Top Overall Seed in NCAA Tournament

[KnightBridgeAZ]

I'm one that doesn't think it matters (this year) who is the #1 over all seed. Because - if you had 2 #1 seeds that were undefeated, they ought to be able to beat the #3 and #4 over all seeds - whoever they are and who ought to be fairly equal in talent in any case.

And what about if the #3 over all seed loses and you get a #2 or #2 regional seed in the slot?

So I don't see it mattering. I still suspect - as it stands now - UConn is the over all #1 seed.

 

[semper]

I'm having a hard time seeing the problem...it used to be really important to me that the 1 seed play the 4 seed, etc., but I think it's going to be Notre Dame and UConn and no way they make us play in the semi-final game. So this year, really it doesn't matter. I think that Duke will continue to slide, and that Louisville will be the no. 4 seed. And I really don't think it matters too much if we play them or Stanford in the final 4, that is, assuming that the top 4 seeds are the final 4 teams. That might not be the case, but the tournament should be set up as if that were the case, of course.

 

[Phil]

I think that the fact that they'll have been unanimous number one for the better part of the season and the fact that they are the defending national champions, it'll be a pretty obvious selection.

I don't think defending national champion carries much weight.

 

[Vowelguy]

I emailed Creme and he responded, confirming that he has NOT bracketed the final 4. Thus he is not predicting a UConn-ND semi. And the semis are determined on Selection Monday.

 

[Phil]

I emailed Creme and he responded, confirming that he has NOT bracketed the final 4. Thus he is not predicting a UConn-ND semi. And the semis are determined on Selection Monday.

Thanks for checking that out.

 

[DobbsRover2]

Those rules were written prior to the decision to allow Regionals to host. In addition, the men have always placed more emphasis on the S-curve than the women, especially recently, so I wouldn't automatically assume that last year's men's rule will be this year's women's rule.

Phil, I'm not quite sure I understand the reasoning. I thought that the S-curve referred more to the snaking level ideal of having the #1-#8-#9-#16 for instance seeds together for the 34 total (not that it ever works out that way exactly) rather than any real significance of how the top 4 seeds should match up #1 vs #4 and #2 vs #3.

And I'm not understanding how the Regionals hosting and any marketing considerations would reflect on how that last FF weekend is matched up and what would set up say a #1 vs #3 and #2 vs #4. Can you explain that concept more?

 

[CompSci87]

I emailed Creme and he responded, confirming that he has NOT bracketed the final 4. Thus he is not predicting a UConn-ND semi. And the semis are determined on Selection Monday.

It's hard to imagine that he hasn't thought about the order of the regional #1 seeds. I think the problem is basically a software limitation -- the ESPN web page doesn't allow him to rearrange the regions each week so that those with the #1 and #4 overall seeds appear on one side and the #2 and #3 on the other (to match the lines that the page draws), so he just doesn't bother. Occasionally when he writes an article to accompany a bracket update, the article says (or at least implies) what order he thinks the overall seeds belong in.

 

[Vowelguy]

. I think the problem is basically a software limitation -- the ESPN web page doesn't allow him to rearrange the regions each week so that those with the #1 and #4 overall seeds appear on one side and the #2 and #3 on the other (to match the lines that the page draws), so he just doesn't bother.

Exactly correct!

 

[Phil]

Phil, I'm not quite sure I understand the reasoning. I thought that the S-curve referred more to the snaking level ideal of having the #1-#8-#9-#16 for instance seeds together for the 34 total (not that it ever works out that way exactly) rather than any real significance of how the top 4 seeds should match up #1 vs #4 and #2 vs #3.

And I'm not understanding how the Regionals hosting and any marketing considerations would reflect on how that last FF weekend is matched up and what would set up say a #1 vs #3 and #2 vs #4. Can you explain that concept more?

The rationale behind the S curve is that the best teams should get the easiest opponents. That can be viewed as a three step concept:

1. The 1 seeds get to play the 16 seeds 2-15, etc,
2. The allocation of the 1 and 2 seeds such that the overall top team should play the overall 8th best team, and
3. When setting up the regional format, the top overall seed should meet the 4th best in the semi.

The first of these three steps still applies, but the second step has definitely given way to costs/geography as a consideration. (which is why I advocate that we should stop calling it the S-curve. Step 1 doesn't require the "S".)

What is less clear (to me at least) is how they handle step 3. Geography plays a role in step 3, but not step three, after all they are all going to the same place. Do they return to the S curve concept - i.e. match up the overall 1 versus 4? Anecdotal evidence suggests this is not the case, so perhaps TV considerations, either timing or desirable match ups, play a role. That what I meant by marketing considerations, that they might identify the best match ups form a marketing point of view, rather than automatically pitting 1 against 4.

As for regionals hosting, it throws a spanner into the works. For the first (and only) time, you can have a team playing a regional on their own floor, while not being a 1 seed. This is certain to happen with Nebraska, and may happen with Louisville. Now the committee has the additional complication of deciding how to factor that in. When they decided to make location the primary consideration in step 2, it meant the possibility that the overall top seed got to play someone other than the 8th best team. That was considered a small price to pay. But now we have the distinct possibility that you not only ask the top overall seed to play the 5th best team, but on their home court. I think it is possible they will decide this is a bridge too far, and will allow considerations other than pure geography in step 2.

 

[Phil]

It's hard to imagine that he hasn't thought about the order of the regional #1 seeds. I think the problem is basically a software limitation -- the ESPN web page doesn't allow him to rearrange the regions each week so that those with the #1 and #4 overall seeds appear on one side and the #2 and #3 on the other (to match the lines that the page draws), so he just doesn't bother. Occasionally when he writes an article to accompany a bracket update, the article says (or at least implies) what order he thinks the overall seeds belong in.

I think you are probably right, but this is astounding. Someone had to generate the chart - how hard can it be to change it, and make the regional location a variable instead of hard-coding?

 

[Vowelguy]

I think you are probably right,

Not 'probably' right. Exactly right. Creme confirmed.

 

[DobbsRover2]

The rationale behind the S curve is that the best teams should get the easiest opponents. That can be viewed as a three step concept:

Ah, thanks for the explanation. Personally, I never got too wrapped up in the exact numbers in an S curve because the estimated differences in the 5-8 and 9-12 etc teams are often so slight and very debatable much of the time, and on occasion I would feel better about UConn playing a specified #5 than a #8 due to matchups. But I realize that certain fans to the southward often believe their team has been insulted and dragged through the dirt by the committee due to the murderer's row they will have to face all the way from top to bottom. But in the best of cases, your team is only going to be playing three of the teams in its region, and the first team is pretty much a gimme unless it's named Ball State.