10/23/2013 12:13 PM
seble did a good job on the projection report, so id hazard a guess that it almost certainly includes home/away in some way. i do think rpi is a factor in and of itself so theres that. i wonder if he included it in like, record vs top 100 calculations?
10/24/2013 8:44 AM
Posted by cburton23 on 10/23/2013 11:46:00 AM (view original):
RPI, if a factor at all, is very small.  There is always some major disparities between RPI and projection standing, and it seems like, though not as much anymore, there is always somebody ******** about their projection standing vs their RPI.  I know because I have done it.
by major disparities, you mean a team being off by around what, 20 spots? 40 spots? out of 300+ teams in D1 that's not a major disparity. Compared to other factors I'd be surprised if RPI's effect is "small".
10/24/2013 9:36 AM

Interesting idea.  A couple of thoughts:

  • Most teams in most worlds are sims.  In Allen there are 120 DIII humans (out of 384) and 140 DI humans (324 total).  Some of those humans aren't active.  I'd bet other worlds are similar.  So really the pool of competent players isn't all that high.
  • With only 100 or so "real teams" the disparity between RPI and projection can be significant and by that I mean being listed in the 70s and "out" on the projection report but having an RPI in the 20s.  Happened last season as referenced in a recent thread.
  • I do believe the projection report is a superior system to what was in the past (RPI).  Great job to wis for coming up with it.
  • Instead of bonus money, why not just make mandate at least 4-5 home games?  Program it so that you couldn't accept or offer road games if already at the max.  Four instead of 5 would give those that schedule the flexibility by not having to have exactly half and would accomplish a desired result.  No team schedules all OOC games on the road year after year.
  • Another idea might be to tweak the RPI formula.  1.4 for a road win is too high in DIII where HCA means nothing.  Winning at a DI sim is not the same as winning at UNC.  IN HD, Winning at Duke is much harder than winning at a good DIII program.
10/24/2013 10:57 AM

It's been mentioned but I want to reiterate now that the Projection Report decides seeding RPI just isn't a very big factor, so the hit to RPI doesn't really matter anymore. This would not have been a bad idea under the old system where RPI was about 80% of the determinant of your NT seeding. Now the RPI of your opponents is the most crucial aspect because you are looking for quality wins.

So if you play in a tough conference you can just schedule 10 sims at home get the extra money and still have it not impact your seeding too much because you'll have enough top 50/100 opponents in conference play.

10/24/2013 1:26 PM
Posted by kmasonbx1 on 10/24/2013 10:57:00 AM (view original):

It's been mentioned but I want to reiterate now that the Projection Report decides seeding RPI just isn't a very big factor, so the hit to RPI doesn't really matter anymore. This would not have been a bad idea under the old system where RPI was about 80% of the determinant of your NT seeding. Now the RPI of your opponents is the most crucial aspect because you are looking for quality wins.

So if you play in a tough conference you can just schedule 10 sims at home get the extra money and still have it not impact your seeding too much because you'll have enough top 50/100 opponents in conference play.

I pulled one example, Knight D2:

Projection Report Ranking has a 98.5% Correlation (R-value) with RPI Ranking. Obviously Correlation doesn't prove causation, and these are both pulling from the same datasets, but the odds that RPI has a "small" effect on projection report is very slim. Just because we anecdotally know of cases where RPI and Projection vary by 50 doesn't mean that much.

As a side note, the reason small schools with no shot at national tournament contention, schedule big schools is for a financial boost. Maybe this could be re-designed as a balancing force for low-D1 teams.
of 2

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.