12/17/2012 10:19 AM
Interesting study.  Thanks for doing this.

My current conclusion, is that like a lot of things at D3, good coaching can overcome a glaring weakness and a team can be competitive and even make the NT with a big shortfall in a key area.  At the same time, while competitive on the whole, the teams with the glaring weakness will get exposed against the truly good and very good teams who game plan.
Currently at D3, there are enough SIM teams combined with teams coached by humans who either aren't very good or don't care to gameplan in regular season that an active team by a seasoned coach should win a fair number of games.

I've seen this across the board for weaknesses.  Teams with awful defense or awful ath, for instance, will finish with 20+ wins, a top 30 RPI but then come NT time, they never make a run and looking at individual game results, get blown out by good teams.

This also happens at the individual level.   It is what causes some of the value of ind. ratings debates.  A player with 20 ath and 70 spd, but 85+ per, CAN avg. 20+ points effeciently by feasting on the good matchups.  But that same player will get completely shutdown when it plays a very good team who game plans.

The same is happening here.  The team is wining the games against SIM teams and being competitive and winning against some human teams, but if it played top teams on a regular basis, it the results would show that the games are not nearly as close as they should be for a team of its abilities.

The takeaway is this:  IQ (and for that matter Ath/Def) can be overcome against the majority of the cannon fodder that permeates this game.   At that level, good coaching and good overall numbers matters more than any individual trait.  BUT, once you get to the elite level (i.e., teams consistently going to Sweet 16+), every rating must be perfectly fine tuned or you will lose and lose badly more often than not.  So, there is a bit of a cliff dividing the elite from everyone else.  The difference between good teams and elite teams don't look like very much on paper (just a few ath points here, some defense there, higher IQs) but make a ton of difference at that level.  Those teams are just more consistent regardless of the opponent.

This all makes sense when you understand that this game is just one large probability equation.  Any individual factor merely affects the probability odds for every calculation.  Thus, a team weak in a key factor, but very strong elsewhere, will remain competitive because the probability factor of maximizing the ratings is lower, but the ratings still carry more often than not.  

A player with an IQ of F will still make the right play 60% of the time (or thereabouts considering that even 1 practice period at 10 or so minutes moves an IQ from F to D-).  But the computer doesn't view it as "right" or "wrong" play they way it happens in the real world.  The computer instead says, how close to best result will this player get.  60% of the time, the player will maximize its ratings, and 40% of the time, it will be something less (but not zero).  The player with A rating, will maximize his ratings 90% of the time.  That said, if the player with IQ of F is sufficiently better than player with IQ of A, even if F is only functioning at reduced capacity, it will still often be better than player A.


12/17/2012 10:38 AM
Very well said.
12/18/2012 12:10 AM
Posted by andrew5975 on 12/16/2012 11:57:00 PM (view original):
hey robot...so another season in the books. how would you say your season played out compared to what you expected going in? how about predictions for next season? and how would your predictions for next season be different if all your players had A/B IQs going into the season?

I feel a team with this much talent would have easily made the Sweet 16 this season if i were putting 20min into OFF/DEF practice, so yes at this point i would say it means something. How much? I still don't know. It is difficult to take out all of the other factors that can contribute to winning and losing to pinpoint an exact value but i'm hoping maybe after this next season we can start to quantify it.

*The crazy thing about this team though is that there is still upside in them. I got pretty lucky on some of the high/highs. Still a good amount of black and blue on my practice page and I think this team will finish next season in the 630s for overall ratings. If i were putting in 20min of IQ practice with the skills they are going to have i would be dissapointed if we weren't playing in the NT championship and taking the trophy home. We'll see how far they get.

12/31/2012 5:40 PM
Anyone care to make a guess as to what the value of IQ is at this point? Just finished up the non-conference schedule. Maybe 10pts a game difference if most of my guys were in the A range?
12/31/2012 6:03 PM
robot...your teams looks to be far superior in talent over everybody you played...even the teams you lost against...so im interested to read what the experienced HD coaches say as far a "value" for IQ goes???
12/31/2012 6:54 PM

A question to consider when thinking about that potential 10 point increase is whether or not your ratings would be nearly as high if you were at A in IQ's. If you have to give up significant ratings points, 10 points per game will be negated and it would even out.

