Page 5 of 6 FirstFirst 1 2 3 4 5 6 LastLast
Results 61 to 75 of 84

Thread: 12.03 / Into the future, overall ratings rise [Fixed BB2K11]

  1. #61
    Join Date
    Feb 2007
    Posts
    1,749

    Re: 12.03 / Into the future, overall ratings rise

    In statistics, distributions are described by several factors: mean and/or median, standard deviation, skew, and "normality", for example. The talent distributions in Mogul are so far off the mark I don't think one or two adjustments will suffice. For example, you could reduce the number of 90+ players but still have too many 80+ -- or vice versa.

    With so many adjustments, also, it would help a lot if you could have a profile to save all your favorite settings. Civilization had such a file, and since Mogul already has the Rookies.ini file it shouldn't be difficult to expand the concept.

  2. #62
    Join Date
    May 2008
    Location
    Republic of Georgia
    Posts
    12,804

    Re: 12.03 / Into the future, overall ratings rise

    Did some more playing last night...

    More evidence that the problem is with the initial set up of the game BEFORE you can even adjust the settings.


    *Players refers to position players

    In this first graph, I simmed 5 then 25 years with all the basic settings, changing nothing.

    The first graph (and data) is the talent distribution (and peaks distribution) when the game is first loaded (fictional players from 2009 database). It looks pretty nice.

    Simming 5 years to get the the second graph. Things are already horribly skewed. Look above the second graph. That is the increase in the number of players at each level of overall. The engine has added OVER 300 players with overall ratings of 80 or more (the last line should read 80+ not 85+). WTF?!

    To put that in context, that is 10 more 80 rated players PER TEAM.

    That's the biggest problem.

    The graphs of overall get flattened out. There are A LOT more 80+ players and a few more low end players. So instead of that nice spike of 70 rated players we have a big blob.

    Simming another 25 years in order to retire all of the originally created players and use only players created during the game the numbers are better. This shows that the INITIAL PLAYER CREATION is worse than the in game creation.

    There are now around 150 more 80+ players than from game start up but 150-200 less than 5 years in. Still way too many but not as bad. You can see the pitchers (yellow line) represent most of those players.


    I did another 30 year sim this time lowering the rookies.ini file so that Pitchers = 50 and All Positions = 75.



    The same things happens at first. The initial set up of the game creates 300 extra 80+ rated players. Even though the values for rookies was slashed. These players haven't had a chance to make any impact yet.

    25 years later. The average rating of players is way down (from 73 to 63 for position players and from 73 to 65-70 for pitchers) but. Look at the number of 85+ rated players. Almost unchanged from the initial set up of the game. So, while the rookies.ini is making these reduced players, somehow, the game engine is still over producing 85+ rated players.

    At set up there were 26 players rated 85 or higher. 30 years later, with rookies.ini at 75 and 50 there were 31.

    I manually checked each of these players (would have been much easier if I could sort by draft date) and confirmed that these were all players created by rookies.ini and NOT left over from the initial game set up.

    Something in the engine is telling the game to make too many 85+ rated players.

    I know some people complained that the engine didn't produce enough Hall of Fame players. The Ty Cobbs, Babe Ruths, etc that clearly stand out from their piers. So I wonder if this is an attempt to do that? Make extra super stars.

    Also, something is with pitchers. Even with pitchers at 50 in the rookies.ini and position players at 75, pitchers are still over represented in the 75+ and 80+ breakdowns. You'd expect pitchers to be worse, but they are better.

    Something is very wrong here.

    The two problems that I see:

    1) The initial set up of the game creates too many 80+ rated players
    2) The engine produces too many 85+ rated players

  3. #63
    Join Date
    Jul 2009
    Location
    Atlanta
    Posts
    1,213

    Re: 12.03 / Into the future, overall ratings rise

    So what would be the best default settings for a stable sim?
    “The bases are loaded, just like [Marlins manager] Jack McKeon probably wishes he was.”
    -Skip Caray
    RIP 8/12/1939- 8/3/2008

    http://forum.sportsmogul.com/showthr...ething-Special

  4. #64
    Join Date
    May 2008
    Location
    Republic of Georgia
    Posts
    12,804

    Re: 12.03 / Into the future, overall ratings rise

    Quote Originally Posted by Francoeurstein View Post
    So what would be the best default settings for a stable sim?
    That's my point.

