02/16/2011

• Adjusted assist odds for a team to create more separation between good passing teams and poor passing teams. This does not change the impact of passing on FG%.
• Added an individual player matchup component to rebounding, so that a mismatch will produce better numbers from the player with the advantage. This will in turn create a little more separation between the truly good rebounding teams and poor rebounding teams.
• Adjusted the distribution of players available for the different divisions/prestiges to better match the distribution prior to the 5/12/2010 release. In particular there were too many recruits slotted for the DII level and not enough at the lower levels of DI. 
  
  Note: This does not change the actual ratings of the players, it just affects what schools a particular quality of player will consider and so affects also what recruits will show up in the search for a school.
• Added a small bump to the average max value for most ratings for recruits. This will only affect recruits generated after today's release.
• Broke up the Scoreboard page by tournament (e.g. DI NT, DI PI, etc.) for postseason games. Also added a note at the bottom of game capsules to indicate the type of game.

any thoughts?

the first two changes sounds good to me (assists/reb). 

will have to think about the distribution of players... my gut is that this will prove to be a good thing. i never liked the impact of the 5/12/10 release on d2 recruiting - letting d2 teams go for an extra couple hundred ranked players, which forced more altercations with d1 schools. i wonder how this will impact d3 - will some of those super studs, fit to lead d2 teams (or some d1 teams even), disappear from d3?

small bump to average ratings... for MOST recruits. i like that. hopefully the most includes the crappy d1 recruits and not the great ones.

i noticed the last bit, the type of game on the office page, i thought it was out of place... i was going... man, has that always been there? did i really drink that much last night? but luckily, it is new :) on the down side, my headache is quite real.

Interesting.

I don't understand why he added a bump to the average max value for all recruits. He was trying to address the same issue as before, and he needed to add it for only a certain cross section of recruits in order to achieve the goal.

The rebounding thing is interesting, I always felt this aspect should be part of the engine. The question is simply if it was implemented correctly, and only time will tell.

I don't think that distribution of players is a good thing at all. All this means is that D2 teams can pull down less players, and the fact that D2 teams were pulling these guys down to begin with means D1 schools weren't going after them anyway. So instead of going to D2 schools they will either go to sim D1s or JUCO.

Posted by coach_billyg on 2/16/2011 12:55:00 PM (view original):
the first two changes sounds good to me (assists/reb). 

will have to think about the distribution of players... my gut is that this will prove to be a good thing. i never liked the impact of the 5/12/10 release on d2 recruiting - letting d2 teams go for an extra couple hundred ranked players, which forced more altercations with d1 schools. i wonder how this will impact d3 - will some of those super studs, fit to lead d2 teams (or some d1 teams even), disappear from d3?

small bump to average ratings... for MOST recruits. i like that. hopefully the most includes the crappy d1 recruits and not the great ones.

i noticed the last bit, the type of game on the office page, i thought it was out of place... i was going... man, has that always been there? did i really drink that much last night? but luckily, it is new :) on the down side, my headache is quite real.

billyg, it doesn't say a small bump for most recruits, it says a small bump for most ratings.

Posted by kmasonbx on 2/16/2011 12:59:00 PM (view original):
I don't think that distribution of players is a good thing at all. All this means is that D2 teams can pull down less players, and the fact that D2 teams were pulling these guys down to begin with means D1 schools weren't going after them anyway. So instead of going to D2 schools they will either go to sim D1s or JUCO.

I disagree. I think DII teams have long been able to pull down players that should've been over their head. Right now low DI is pretty barren, which is the only reason most of these players are available. But it was resulting in DII teams having as much or more talent as DI teams. I like the change.

hmmm, interesting girt. i must have been reading what i expected to see. seble, if you are reading this... for your small bump in max ratings - is that even across the board? like if max per went up 2, does that mean the best and worst and everybody in between is expecting to go up 2 per? or is this like your first change, where you also adjusted variance, so that low d1 players are going to catch up to high d1 players a bit?

Posted by kmasonbx on 2/16/2011 12:59:00 PM (view original):
I don't think that distribution of players is a good thing at all. All this means is that D2 teams can pull down less players, and the fact that D2 teams were pulling these guys down to begin with means D1 schools weren't going after them anyway. So instead of going to D2 schools they will either go to sim D1s or JUCO.