Obviously the experiment is not over, but I would wager that a team that could achieve ratings this high and manage even a C+ or B- in IQ (which might take 5-10 minutes per season if you wanted this IQ by junior year) would dominate. I have my doubts that the difference between B IQ and A IQ are significant enough point wise to take away minutes from creating the high skill ratings.

Robot, what are your next steps this experiment once this class is about complete? I know you have a couple more seasons in this phase. Would you consider in the next phase recruiting another solid class and only practicing 5 minutes of O/D (or at least something less than 10). In that scenario you can still achieve high ratings and gain a respectable IQ level. 

12/31/2012 9:50 PM
Posted by poncho0091 on 12/31/2012 6:54:00 PM (view original):

A question to consider when thinking about that potential 10 point increase is whether or not your ratings would be nearly as high if you were at A in IQ's. If you have to give up significant ratings points, 10 points per game will be negated and it would even out.

Obviously the experiment is not over, but I would wager that a team that could achieve ratings this high and manage even a C+ or B- in IQ (which might take 5-10 minutes per season if you wanted this IQ by junior year) would dominate. I have my doubts that the difference between B IQ and A IQ are significant enough point wise to take away minutes from creating the high skill ratings.

Robot, what are your next steps this experiment once this class is about complete? I know you have a couple more seasons in this phase. Would you consider in the next phase recruiting another solid class and only practicing 5 minutes of O/D (or at least something less than 10). In that scenario you can still achieve high ratings and gain a respectable IQ level. 

I'm not quite sure yet where to go with it after this. I'd like to play out one more season after this one with no IQ practice to really get a good idea of how it is affecting the team. Maybe go the other extreme with 30+ minutes in team practice and just enough in individual skills to keep them growing. Or maybe like you said put 7 or 8 minutes in each and see what happens. What do you think?     
1/1/2013 5:41 PM
I think finishing this class with no IQ is the best way to truly test your current experiment, but like I said, to try the next phase of the experiment, you'll have to pull in another really good class. Personally, I'd like to see how you do with less than 10 minutes with elite recruits. I know the general consensus for most coaches is between 15-25 minutes. 15-20 will get you usually in that A range by their senior year, so like I said 7-8 should get you in the C range by year one, and maybe B+ by year 4.

Since your team is already so much more dominant than most in ratings and we all feel fairly confident at this point that IQ is possibly holding your team back, it would be nice to see what your team can do when your team is still elite in the ratings, but has enough IQ to not be elite, but also not retarded. Personally, I think with the team you have with B IQ, you could go far.

1/2/2013 8:29 PM
Thanks for doing this experiment. Very insightful.  Curious...were these players you recruited D2 drops/pulldowns, D3 only, or combination of? Did they all have high potential areas? What was the method to your recruiting?  Thanks! 
1/2/2013 10:45 PM
Yes, the majority but not all were D2 pulldowns. The key is to recruit players not for what they are now, but for what they will be as Jrs and Srs. You want as much upside as you can get in the primary skills for each position.
1/28/2013 1:53 PM
i forgot about this experiment...taking into account your strong season but early NT exit...any further thoughts on IQ robot? any changes to your experiment going forward?
2/24/2013 6:01 PM
it looks like robot has no HD teams, has he given up on this study?
4/22/2013 8:38 PM
This, in response to my new thread...
1/29/2014 3:36 AM
seems kinda unfinished...wish there was one more summary post from OP
1/29/2014 3:41 AM
I just finished the recap moments ago, and I am itching for the same thing.
of 12

Terms of Use Customer Support Privacy Statement

Popular on WhatIfSports site: Baseball Simulation | College Basketball Game | College Football Game | Online Baseball Game | Hockey Simulation | NFL Picks | College Football Picks | Sports Games

© 1999-2014 WhatIfSports.com, Inc. All rights reserved. WhatIfSports is a trademark of WhatIfSports.com, Inc. SimLeague, SimMatchup and iSimNow are trademarks or registered trademarks of Electronic Arts, Inc. Used under license. The names of actual companies and products mentioned herein may be the trademarks of their respective owners.