvious high risk/high reward picks, that your scouts could communicate could be first a liner or could never make the NHL depending on physical development etc. I guess the limitations of the PA/CA system make that harder to program in?

What I would say to Tran is the development isn't that bad but depends on many factors. Some players on the game that get rushed to the NHL develop remarkably quickly, and others stagnate. Having the right role on a reasonably successful team is a big part of it - a key thing which he/she may have missed is that some players can perform at a higher level than their attributes would suggest, and if they do so can develop very quickly. I have seen players jump by 25 CA in a season if put in the right position, so that a borderline 4th liner (judging by their attributes) can become a bona-fida 2nd liner in a year if things go well. I personally find a little power play time and starting on the 2nd/3rd or 4th lines with good line-mates can make a difference.

 

Share this post


Link to post
Share on other sites
26 minutes ago, lomekian said:

Why get so worked up when it is very difficult to communicate tone in a text based vacuum. Benefit of the doubt and all that.

Because I put so much time into it (literally thousands of hours of database editing/testing) and I'm tired of the arrogant/ignorant/rude responses (and I have a right to speak up about it as far as I'm concerned); I have a history that I believe has earned me more respect than I'm sometimes given by some, and I believe strongly I deserve more than the "benefit of the doubt" based on my history (and Tran has a history too that has earned him the responses he gets)

Maybe you have no idea who I am/what I've done and aren't giving me the "benefit of the doubt" I've earned

 

 

 

30 minutes ago, lomekian said:

From my tests I've seen...

I'd love to see your test data compilations, do you have any links to where you've posted the data/results?

I've done hundreds of hours of testing yearly for years, and post the data compiled into tables/charts + provide links to spreadsheets (I used to do so mostly in just the TBL Researchers Forum, but since this summer I've been posting all my results openly/publicly).......so I believe I've supplied significant "proof" & I did so because I thought without it how could anyone tell the difference between actual testing and anecdotal evidence/opinion

 

Share this post


Link to post
Share on other sites

Ok Nino -

Firstly - put the guns down! I know exactly who you are, have seen your input on more forums on this and appreciate your work. I'm certainly not attempting to belittle your efforts in any way, but do you have to be so aggressive? My comment was not just aimed at you, despite your decision to take it as such. I have no idea who Tran is, but short of being a bit blunt and perhaps imprecise, I didn't see anything too bad, bar being too easily offended and a little frustration bluntness. Perhaps Tran has a major history of being a keyboard warrior, but if so, I haven't encountered it, despite being a forum trawler.

I can appreciate how tiring and frustrating it must be to deal with people who haven't spent as much time looking into this as you, but then that will apply to 99.9% of all users of the game, so I'm not sure what exactly you expect. Almost everyone is going to be more ignorant on the subject than you, so dealing with it is inevitable. Your work has 'earned' you the benefit of the doubt of me listening to your input, taking it on face value, and appreciating that you know more than me about this. Because I know who you are. If I didn't I wouldn't!

Crucially, I didn't at any point refute anything you said, and merely posted a musing of my observations in my own way. At no point did i set myself up as a rival to you, so I don't know why you felt the need to wave your credentials around, as if this were some competition. And is there anything I said that is fundamentally incorrect anyway? CA for individuals on the game do sometimes develop at wildly different rates despite similar hidden stats for a reason I don't understand, but I didn't say this was a bad thing, just that the reasoning could be clearer or communicated in subtle ways.

Indeed my post, should you choose to re-read it, was largely in agreement with most of what you had written, and actually was in part an explanation to Tran to why I think he/she might be wrong.

So, in order to leave this post with a degree of purpose and constructive output, do you think the comparative lack of generated players entering the draft with Eichel/Mcdavid or not to far behind levels of CA is a problem with balancing with the TBL database to avoid too many superstars at once, or an issue with sub 18 player development in the game? Or has my experience to date of not seeing 18 year olds with CA's in the 120-150 range been an anomalous one? If so, feel free to say so and I'll leave the subject. If not, are either of my conjectures the reasoning in your view?

And did any of my other ideas for future development of draft scouting have any merit in your eyes?

 

 

Share this post


Link to post
Share on other sites
2 hours ago, lomekian said:

I can appreciate how tiring and frustrating it must be to deal with people who haven't spent as much time looking into this as you, but then that will apply to 99.9% of all users of the game, so I'm not sure what exactly you expect. Almost everyone is going to be more ignorant on the subject than you, so dealing with it is inevitable.

