Page 3 of 6
Re: Working, Real NCAA Model
Posted: Wed Mar 08, 2017 2:37 pm
by Alessandro
BKarchitect wrote:Alessandro wrote:I also noticed all games are at "unknown arena".
How does the host is picked IRL?
The Frozen Four is bid on by cities - generally ones with an NHL infrastructure (every host since 2006 has been an NHL city) although more regional cities with close proximity to NCAA programs like Milwaukee, Providence, Orlando and Omaha have bid and in the case of the former two, have hosted in the past. In modern times it has not been hosted by NCAA university home arenas. The infrastructure to sell more tickets and have the hotel and host facilities is generally not there in most campus towns unless they are an urban setting like Columbus or Providence.
Ah, ok, I thought it was something like the Memorial Cup.
I'll try and see what can be done.
Re: Working, Real NCAA Model
Posted: Thu Mar 09, 2017 12:03 am
by Primis
So, I played through a season (not quick sim, actually played through one season as Michigan State in the Big Ten). And I have to say, it all works very well. Boston U beat Minnesota in the Frozen Four finals (Northern Michigan and Robert Morris were the other 2.... Denver and Quinnipiac didn't make it out of the Round Robin).
The North American signing method works very well. Guys interested in Major Junior tend to ignore my overtures, and guys in say in the USHL who want to play collegiate hockey are very interested and will usually sign/commit. There are no actual salaries, whatever black magic it is that BKarchitect used to get that to work. It's rather brilliant -- no budget, no cap, no anything financial to worry about. All you're essentially doing is getting a commitment from guys, and no salaries to pay for players or coaches.
The one-day trade window I set resulted in only 2 guys being put on the block through the entire season, and zero trades. I'd call that a success for eliminating trading. Not that trades couldn't still happen, but they sure look super unlikely.
I did run into an issue where I had 23-year-olds that couldn't play the last part of the season because I didn't set the age limit to 24, but had it on 23. That's my own fault. Man, Rhett Holland got mad when I couldn't play him anymore...
Great job with this though, BKarchitect. This is what I could never quite manage in 2007, trying to replicate NCAA. And it sure is fun to then use the ability to create exhibitions to create things like the Great Lakes Invitational tourney, or my own made-up on between the Michigan-based NCAA and OHL teams. I haven't played this much EHM is quite some time.
I'm super-excited about my recruiting class for next season though. Looks like they come on board in August.
By the way: I think the "random-seeming" commits happen because they're manually put into the EXTRA_CONFIG file under FUTURE TRANSFERS and processed when the game is loaded. I noticed there are some 2018 recruit commits in there for quite a few teams.
Re: Working, Real NCAA Model
Posted: Thu Mar 09, 2017 12:53 am
by BKarchitect
I put in a ton of actual commits into the config

...there are some 2020 commits even for really young top prospects who have committed in real life. You can tell the difference between the CONFIG commits I added and the "random CPU commits" when you look at future transactions. The real CONFIG ones are listed as "CTR" while the ones the CPU adds throughout the progress of the sim say "free". Eventually once the CONFIG file runs its course, all the CPU-determined commits simply say "free".
One other thing I did in the config...I made a lot of the top drafted prospects stay the entire 2016-17 season (instead of signing immediately in the sim in the fall of 2016) by giving them future transfers to their NHL team after the 2017 season (or later in a few cases where the prospect is less "hyped"). This way Clayton Keller and Tyson Jost and Colin White all play at least the 2016-17 in the NCAA like real life and then sign pro. Obviously if a guy like Jost decides to spend an extra year in school that will have to be updated but it made sense to me in order to create a realistic 2016-17 NCAA season.
Great work on the "non" trade/transaction system! Sounds like that'll work pretty darn well. That's really super exciting!
Did you take a look at my Frozen Four issue on the preceding page? Have you noticed this?
Re: Working, Real NCAA Model
Posted: Thu Mar 09, 2017 1:16 am
by Primis
BKarchitect wrote:Did you take a look at my Frozen Four issue on the preceding page? Have you noticed this?
No problems here: the 6 teams that made it all won their respective conference tourneys.