    There are NO settings that will allow for a stable sim. The instability comes from the initial set up of the game over which the user has zero control.
    Well you could create a default player database where all the players had lower peak ratings. The fact that the game levels out "pretty well" after 30 years shows that the problem isn't with the engine as much as with the initial database.

    (Yes I realize there are too many 85+ players after 20-30 years, but the settings can adjust that -- the real problem is with the initial database).
    Last edited by Clay Dreslough; 01-04-2010 at 10:05 AM.

  5. #65
    Join Date
    Feb 2007
    Posts
    1,749

    Re: 12.03 / Into the future, overall ratings rise

    Yeah, I've been playing a 1950's dynasty. I originally simmed a 1975 game for 5 years and it looked like the ratings didn't go through the roof, so I decided to try the 1950's and it's taken 5 years instead of 2 but still the league average team rating is now 90 in 1955 (starting from 1950.) We need to be able to adjust player development rather than player creation; if we could reduce the number of players who actually hit their peak we wouldn't need to crank down the amateur draft talent.

  6. #66
    Join Date
    Jul 2009
    Location
    Atlanta
    Posts
    1,213

    Re: 12.03 / Into the future, overall ratings rise

    Quote Originally Posted by filihok View Post
    That's my point.

    There are NO settings that will allow for a stable sim. The instability comes from the initial set up of the game over which the user has zero control.

    The only way I can stand to play the game now is to start a game and sim 30 years into the future so all the players initially created by the game are retired.

    THEN your settings take effect.
    This makes the game literally unplayable for me. It makes me sick seeing guys with overall rating of 85-90 consistently riding the pine. Most of the time I will see a guy rated around 95 and he will perform just above league average... That is just ridiculous. This needs fixing.
    “The bases are loaded, just like [Marlins manager] Jack McKeon probably wishes he was.”
    -Skip Caray
    RIP 8/12/1939- 8/3/2008

    http://forum.sportsmogul.com/showthr...ething-Special

  7. #67
    Join Date
    Oct 2009
    Posts
    381

    Re: 12.03 / Into the future, overall ratings rise

    If you want a fix by force, you can try this:

    I played just one year and the average overall team score jumped from 77 to 81 after one year. Then I looked into each player on team rosters and found it was very very obvious to spot the players that were given an artificial "spike" of talent for no good reason (players that were 33 years old and increasing from 81 to 87 or players that were young and jumped 12 points in talent despite never having had a peak nearly that high). There were some increases that were more subtle as well. But there were at least a few examples of scrubs becoming superstars in one season (i.e., someone who had a overall/peak of approximately 63/65 their entire 8 year career and suddenly the next year they were an 88). So I just manually turned all those players down to where they should be. If you do this right after the World Series ends (and before the many extensions and free agent signings), it doesn't screw with any teams finances. It wasn't too hard. If you take a few weeks with each season, than taking 45 minutes to do this at the end of a season isn't too big a deal. If you do this for the first 5 seasons or so that should most likely completely take care of the massive bulge of early talent. Then your setting can take over.

    BTW what should those settings be so that the overall ratings in the future of teams are around 78 as they were on the beginning. Should I turn down Overall/Peak -20% or something? I can live with these manual turn downs for the first few years, but not infinitely into the future.

  8. #68
    Join Date
    Jul 2009
    Location
    Atlanta
    Posts
    1,213

    Re: 12.03 / Into the future, overall ratings rise

    Quote Originally Posted by Merlin401 View Post
    BTW what should those settings be so that the overall ratings in the future of teams are around 78 as they were on the beginning. Should I turn down Overall/Peak -20% or something? I can live with these manual turn downs for the first few years, but not infinitely into the future.
    Where is the overall/peak option?
    “The bases are loaded, just like [Marlins manager] Jack McKeon probably wishes he was.”
    -Skip Caray
    RIP 8/12/1939- 8/3/2008

    http://forum.sportsmogul.com/showthr...ething-Special

  9. #69
    Join Date
    May 2008
    Location
    Republic of Georgia
    Posts
    12,804