I expect/wish for the ignorance (lack of knowledge) to NOT be accompanied by the arrogance/rudeness/lack of appreciation, especially when real/actual help is offered!.....there's only a small % of people that are like this, and I'm just not going to take it anymore

My very first words in this thread said "I've got three "regen tests" going right now, two are in 2025 (heading towards 2035/2045) but I haven't compiled the data yet. When I do I'll look for examples of support"  &  then when I faced more arrogance/rudeness (and none of my questions were answered/no reference was made to the data in EHM and IRL that didn't support the claims) I still said "Like I said, I'll try to look for support/evidence of your concerns"  &  then (when no reasonable debate was occurring/abuse was still occurring) I said "For myself, I've got lots on the go and have decided given your argumentative responses/lack of appreciation for actual IRL data and testing EHM (and the many hours it takes to compile testing results) I'm not going to even bother looking into the issue"

I offered to help more than once before giving up! As I noted above "I offered to help, but I'm not interested in the hyperbole/rudeness and when it continued I chose not to (#1 I've got LOTS on the go regarding EHM and #2 and maybe most importantly, I think I've already covered the concerns (see below regarding Player Role testing) and #3 I lost interest in helping/looking at the specific issue of top draftees further when the response was repeatedly so negative)"

 

 

 

 

2 hours ago, lomekian said:

Firstly - put the guns down! ..........At no point did i set myself up as a rival to you, so I don't know why you felt the need to wave your credentials around, as if this were some competition.

As far as I'm concerned no data means you're not "testing" you're at best looking into (so if I'm testing and posting data/links that have taken weeks to do and you come along and say "from my testing" aren't you saying you're doing the same thing as me when you're not?)

I didn't think in any way it was a competition/I honestly though maybe you didn't know the efforts I've put in (for years) to help the EHM Community

And I'm tired of being told I'm the problem or have a problem because I eventually react to negativity/lack of appreciation.....if others have the right to be rude, and others have the right to tell me I'm doing something untoward, don't I have the right to post my own thoughts/feelings like others? I think I do

 

 

 

 

2 hours ago, lomekian said:

CA for individuals on the game do sometimes develop at wildly different rates despite similar hidden stats for a reason I don't understand, but I didn't say this was a bad thing, just that the reasoning could be clearer or communicated in subtle ways.

I don't "know" this from testing results or talking to Riz, but I suspect what you're seeing is intentional in an attempt to simulate real life where players don't all develop at the same rate and I honestly don't know how well it's communicated ingame or not.....while very interesting to me, the validity/correctness/reasonableness testing of ingame text is by far the toughest/most time consuming type of testing I can imagine doing [as you'd have to "connect" actual data (that the EHM Assistant thankfully provides!) with screenshots or notes taken regarding what is said] so alas I don't think I'll be able to add much of value on the reasoning anytime soon if ever

Maybe someday if I ever get to look at Scouting in further detail (with testing results of course! HaHa) I'll be able to look learn more at the ingame text in light of actual data, but realistically unless I get hired by SI HaHa I suspect that won't happen (I'm off in July/August as I'm a Teacher Assistant, so soon enough I'll have real life taking up much of the time I'm currently spending testing/editing)

I am doing Attribute Development testing again right now (like I did last summer http://www.ehmtheblueline.com/forums/viewtopic.php?f=110&t=14804&start=25) but it takes time to compile the data (2-3 weeks of daily work).....but it's coming/I expect it'll be done sometime this month! so maybe we'll get some more insight from the testing results (or maybe/hopefully others will review the data and see some connections/share some insights! as I've posted links to the actual spreadsheets/data from last year & I'll be doing the same for this year's testing)

 

 

 

 

2 hours ago, lomekian said:

do you think the comparative lack of generated players entering the draft with Eichel/Mcdavid or not to far behind levels of CA is a problem with balancing with the TBL database to avoid too many superstars at once, or an issue with sub 18 player development in the game? Or has my experience to date of not seeing 18 year olds with CA's in the 120-150 range been an anomalous one? If so, feel free to say so and I'll leave the subject. If not, are either of my conjectures the reasoning in your view?

And did any of my other ideas for future development of draft scouting have any merit in your eyes?

Regarding Scouting I did already say above "I myself think scouting "needs improvement" but I like data based proof before I draw conclusions and make suggestions on fixes, and I've not done any real testing on scouting"

And my thoughts regarding top end draft quality are the same...I don't know for sure because I haven't looked into it, but I do think improvements can be made (if Tran hadn't been so rude I was going to, but I have LOTS on the go now, and things take time, and while important the issue connects to a handful of players yearly so the "bang for the buck" isn't worth dropping what I'm working on now, which is #1 regen testing #2 Attribute Development testing #3 improving they Staff in the 1998 database for use in the next TBL Challenge) - I did already say above "IMO it's likely that EHM needs to improve in it's ability to generate players (the handful at best per year this applies to) that can play in the NHL at 18-19"

To be honest I'm hoping to use the testing I have to look into the actual data regarding NHL Drafts (I'm also curious about the rankings and how "correct" they are), but it takes days/weeks to do things (so it can be months before answers can be obtained)..... just like it takes many years to create a retro database (and I'd like to return to work on publicly know about and unknown about retro projects too!.....someday HaHa)

So though I'd like to, I can't directly answer your questions at this time [as I said earlier in this thread "I myself like to do extensive testing and obtain real "proof" that results in a complete/thorough understanding of an issue before pushing for change/improvement (for myself I want to have solutions I'm pretty sure of, as opposed to just complaining)]"

 

 

 

 

I believe EHM is more than reasonably priced, and is still being worked on/updated, and Riz/SI have never lied about the prioritizing it gets (it's alas is not high on the list HaHa), and so that means things take time.....but IMO clearly "things are happening" (and I think people like Riz and myself and those at TBL are all doing "the best we can" - it's a niche game, so not everything can be figured out and explained and/or fixed as fast as some may like [for retro databases I've been asking (begging! HaHa) for the ability to change the # of teams in a league/change rules for a long time, so I really do understand it can be frustrating waiting for improvements HaHa]

 

Regarding TBL and the TBL Rosters I already noted that it's an ongoing process of learning for the researchers/editors too (below)

On ‎7‎/‎30‎/‎2016 at 14:21, Nino-33 said:

Actually there is a big difference already in how some players are rated, and changes are ongoing as researcher understanding increases.....the NHL CA average is about 138 in the TBL Rosters, and Major Junior players used to be given much lower CAs in EHM07 because they almost immediately/immediately progressed to NHL caliber in EHM07 - but now development is different (slower/more realistic)  the Top draftees are given a CA that's much higher than they ever were with EHM07 (McDavid's CA in TBL 8.2 is 142 and Eichel's is 138, by far the highest starting CA for top draftees ever.......looking at the top 5 picks from TBL 7.0 their CAs were 96-96-98-99-90 & looking at the top 5 picks from TBL 6.1 their CAs were 118-115-88-121-95)

And it's really important to remember that researchers were still learning LOTS about how the game's changed & an enormous amount more has been learned since the ratings being discussed were released! The 2015-16 TBL rosters were "start of season" rosters and the CA/PA and Attributes haven't been edited since last November! So I'm quite sure you'll see improvement in the 2016-17 rosters due to researchers having gained more knowledge/understanding since the game came out

Improvements made by the TBL Roster team are ongoing like the improvements/tweaks Riz makes (and the regen testing and Attribute Development testing I'm doing currently is IMO more helpful "overall/at this time" than other things that also need looking at)

 

 

Nino :)

 

Edited by Nino-33
fixed typos

Share this post


Link to post
Share on other sites
1 hour ago, Nino-33 said:

I expect/wish for the ignorance (lack of knowledge) to NOT be accompanied by the arrogance/rudeness/lack of appreciation, especially when real/actual help is offered!.....there's only a small % of people that are like this, and I'm just not going to take it anymore

Ok - I guess we just have differing views on what counts as notable rudeness on an internet forum. Maybe I'm just dulled to it after so many years, so I don't notice it! At least no one started talking about Hitler... ;-)

 

 

As far as I'm concerned no data means you're not "testing" you're at best looking into (so if I'm testing and posting data/links that have taken weeks to do and you come along and say "from my testing" aren't you saying you're doing the same thing as me when you're not?)

Subtle difference - I said 'from my tests' , which I had no intention of being interpreted as a serious long term testing exercise. Otherwise I would have offered firm data as you suggest. Perhaps another term would have less scope for misinterpretation, but given the inherent ambiguity of English as a language...(one thing the Germans got right I guess!)