Re: Working, Real NCAA Model
Posted: Thu Mar 09, 2017 1:26 am
by BKarchitect
No it's the seeding of the final four...I have years where the results of the top 4 of the round robin aren't the four in the final playoff leg. Like the Gophers finished 3rd and Holy Cross finished 5th and somehow it was Holy Cross as the 4th seed in the final leg. I'm sure there is a setting wrong in selecting the top 4 to the second leg or something.
Re: Working, Real NCAA Model
Posted: Thu Mar 09, 2017 12:39 pm
by Primis
BKarchitect wrote:No it's the seeding of the final four...I have years where the results of the top 4 of the round robin aren't the four in the final playoff leg. Like the Gophers finished 3rd and Holy Cross finished 5th and somehow it was Holy Cross as the 4th seed in the final leg. I'm sure there is a setting wrong in selecting the top 4 to the second leg or something.
Ahh ok, I see.
Now looking back at it, the seeding took the top 4 teams out of the Round Robin just fine, in the correct seeding order.
Re: Working, Real NCAA Model
Posted: Mon Mar 13, 2017 5:38 pm
by BKarchitect
Hey Primis...
I tried the contract thing (2 window) and it seems to work very well but one thing I noticed is that the CPU generally will offer 3 year contracts as a standard...at least in my game they do. Which of course begs for a new flag that is "Minimum Contract Years" so we can set that at 4 just like the max contract years at 4. It's not a game-breaking fault but it's kind of annoying to see a guy sign somewhere for 3 years and then his 4th year he leaves and it still says he wants to play college hockey.
Other than that...it's pretty awesome...I do notice it is quite hard to recruit/sign USHL level players to more modest NCAA programs which is probably fine but even some of the lesser major conference teams...I find myself (more correctly, my scouts) trolling the depths of the AJHL, etc for bottom-level recruits. I switch to Minnesota or BC though and I can offer anybody I want

I'm sure it's based on program reputation in the database which of course we can edit. Just would be nice as a lower level program to be able to recruit a top college prospect at least once in a while. The other thing is if a player is too young they of course will not commit to anywhere...it would be cool if you could recruit years in advance like in real life but that is probably asking for too much out of this.
I also still have the bug with the Frozen Four not always selecting the top 4 teams in terms of points from the round-robin - but perhaps you changed something to make it work every time? Alessandro - have you noticed this issue at all? Mind looking at the Stage 1/2 settings for the "Frozen Four" league and seeing if there is an error?
Re: Working, Real NCAA Model
Posted: Mon Mar 13, 2017 6:59 pm
by Manimal
Has any work been done on team reputations? That would improve recruitment.
Re: Working, Real NCAA Model
Posted: Mon Mar 13, 2017 7:37 pm
by BKarchitect
Manimal wrote:Has any work been done on team reputations? That would improve recruitment.
I have in the test DB I uploaded - it probably just needs more testing and vetting over extended sims to determine the effects. I think I have the big programs right now at about a 5000 (your Minnesotas, BC, BU, North Dakotas, etc) and the weaker schools (Atlantic Hockey, some of the weaker WCHA/ECAC programs) at like 3000. The problem may actually be more that the lower leagues are not as fleshed out and if you use additional fake players at the start - maybe it is fine. I like to start with no fake players generated - the result is the pool of NAHL/Canadian Jr. A and high school/prep talent is not exceptionally large - certainly not large enough to feed into 60 NCAA schools. So maybe you need fake junior players on for best results.
Re: Working, Real NCAA Model
Posted: Mon Mar 13, 2017 9:28 pm
by Primis
BKarchitect wrote:I also still have the bug with the Frozen Four not always selecting the top 4 teams in terms of points from the round-robin - but perhaps you changed something to make it work every time? Alessandro - have you noticed this issue at all? Mind looking at the Stage 1/2 settings for the "Frozen Four" league and seeing if there is an error?
I've only done the one season all the way through, am about to come to a close with the second so I'll monitor the Frozen Four seeding and let you know, but my first season it all worked perfectly.
One of the things I want to do after I'm done fiddling with my Michigan State game is to try a smaller program like UA-Huntsville or something and see what recruiting is like there. Some tweaking of NCAA reps could be in order, to make the lesser programs still playable. It's probably a lot more difficult as a "smaller" program, and it should be, but not impossible. I also don't know what your starting GM reputation has in terms of influence, if any.
Re: Working, Real NCAA Model
Posted: Mon Mar 13, 2017 10:31 pm
by Alessandro
Write here what you'll discover with reputations. I'll set as BKarchitect say in the meantime.