    Re: 12.03 / Into the future, overall ratings rise

    Quote Originally Posted by Merlin401 View Post
    If you want a fix by force, you can try this:

    I played just one year and the average overall team score jumped from 77 to 81 after one year. Then I looked into each player on team rosters and found it was very very obvious to spot the players that were given an artificial "spike" of talent for no good reason (players that were 33 years old and increasing from 81 to 87 or players that were young and jumped 12 points in talent despite never having had a peak nearly that high). There were some increases that were more subtle as well. But there were at least a few examples of scrubs becoming superstars in one season (i.e., someone who had a overall/peak of approximately 63/65 their entire 8 year career and suddenly the next year they were an 88). So I just manually turned all those players down to where they should be. If you do this right after the World Series ends (and before the many extensions and free agent signings), it doesn't screw with any teams finances. It wasn't too hard. If you take a few weeks with each season, than taking 45 minutes to do this at the end of a season isn't too big a deal. If you do this for the first 5 seasons or so that should most likely completely take care of the massive bulge of early talent. Then your setting can take over.

    That's interesting.

    Looking at this graph I made, you can see that in Year 0 the number of players with peaks 85+ is about 10 TIMES as many as players with overall ratings of 85+. That means that, at most, 1/10 of those players should reach their peak. I always assumed that too many players were reaching peak, as well as the current 85+ players not retiring fast enough.

  10. #70
    Join Date
    Oct 2009
    Posts
    381

    Re: 12.03 / Into the future, overall ratings rise

    Quote Originally Posted by filihok View Post
    That's interesting.

    Looking at this graph I made, you can see that in Year 0 the number of players with peaks 85+ is about 10 TIMES as many as players with overall ratings of 85+. That means that, at most, 1/10 of those players should reach their peak. I always assumed that too many players were reaching peak, as well as the current 85+ players not retiring fast enough.

    It seemed to me that most of the "old players" were declining as they should be. In this one year, Chipper went from a 90s player to a high 80s. Helton went from high 80s to 77. Magglio Ordonez went from 88 to 83 (to name a few examples). I think its a combination of random weird increases at the beginning and what you said. Maybe another thing you could do right at the beginning of a sim is randomly delete one player from each farm system that has a peak in the low 90s. This should make the amount of talent in the minor leagues realistic and keep the sim stable until you get to your settings a few years later.

  11. #71
    Join Date
    Oct 2007
    Posts
    326

    Re: 12.03 / Into the future, overall ratings rise

    Ummmm......wouldn't the easiest option be just to make new rosters where current prospects peaks are lower? I mean, there aren't THAT many of them.

  12. #72
    Join Date
    May 2008
    Location
    Republic of Georgia
    Posts
    12,804

    Re: 12.03 / Into the future, overall ratings rise

    I don't know that it's the easiest option, but the BEST option would be for Clay to fix the problem.

    If the user could change the settings BEFORE the game creates the initial players that should solve a big part of the problem.

  13. #73
    Join Date
    Mar 2007
    Posts
    30

    Re: 12.03 / Into the future, overall ratings rise

    I could deal with the higher average. The problem I have is the AI doesn't adjust to the new talent level.

  14. #74
    Join Date
    Oct 2009
    Posts
    381

    Re: 12.03 / Into the future, overall ratings rise

    Quote Originally Posted by rwperu34 View Post
    I could deal with the higher average. The problem I have is the AI doesn't adjust to the new talent level.
    I'm not sure how anyone could enjoy playing with team overall's in the 90s. If you have Albert Pujols on your team, he may put up Pujols type numbers. But if he has an off year, he'll now be an average outfielder. You need some sort of spread to distinguish good players from bad.

  15. #75
    Join Date
    Dec 2008
    Posts
    6,390

    Re: 12.03 / Into the future, overall ratings rise

    Well you talk about players overall ratings raise and how their performance isn't as good as their rating. The vitals are more important to a player's performance. I know this prob won't help the conversation so carry on.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •