Posted by rogelio on 9/14/2013 8:42:00 PM (view original):
I'll be glad to delete this, it's going to take up some space, but here's a quick idea on scheduling that may quickly address mmt/llamanunts concerns. The idea would be that the conferences would be ranked A-F (A the winner thru F - relegation entrant) and the A, B & C and D, E, F conferences would play one another 1-6, but the games between any of ABC & DEF would not be purely 1-6. Here's the chart [Rank is the sum of each of the seeds played, which in a straight 1 only plays other 1s schedule would be 5, 10, 15, 20, 25, 30): [Edited to even out the brackets and rankings]
A |
A |
B |
C |
D |
E |
F |
Rank |
1 |
|
1 |
1 |
2 |
3 |
4 |
11 |
2 |
|
2 |
2 |
1 |
5 |
3 |
13 |
3 |
|
3 |
3 |
6 |
1 |
2 |
15 |
4 |
|
4 |
4 |
5 |
6 |
1 |
20 |
5 |
|
5 |
5 |
4 |
2 |
6 |
22 |
6 |
|
6 |
6 |
3 |
4 |
5 |
24 |
B |
|
|
|
|
|
|
|
1 |
1 |
|
1 |
4 |
2 |
3 |
11 |
2 |
2 |
|
2 |
3 |
1 |
5 |
13 |
3 |
3 |
|
3 |
2 |
6 |
1 |
15 |
4 |
4 |
|
4 |
1 |
5 |
6 |
20 |
5 |
5 |
|
5 |
6 |
4 |
2 |
22 |
6 |
6 |
|
6 |
5 |
3 |
4 |
24 |
C |
|
|
|
|
|
|
|
1 |
1 |
1 |
|
3 |
4 |
2 |
11 |
2 |
2 |
2 |
|
5 |
3 |
1 |
13 |
3 |
3 |
3 |
|
1 |
2 |
6 |
15 |
4 |
4 |
4 |
|
6 |
1 |
5 |
20 |
5 |
5 |
5 |
|
2 |
6 |
4 |
22 |
6 |
6 |
6 |
|
4 |
5 |
3 |
24 |
D |
|
|
|
|
|
|
|
1 |
2 |
4 |
3 |
|
1 |
1 |
11 |
2 |
1 |
3 |
5 |
|
2 |
2 |
13 |
3 |
6 |
2 |
1 |
|
3 |
3 |
15 |
4 |
5 |
1 |
6 |
|
4 |
4 |
20 |
5 |
4 |
6 |
2 |
|
5 |
5 |
22 |
6 |
3 |
5 |
4 |
|
6 |
6 |
24 |
E |
|
|
|
|
|
|
|
1 |
3 |
2 |
4 |
1 |
|
1 |
11 |
2 |
5 |
1 |
3 |
2 |
|
2 |
13 |
3 |
1 |
6 |
2 |
3 |
|
3 |
15 |
4 |
6 |
5 |
1 |
4 |
|
4 |
20 |
5 |
2 |
4 |
6 |
5 |
|
5 |
22 |
6 |
4 |
3 |
5 |
6 |
|
6 |
24 |
F |
|
|
|
|
|
|
|
1 |
4 |
3 |
2 |
1 |
1 |
|
11 |
2 |
3 |
5 |
1 |
2 |
2 |
|
13 |
3 |
2 |
1 |
6 |
3 |
3 |
|
15 |
4 |
1 |
6 |
5 |
4 |
4 |
|
20 |
5 |
6 |
2 |
4 |
5 |
5 |
|
22 |
6 |
5 |
4 |
3 |
6 |
6 |
|
24 |
Ok, I understand the thinking here to a degree, especially (I guess specifically really) as it matters more as the seeding accuracy decreases. Since we won't ever really probably be able to determine a method that with any real degree of accuracy reflects the real overall strength of teams for the next season, perhaps something like this is necessary, but I don't like it. Here's my issues:
Assuming that the seedings are even close to accurate, (which, I concede, may not be possible, but I'm first looking at the issue in the ideal) what makes the #1 seeds any more or less special than every other seed? If we are anything near accurate in seeding, then in actuality the 1 seeds probably have more equal competition among the other 1 seeds then say the 4 seeds do due to the fact that some conferences only have the 6 or 7 coaches participating while others have 9 or 10. It stands to reason that the deeper conferences #4 seeds are likely to be tougher than the 6 or 7 team conferences' #4s or #5s, or whatever... this isn't always the case I know, and maybe my assumption on this is just off base, but that's one of my concerns.
The other is the system itself. As outlined by rogelio, under the current like seeds play each other method the total of the seeds that each team faces goes 5, 10, 15, 20, 25 and 30, while under his proposal it goes 11, 13, 15, 20, 22, 24. Since lower totals imply tougher match-ups, the #1 seeds get a boost from 5 to 11, while the #3 and #4 seeds get no boost at all, and the #5 and #6 seeds actually get tougher match-ups... The #1 seeds for this MUSDUC II are Dominguez Hills (started season ranked #1), NC Central (#21), Findlay (#2), Montevallo (#10), Paine (#4) and Central St (#9). All those teams must be pretty darned good. How are the #4 seeds supposed to stand a chance against that? So, while the #1 seeds get more likely to improve their MUSDUC win-loss records on average, the lower seeds have it tougher. My idea is that all the teams have roughly the same difficulty level.
That said, for a number of reasons (different numbers of participating teams per conference, the vagaries of seeding, etc.) it might not be realistic or possible to achieve my ideal. I just want to make sure that we don't make things easier for the 6 #1 seeds at the expense of 24 3-6 seeds...