Re: Working, Real NCAA Model
Posted: Tue Mar 14, 2017 1:02 am
by Primis
Sooooo, Season 2 ended, and in the Frozen Four round robin the #5 team made it into the final four over the #4. And it's not a matter of a tie-breaker -- the #4 team had 5 pts, and the #5 had only 4 pts. Yet the #5 team advanced.
In fact, the #4 team even beat the #5 team head to head.
I.... I have no idea or explanation. *shrug* Never seen anything quite like it.
I looked at the editor and the Frozen Four sort rules are set as:
1. Games Won
2. Goal Difference
3. Goals For
4. Own Match (What the #&*% does this mean anyways?)
5. Pts divided by possible pts.
Which is what all the other collegiate conferences are also set to.
Re: Working, Real NCAA Model
Posted: Tue Mar 14, 2017 6:18 am
by Lumpjaw
Primis wrote:Sooooo, Season 2 ended, and in the Frozen Four round robin the #5 team made it into the final four over the #4. And it's not a matter of a tie-breaker -- the #4 team had 5 pts, and the #5 had only 4 pts. Yet the #5 team advanced.
In fact, the #4 team even beat the #5 team head to head.
I.... I have no idea or explanation. *shrug* Never seen anything quite like it.
I looked at the editor and the Frozen Four sort rules are set as:
1. Games Won
2. Goal Difference
3. Goals For
4. Own Match (What the #&*% does this mean anyways?)
5. Pts divided by possible pts.
Which is what all the other collegiate conferences are also set to.
It shouldn´t, but is it possible the game goes after Games Won before points? As both 4th and 5th team had 1 win each. It shouldn´t do that, but maybe it does? With that said, it should not matter as the 4th placed team had better GD and GF.
Re: Working, Real NCAA Model
Posted: Tue Mar 14, 2017 8:15 am
by Alessandro
I'll have a look when I'll have some more time.
I'm now finalizing the DB before the release.
EDIT: I added even more college commitments
Re: Working, Real NCAA Model
Posted: Thu Mar 16, 2017 6:19 pm
by Primis
BKarchitect... I think I *maybe* figured out some of what's happening in the Frozen Four, finally.
3rd year my team finally made it, and I noticed that it scheduled me for 5 round-robin games, one against the other 5 teams each. However, after only the 3rd game, it set the bracket with the 4 semi-finalists. I'm assuming the round robin standings then get messed-up because the round robin games then still finish out. The #5 and #6 teams have 0 pts currently, but the #4 only has 3 pts and theoretically one of those bottom two teams could win out and overtake those 3 pts, but it wouldn't matter because the bracket of four has already been set.
EDIT: Basically... the game is setting the bracket after only 3 games and it needs to delay it until the full slate has been played.
Re: Working, Real NCAA Model
Posted: Thu Mar 16, 2017 6:30 pm
by Alessandro
Something wrong with dates?
Re: Working, Real NCAA Model
Posted: Thu Mar 16, 2017 9:36 pm
by Primis
Alessandro wrote:Something wrong with dates?
I'm cracking open the DB in the editor now. I just had time to post that before I needed to leave for a few hours earlier.
Taking a look... Stage 1 occurs from March 24 to April 4, and Stage 2 (the actual semifinals) runs from April 7 to April 9.
April 4 apparently does not allow enough time. By April 4th, in my current game I'll only be playing my 4th Round Robin game, with the final one on the 6th. And April 5th is the first semifinals bracket game too.
So it seems to be a dates issue. Stage 1 needs a longer runtime, to say another week later. And then Stage 2's dates be adjusted also to reflect the change.
Re: Working, Real NCAA Model
Posted: Thu Mar 16, 2017 9:39 pm
by Alessandro
Primis wrote:Alessandro wrote:Something wrong with dates?
I'm cracking open the DB in the editor now. I just had time to post that before I needed to leave for a few hours earlier.
Taking a look... Stage 1 occurs from March 24 to April 4, and Stage 2 (the actual semifinals) runs from April 7 to April 9.
April 4 apparently does not allow enough time. By April 4th, in my current game I'll only be playing my 4th Round Robin game, with the final one on the 6th. And April 5th is the first semifinals bracket game too.