 

I don't "know" this from testing results or talking to Riz, but I suspect what you're seeing is intentional in an attempt to simulate real life where players don't all develop at the same rate and I honestly don't know how well it's communicated ingame or not.....while very interesting to me, the validity/correctness/reasonableness testing of ingame text is by far the toughest/most time consuming type of testing I can imagine doing [as you'd have to "connect" actual data (that the EHM Assistant thankfully provides!) with screenshots or notes taken regarding what is said] so alas I don't think I'll be able to add much of value on the reasoning anytime soon if ever

Maybe someday if I ever get to look at Scouting in further detail (with testing results of course! HaHa) I'll be able to look learn more at the ingame text in light of actual data, but realistically unless I get hired by SI HaHa I suspect that won't happen (I'm off in July/August as I'm a Teacher Assistant, so soon enough I'll have real life taking up much of the time I'm currently spending testing/editing)

 

My conclusion on this is exactly the same as yours (partly from having read your exhaustive stuff elsewhere). I don't think for a second its unintentional given the general robustness of the process. What is a problem however, is that despite Ivan's sound observations, whatever is determining this differentiation appears to be very difficult to identify, judging by the uncertainty on the subject on all the forums I have stumbled across. As I said, I have absolutely no problem with the model, and I'm sure things like coaching, quality of competition, ice time etc must factor, in addition to hidden attributes (EHM Assistant again!), but much of it is communicated very ambiguously or not at all. I don't know if your tests would corroborate this, but I get the impression that scoring totals and development in the games junior leagues is a little off. Its still a lot better than earlier iterations of the game where college prospects would often stall horribly compared to CHL or European prospects, who would develop too quickly.

 

Regarding Scouting I did already say above "I myself think scouting "needs improvement" but I like data based proof before I draw conclusions and make suggestions on fixes, and I've not done any real testing on scouting"

And my thoughts regarding top end draft quality are the same...I don't know for sure because I haven't looked into it, but I do think improvements can be made (if Tran hadn't been so rude I was going to, but I have LOTS on the go now, and things take time, and while important the issue connects to a handful of players yearly so the "bang for the buck" isn't worth dropping what I'm working on now, which is #1 regen testing #2 Attribute Development testing #3 improving they Staff in the 1998 database for use in the next TBL Challenge) - I did already say above "IMO it's likely that EHM needs to improve in it's ability to generate players (the handful at best per year this applies to) that can play in the NHL at 18-19"

To be honest I'm hoping to use the testing I have to look into the actual data regarding NHL Drafts (I'm also curious about the rankings and how "correct" they are), but it takes days/weeks to do things (so it can be months before answers can be obtained)..... just like it takes many years to create a retro database (and I'd like to return to work on publicly know about and unknown about retro projects too!.....someday HaHa)

So though I'd like to, I can't directly answer your questions at this time [as I said earlier in this thread "I myself like to do extensive testing and obtain real "proof" that results in a complete/thorough understanding of an issue before pushing for change/improvement (for myself I want to have solutions I'm pretty sure of, as opposed to just complaining)]"


Sure, Fair enough all round. I was just wondering if you had noticed any definite patters with this, because in my much more limited way its been pretty consistent that your statement in red is true. I guess until some of the uncertainty about the factors that affect prospect development can be clarified, theorising about methods that can communicate these in a more intuitive and immersive way is very much long-term thinking. I do think, however that some sort of 'draft combine'-like extended scouting element that reveals more about hidden attributes in a similar narrative style to that which already exists would be a comparatively simple addition to the existing model, as essentially it would be just further report generation, where numeral ranges would generate a small range of stock written word reports.

 

I believe EHM is more than reasonably priced, and is still being worked on/updated, and Riz/SI have never lied about the prioritizing it gets (it's alas is not high on the list HaHa), and so that means things take time.....but IMO clearly "things are happening" (and I think people like Riz and myself and those at TBL are all doing "the best we can" - it's a niche game, so not everything can be figured out and explained and/or fixed as fast as some may like [for retro databases I've been asking (begging! HaHa) for the ability to change the # of teams in a league/change rules for a long time, so I really do understand it can be frustrating waiting for improvements HaHa]

I couldn't agree more. I have absolutely no complaints about the efforts being made. Hell I bought both previous iterations more than once after the demise of various old PCs. I'm not particularly frustrated about the progress, and indeed, some of it over the last 12 months has been so noticeable as to put its larger more profitable stablemate to shame, where underlying issues are often left unresolved in favour of cosmetic or similar upgrades. RIz deserves a medal and TBL and yourself take a lot of credit. Behind CM1-FM16, CIVs 1-5, The Edler Scrolls games and maybe the first maxis and side meier games 25+ years ago, this has been my biggest time-sink.

Frankly, I think the vast majority of users would pay a higher price. Certainly all the returning users would.

1 hour ago, Nino-33 said:

 

Regarding TBL and the TBL Rosters I already noted that it's an ongoing process of learning for the researchers/editors too (below)

Improvements made by the TBL Roster team are ongoing like the improvements/tweaks Riz makes (and the regen testing and Attribute Development testing I'm doing currently is IMO more helpful "overall/at this time" than other things that also need looking at)

Nino :)

 

Can't argue with that. Thanks for taking the time to respond in full.

 

I've bolded my repsonses, since the SI forums quoting mechanism doesn't appear to be working in the same way as before.....

Share this post


Link to post
Share on other sites
9 hours ago, Nino-33 said:
11 hours ago, lomekian said:

I can appreciate how tiring and frustrating it must be to deal with people who haven't spent as much time looking into this as you, but then that will apply to 99.9% of all users of the game, so I'm not sure what exactly you expect. Almost everyone is going to be more ignorant on the subject than you, so dealing with it is inevitable.

I expect/wish for the ignorance (lack of knowledge) to NOT be accompanied by the arrogance/rudeness/lack of appreciation, especially when real/actual help is offered!.....there's only a small % of people that are like this, and I'm just not going to take it anymore

7 hours ago, lomekian said:

Ok - I guess we just have differing views on what counts as notable rudeness on an internet forum. Maybe I'm just dulled to it after so many years, so I don't notice it! At least no one started talking about Hitler... ;-)

 

I openly admit my standard is NOT that of an "internet forum" in general

I don't own a cell phone (never have), don't have a Facebook account (never have), I don't "do" social media at all (never have, can't imagine I ever will). My "internet forum presence" is limited to EHM related forums (here, at TBL, at HF Forums and reddit)

So given this, for me the standard for "seems unreasonable/problematic" starts well above talking about Hitler HaHa

I've always tried to answer/help people regarding EHM the way I would want to be answered/helped & I avoid the hyperbole, I don't post things as "facts" that are wrong (and are easy to check), I don't make "authoritative statements" (that are actually wrong/incorrect) that attributes knowledge/expertise that I don't have (I use phrases like "it seems like" and "it appears to be" as I'm gathering data/knowledge and formulating an educated/fact based opinion & if I don't "know" something I openly state so).....and I try to engage in meaningful/helpful dialogue/debate as I both first take the time to actually ensure what I'm posting is correct and actually supports my views & when I respond back to respond to what others say/what they post to support their reviews

That "benefit of the doubt" you mentioned initially - because I'm essentially "EHM exclusive" online, I figure I should have long ago earned that benefit of the doubt for myself, because my knowledge regarding EHM and my earnestness to help are very real (as is my years of actually producing things helpful to the EHM Community)

 

 

Now.....I'll get back to editing/testing EHM! :)

Edited by Nino-33
corrected poor wording/typos

Share this post


Link to post
Share on other sites
10 hours ago, Nino-33 said:

Maybe someday if I ever get to look at Scouting in further detail (with testing results of course! HaHa) I'll be able to look learn more at the ingame text in light of actual data, but realistically unless I get hired by SI HaHa I suspect that won't happen (I'm off in July/August as I'm a Teacher Assistant, so soon enough I'll have real life taking up much of the time I'm currently spending testing/editing)

Nino :)

 

My experience (from playing the game - not running tests ;-) ) is that EHM is working very well in this area.

If a prospect looks weak in (let's say) skating when looking at the attributes screen but your scouts repeatedly tell you that it's not a problem (good/above average/very good skater) then that player will improve rapidly in that area. If, on the other hand, your scouts point out skating as a problem area then it doesn't seem to matter how much you have the guy practice, he'll always be 2nd in a race to the puck.

IMHO the scouting reports are very good at imparting information about the player. The only part that can seem random is the future projection (stars system) but that is probably close to real life as only half of first round drafts ever make it to the NHL and they were likely all rated 5 stars by their teams on draft day.

A way for the scouts to let us know if they're only guessing when claiming a prospect is key material or if they're sure about it would be a nice improvement.

Share this post


Link to post
Share on other sites

My own thoughts of scouting could use some improvement are based wholly on the numbers (like CA/Attributes) and not at all based on text ingame (which I've never looked at in any way in my testing); I do know from testing that my concerns were validated but "how important/how much they affect things" is an opinion at this point.....I'm not going to go into details (some are mentioned in my underdatabase thread at TBL) but the knowledge I think has helped us researchers/editors to be able to better create/edit Scouts (and other Staff)

I've seen reports of Scouts changing their views on a player immediately after they player's drafted - have you noticed this Ivan?