Well, I think that's a good thing, not a bad thing.  Low-level Simmy D1s SHOULD be filled with these kind of guys...not walk-ons, as is the case with many of them right now.

Posted by girt25 on 2/16/2011 12:59:00 PM (view original):
Interesting.

I don't understand why he added a bump to the average max value for all recruits. He was trying to address the same issue as before, and he needed to add it for only a certain cross section of recruits in order to achieve the goal.

The rebounding thing is interesting, I always felt this aspect should be part of the engine. The question is simply if it was implemented correctly, and only time will tell.

I beleive the old engine focused more on idividual matchups. This has been one of my peeves with the current engine. Hopefully we will see the teams with superior Ath and REB actually getting mroe rebounds now.

Posted by kmasonbx on 2/16/2011 12:59:00 PM (view original):
I don't think that distribution of players is a good thing at all. All this means is that D2 teams can pull down less players, and the fact that D2 teams were pulling these guys down to begin with means D1 schools weren't going after them anyway. So instead of going to D2 schools they will either go to sim D1s or JUCO.

Agree 100%...This just might drive me out of HD personally...

Posted by gillispie on 2/16/2011 1:04:00 PM (view original):
hmmm, interesting girt. i must have been reading what i expected to see. seble, if you are reading this... for your small bump in max ratings - is that even across the board? like if max per went up 2, does that mean the best and worst and everybody in between is expecting to go up 2 per? or is this like your first change, where you also adjusted variance, so that low d1 players are going to catch up to high d1 players a bit?

Any change to recruit ratings is going to be across the board.  It may affect certain levels more than others, but recruits are not generated by division, they are created as a large pool all at once. 

Posted by mullycj on 2/16/2011 1:24:00 PM (view original):

Posted by girt25 on 2/16/2011 12:59:00 PM (view original):
Interesting.

I don't understand why he added a bump to the average max value for all recruits. He was trying to address the same issue as before, and he needed to add it for only a certain cross section of recruits in order to achieve the goal.

The rebounding thing is interesting, I always felt this aspect should be part of the engine. The question is simply if it was implemented correctly, and only time will tell.

I beleive the old engine focused more on idividual matchups. This has been one of my peeves with the current engine. Hopefully we will see the teams with superior Ath and REB actually getting mroe rebounds now.

Actually the opposite is true.  The old engine had very little individual player matchup logic in it and had no individual matchup logic for rebounding.  I originally had that individual matchup adjustment in the engine but then removed it during testing.  So I'm just adding it back in there after seeing that rebound margins were not quite what they should be.

Posted by kmasonbx on 2/16/2011 12:59:00 PM (view original):
I don't think that distribution of players is a good thing at all. All this means is that D2 teams can pull down less players, and the fact that D2 teams were pulling these guys down to begin with means D1 schools weren't going after them anyway. So instead of going to D2 schools they will either go to sim D1s or JUCO.

What was happening is that DII teams had a huge pool of players to recruit, more than they should have had.  So a lot of the top DII teams were taking recruits that should have gone to the mid-to-lower level DI teams.  That was making DI recruiting pretty challenging because those schools had to compete with the top DII teams in addition to the normal competition.  There were also too few recruits for the lesser DIII teams.

Adjusted assist odds for a team to create more separation between good passing teams and poor passing teams. This does not change the impact of passing on FG%.

Can you explain that a little bit more for the simpletons among us?

If it does not impact FG%, what (if anything) does it actually impact? Turnovers perhaps?

Will it impact different positions differently, or will it have the same impact on equally-rated players regardless of whether they are playing PG or PF?

The change on 2/3 raised the assist rate on made shots.  What it didn't do was create enough of a difference between good passing teams and bad passing teams.  So bad passing teams were getting too many assists and good passing teams not enough.  This change will not affect the average, but will separate the good passing teams from poor.  This is still purely a cosmetic change, meaning it doesn't affect turnovers, FG% or anything else. 

The change on 2/3 also affected distribution of assists to players including a position factor.  This release hasn't changed that logic at all.