So it seems to be a dates issue. Stage 1 needs a longer runtime, to say another week later. And then Stage 2's dates be adjusted also to reflect the change.
We have dates problems with other leagues too... sometimes I simply think that the game bites more that it can chew (or is the saying the other way around?)
Re: Working, Real NCAA Model
Posted: Thu Mar 16, 2017 9:42 pm
by BKarchitect
Primis wrote:Alessandro wrote:Something wrong with dates?
I'm cracking open the DB in the editor now. I just had time to post that before I needed to leave for a few hours earlier.
Taking a look... Stage 1 occurs from March 24 to April 4, and Stage 2 (the actual semifinals) runs from April 7 to April 9.
April 4 apparently does not allow enough time. By April 4th, in my current game I'll only be playing my 4th Round Robin game, with the final one on the 6th. And April 5th is the first semifinals bracket game too.
So it seems to be a dates issue. Stage 1 needs a longer runtime, to say another week later. And then Stage 2's dates be adjusted also to reflect the change.
That's got to be it...nice sleuthing, Sherlock! Hope that simple fix works...
Re: Working, Real NCAA Model
Posted: Thu Mar 16, 2017 9:45 pm
by Alessandro
BKarchitect wrote:Primis wrote:Alessandro wrote:Something wrong with dates?
I'm cracking open the DB in the editor now. I just had time to post that before I needed to leave for a few hours earlier.
Taking a look... Stage 1 occurs from March 24 to April 4, and Stage 2 (the actual semifinals) runs from April 7 to April 9.
April 4 apparently does not allow enough time. By April 4th, in my current game I'll only be playing my 4th Round Robin game, with the final one on the 6th. And April 5th is the first semifinals bracket game too.
So it seems to be a dates issue. Stage 1 needs a longer runtime, to say another week later. And then Stage 2's dates be adjusted also to reflect the change.
That's got to be it...nice sleuthing, Sherlock! Hope that simple fix works...
I'm not sure it will definitely work. It's a crapshoot, I fear. But at least NA leagues have more time (european season ends april 30), so it may have more chances.
Re: Working, Real NCAA Model
Posted: Fri Mar 17, 2017 7:40 pm
by Primis
So I'm now in July of 2019 in my game. And I know BKarchitect mentioned before about a few random players ended up on collegiate rosters. I have not had or experienced this problem.
Not until now.
And booooooy is it a doozy.
You can have a roster of up to 30 players on a team. I had a full 30 and I lost 8 of them to NHL signings (I won a national championship this past season, and a bunch of guys were ready to move on, either graduating or to the NHL). That left me with 22 players, and a class of maybe 8 or 9 guys coming in come August.
On the rollover from July 15th to July 16th, I suddenly had FORTY-SIX of these players show up on my roster all at once, with no explanation, making my roster SIXTY-EIGHT players. All of the mystery players have the "Ctr" icon next to their names. If you look under Team Transactions, it shows they are NOT listed under Players In or anything, yet they all have 4-year contracts. And they all seem to have pretty random birth places (multiple players hailing from Franklin, TN would be the big hint to that).
*shrug*
I have no ideas, short of a flag somewhere telling the game to auto-sign these guys for the upcoming season like it would normally if they weren't playable. I do know I have a whole lot guys to Release now...
Re: Working, Real NCAA Model
Posted: Mon Mar 20, 2017 12:09 am
by Named
Have You solved "Unknown Arena" bug for NCAA Frozen Four? I think i know how to fix it.
Re: Working, Real NCAA Model
Posted: Mon Mar 20, 2017 7:18 am
by Alessandro
Named wrote:Have You solved "Unknown Arena" bug for NCAA Frozen Four? I think i know how to fix it.
It's not really a bug, it's just that it happens IRL in a "random" arena.
Re: Working, Real NCAA Model
Posted: Tue Mar 21, 2017 1:12 pm
by Alessandro
We fixed the playoffs draw bug (max league games set as 3 - should have been 5, that's why playoff seeds were drawn after 3 games only).
We're ready to release NCAA now.
Re: Working, Real NCAA Model
Posted: Tue Mar 21, 2017 4:14 pm
by Javs
Alessandro wrote:We fixed the playoffs draw bug (max league games set as 3 - should have been 5, that's why playoff seeds were drawn after 3 games only).
We're ready to release NCAA now.
CIS next?