 

 

42 minutes ago, IvanIvanovich said:

If a prospect looks weak in (let's say) skating when looking at the attributes screen but your scouts repeatedly tell you that it's not a problem (good/above average/very good skater) then that player will improve rapidly in that area. If, on the other hand, your scouts point out skating as a problem area then it doesn't seem to matter how much you have the guy practice, he'll always be 2nd in a race to the puck.

Thanks for posting, as this kind of thing I find helpful! It's easier to look into things that you're not totally sure about if you're trying to validate something rather than "hunting for the unknown in the dark" HaHa (but as mentioned, I don't see being able to look into the text anytime soon if at all)

The thing about skating Attributes and their development is it is NOT the same at all as it was in EHM07! In EHM07 only 1 of the 4 Attributes ever developed, while in the new EHM they all develop! And I suspect (but most certainly don't "know") that it might connect as much or more to age as it does to CA/PA

The other thing is some Attributes (including some skating Attributes) are Non-Essential/Irrelevant Attributes for a few Player Roles, and so this too will affect the development (or lack of) in those cases

That's why I do the Attribute Development testing I do, as a key starting point for editing (which requires some understanding obviously! HaHa) is which Attributes develop and which don't and how much do they develop (some that develop don't develop much at all, while others develop "normally") + it gives a baseline for development

Have you noticed a high number of players having ratings of 18-20 in their skating (particularly Speed/Acceleration)? This too has been reported and noticed by CJ at TBL as well

Edited by Nino-33

Share this post


Link to post
Share on other sites

I've seen it mentioned here and on reddit that scouts change their reports after the draft but it's not clear to me exactly what it is that is being said. - Are the reports changed without the player being scouted anew or is this a new scouting report coming in? 

If it is the second (a new scouting report) then I think I know what is happening. Any scouting report made during off-season will be very different compared with one where the scout watches the player in a game. This I think is fair, there are things that show up in in-game situations that cannot be "seen" outside of game situations.

From my experience, most if not all, scouting reports during the off-season (or for players in un-scoutable leagues like college or minor Europeans) are significantly downplaying the player (or maybe just not reporting things they cannot see).

I have repeatedly have players on tryout and when scouting them then they (may) get high grades but if I scout them after the tryout is finished they will get very much less praise. This I also think is fair, having the player on your team should give you a very much more accurate impression of him. I do think though that the scout should remember his previous assessment (if it is the same scout doing both reports), having the same scout give two reports with only a few days in between shouldn't give different results just because the player moved to a different team. 

Where the player happens to be when scouted definitely has an impact on the report.

About skating attributes. - I'm partial to good skaters (and playmakers) so I cannot really say if there are more or less of them. All my prospects need to be good skaters or they won't be my prospects. I guess that that means that I don't have anything to compare with and cannot really comment.

I'm guessing that since speed is currently "in fashion" IRL then this will favor speedy forwards but I still see "decent" skaters often, especially defenders. Enforcers and others that usually were the ones that were bad skaters don't really figure in today's (top level) hockey anymore. This could just be a reflection of reality but, as I said, I'm not really qualified to do anything but guessing here.

PS. As a side note - There is a situation that may be related. If you have your (NHL) coaching staff give a report on one of your AHL-players while he is on the AHL-team and then immediately promote him and ask your coaches once again you will get two totally different reports.

Share this post


Link to post
Share on other sites
17 hours ago, IvanIvanovich said:

I've seen it mentioned here and on reddit that scouts change their reports after the draft but it's not clear to me exactly what it is that is being said. - Are the reports changed without the player being scouted anew or is this a new scouting report coming in?

Here http://hfboards.hockeysfuture.com/showpost.php?p=121124331&postcount=126 it's reported as occurring on the same day, before the draft/after the draft

Share this post


Link to post
Share on other sites
4 hours ago, Nino-33 said:

Here http://hfboards.hockeysfuture.com/showpost.php?p=121124331&postcount=126 it's reported as occurring on the same day, before the draft/after the draft

It is still not clear if they are looking at a particular scout that has actually scouted the player or if they are looking at the head scout who aggregates all scouts opinions or if they are just opening the scouting tab and it happens to default to a different scout than it did last time. Too many unknowns to draw any conclusions.

It is also not clear if they are talking about any player or if they are talking about a player they have drafted. If it is a player they have drafted then there is a change in circumstance that could be significant - the player is now in a prospect in your system. (See my point about coaching assessments between AHL/NHL in previous post.)

Besides, the star system, is an assessment of probable potential and we have no way of knowing the confidence level of the scout making that assessment. This is what I requested in an earlier post, a way for the scout to tell us how sure he is of his assessment  - If he is just guessing or if he is betting his life on it.

The star system is, IMHO, the weakest part of the scouting system and should be taken with a whole truckload of salt. It's much more important to look at the details of what they actually say about the player.

Share this post


Link to post
Share on other sites

Thanks for the feedback Ivan! For me, every little bit helps to try and get a solid understanding (either to ask better questions of posters when they post about issues and/or to better able myself to look into things (trying to minimize my "hunting for the unknown in the dark"). I've got a wall chart up now (put up tonight!) so I don't forgot what to look into/keep my eye out for related posts & I can add notes to what is likely and/or known.....and hopefully when passing it on to Riz, it can be done in the most clear/concise helpful way for him and maybe even have some suggestions/thoughts on possible methods of improvement (hopefully reducing somewhat his required time when looking into things, thereby increasing what he can get to!)

Share this post


Link to post
Share on other sites
On ‎8‎/‎5‎/‎2016 at 00:25, IvanIvanovich said:

It is still not clear if they are looking at a particular scout that has actually scouted the player or if they are looking at the head scout who aggregates all scouts opinions or if they are just opening the scouting tab and it happens to default to a different scout than it did last time. Too many unknowns to draw any conclusions.

It is also not clear if they are talking about any player or if they are talking about a player they have drafted. If it is a player they have drafted then there is a change in circumstance that could be significant - the player is now in a prospect in your system. (See my point about coaching assessments between AHL/NHL in previous post.)

I followed up on the HF EHM Forum and got some clarification...

 

 

Nino33 - I'm hoping you can clarify a few things regarding the issue you're noting...
Are looking at a particular scout that has actually scouted the player or are you looking at the head scout who aggregates all scouts opinions? or are you just opening the scouting tab and it happens to maybe default to a different scout than it did last time?
Also, are you talking about any player or are you talking about a player that you have drafted

 

Cap'n Flavour - I always have at least 5 scouts with 19+ JPT/JPA scouting draftees, and some others permanently assigned to junior leagues.

During the draft, I put up the scout report view and sort by stars. I then look at scout reports for players I consider picking. Usually they're pretty consistent between scouts but sometimes they'll all say the same player as a comparable but half will be "a better version" and the other half will be "poor man's".

Anyway, I draft players, exit the draft, look at the scout reports for the players I just drafted, and the player projections all change - usually to point to an inferior player. The first player I took this year went from "like <guy who won the Norris with a 68 PT season>" to "like <mediocre depth D with 6.7 avg. rating>". Sometimes the projected career role changes to
o. Not all players change but around half do and it's usually the top picks.

 

 

Lebowski - For me it affects the actual scouting reports, so obviously the scout actually scouted the player. I don't think it's because the game defaults to one of my scouts, because I tend to look at every scouting reports from each individual scout and the same happens for everyone of them. I'm not sure about what happens to other teams as I don't really look into their players after the draft, but that's something I'll look for in the future.

For example, during the draft, when it's your turn to pick, they can suggest you to draft player X. You then go see player X's scouting reports, and from the available scouting reports (meaning, from scouts that actually scouted the player), you can sometimes have players descriptions like "Like Jonathan Toews, franchise forward". So you pick that player, thinking your scouts actually feel like he has franchise potential. Right after you've picked said player, you go back to his scouting recommendations, and 9 times out of 10, the suggested players will have changed to something like "Like Travis Zajac, 2nd line upside", and more often than not, that updated player projection is more reliable, even if you go in the scouting history of a player and he's been recommended as an impact player/excellent prospect all year long.

 

BraveCanadian - I think it has something to do with the fog of war?
When they are scouting they don't have all the info but once they are on your team they have more?

Lebowski - I'm thinking it probably has to do with something along those lines, but that doesn't really make sense in that context, because you haven't actually taken the time to scout the player while he was on your team. These changes to the player projections happen before I even exit the draft, meaning I haven't had the chance to affect a scout to my own team.

 

 

IMO the scouting reports shouldn't change the same day just because the player is now been drafted and is "on your team" - doesn't seem logical/realistic to me.....it'd make sense to me if any changes occurred after Training Camp though (as long as the player participated in Training Camp) as you'd actually see/scout the new draftee amid professional players

Share this post


Link to post
Share on other sites

So it's a player that has been drafted by their teams?

I agree that a scout shouldn't change their assessment of future ability on the same day but does the game know that it is the same day? It is quite possible that all the game can see is that the circumstances have changed.

Look at it this way (possible scenario): There is a subroutine that does the assessment that gets called every time a scout (or coach) report is opened - the details regarding abilities don't change but the future potential gets done anew every time you open the report (i.e. it is dynamically generated) where circumstances have changed. The effect of this would be exactly what we are seeing here. 

A player that is undrafted and a player that is drafted and inside your organisation are NOT the same thing. A player that is playing in the AHL and one that is playing in the NHL is NOT the same thing (with regards to the coaching reports that I mentioned in previous post). That the two reports are opened on the same day may be unknown to the program. It is quite possible that the subroutine keeps no record of when that scout last accessed it (it wouldn't make sense to code that into it - a new assessment is a new assessment and when the last one was made is irrelevant).

The reason I believe that the assessment of  "potential" is dynamically generated is because that assessment will change during the season for your own players regardless if you scout them or not. If you look at a scouting report of (for example) an aging veteran this "potential" will deteriorate during his last couple of seasons regardless if you scout him or not.

This means that the part of the scouting report that has to do with "season projection", "career role", "Player comparison" and the stars "current and future" are not "created and fixed" when a scout watches the player but is instead generated new every time you open the report (or whenever there is a change).

If those parts of the scouting report were "static and fixed" by a scout watching the player then it would be impossible for the game to change that during the season (or over the season change or yearly change or with career threatening injuries), this would lead to your own players all having "old" assessments unless you continuously scouted and re-scouted them.

If the choice is between having a slightly odd behaviour when scouting the same player twice in the same day under different circumstances and having to scout all your own players all the time as to avoid ending up with reports that aren't representative, then I will prefer the first alternative.

If a player is undrafted then an assessment of him should be less accurate than if he is drafted by you and is one of your unsigned prospects. I hope you agree with this. A player you have drafted will be "better known" to your scouting staff than a player on someone else team. It is unfortunate though, that it is not communicated that such "changes in circumstances" will cause changes in how the scouts view them even if there is no new scouting assignment between the changes.

I am assuming that no one told you that they had seen this behaviour on players that they didn't draft or players where such circumstance changes didn't take place. If they did my reasoning is obviously incorrect.

PS. The same behaviour can be seen when you trade for a player. You can get one assessment before the trade goes through and another, different one, after the player has joined your team. I think this is also OK. and a lot less "problematic" to understand.

Edited by IvanIvanovich
typo

Share this post


Link to post
Share on other sites
3 hours ago, Nino-33 said:

 

IMO the scouting reports shouldn't change the same day just because the player is now been drafted and is "on your team" - doesn't seem logical/realistic to me.....it'd make sense to me if any changes occurred after Training Camp though (as long as the player participated in Training Camp) as you'd actually see/scout the new draftee amid professional players

I (respectfully) disagree with this.

I believe that Toronto Maple Leafs have a much better grasp of the qualities of Auston Mathews than they would have had had he been drafted by some other team. This regardless if he has been signed or attended a training camp or not. The fact that he is part of their organization is reason enough to believe that they have more knowledge of him than other teams do.

Share this post


Link to post
Share on other sites
5 hours ago, IvanIvanovich said:

I (respectfully) disagree with this.

I believe that Toronto Maple Leafs have a much better grasp of the qualities of Auston Mathews than they would have had had he been drafted by some other team. This regardless if he has been signed or attended a training camp or not. The fact that he is part of their organization is reason enough to believe that they have more knowledge of him than other teams do.

Sorry, but that makes no sense to me at all

My suggestion was just a way of having there be some reason/logic behind it.....what exactly do you think the Leafs would have learned in the couple/few months between draft day and Training Camp? What EHM Attributes did they gain a better understanding/knowledge of in the offseason, before he's ever been part of the team/in the dressing room/on the ice?

 

 

And as to your prior post, IMO you're minimizing the amount of information/data that teams actually have and the thoroughness of scouting.

Also NHL/AHL Staff are IRL all part of the same organization; the NHL AGM is often the GM of the AHL team and the Scouts are often "organizational" and not divided between NHL and AHL (which can make database editing a challenge sometimes!)

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...