Toughest schedule in the Country

[Marquette84]

So if there's not much difference between 1 and 50 in the SC's eyes, but there is a difference between 1 and 50 in the W/L column's eyes, who would you rather play?

Really?

#1 Kansas was awarded a #1 seed (as was #10 Pitt). 
#50 Colorado State didn't even get a bid.

So I think the difference is a bit greater in the SC's eyes than you are willing to admit.

[brewcity77]

Really?

#1 Kansas was awarded a #1 seed (as was #10 Pitt). 
#50 Colorado State didn't even get a bid.

So I think the difference is a bit greater in the SC's eyes than you are willing to admit.

*Sigh*

And here I thought we could let our differences on this die. Guess not.

#1 and #50 RPI are both considered quality wins. Is there a difference? Yes. However, unless you disagree with any of these points, you're missing the point:

1) It is much easier to beat the #50 RPI team than it is to beat the #1 RPI team despite both being considered "quality wins"

2) There is a smaller difference in RPI percentage between the #1 and the #50 teams than there is between the #150 and #340 teams

3) An at-large bid quality team should have relatively little difficulty beating either the #150 or #340 team at home (margin of victory is irrelevant)

If you disagree with any of the above points, fine, I guess you can make your case. Otherwise, it's pretty hard to argue against it making more sense to fill the top of your schedule with teams you expect to be 30-50 in RPI and the meat of it with teams in the 100-200 range than it is to stack the top with top-25 teams and the bottom with sub-250s.

[Marquette84]

*Sigh*

And here I thought we could let our differences on this die. Guess not.

Well, if you're going to continue pressing the argument. . .  

In this case, you're now applying a double standard to two different groups.

On the one hand, you claim that because the selection committee considers any team 1-50 regardless of who they are to be "good wins", then there must not be any difference between them.

Meanwhile, the very same committee treats everyone 101 to 345 as a bad loss.  In the eyes of the committee there is no difference between the #101 and #345 team.  At least no more so than between #1 and #50.

[Lennys Tap]

*Sigh*

And here I thought we could let our differences on this die. Guess not.

#1 and #50 RPI are both considered quality wins. Is there a difference? Yes. However, unless you disagree with any of these points, you're missing the point:

1) It is much easier to beat the #50 RPI team than it is to beat the #1 RPI team despite both being considered "quality wins"

2) There is a smaller difference in RPI percentage between the #1 and the #50 teams than there is between the #150 and #340 teams

3) An at-large bid quality team should have relatively little difficulty beating either the #150 or #340 team at home (margin of victory is irrelevant)

If you disagree with any of the above points, fine, I guess you can make your case. Otherwise, it's pretty hard to argue against it making more sense to fill the top of your schedule with teams you expect to be 30-50 in RPI and the meat of it with teams in the 100-200 range than it is to stack the top with top-25 teams and the bottom with sub-250s.

Your points about the RPI are correct, but it's such a flawed system that most experts now dismiss it and the committee relies on it less each year.

[Benny B]

Your points about the RPI are correct, but it's such a flawed system that most experts now dismiss it and the committee relies on it less each year.

With exception to the 2010-11 committee.

[brewcity77]

Well, if you're going to continue pressing the argument. . .  

In this case, you're now applying a double standard to two different groups.

On the one hand, you claim that because the selection committee considers any team 1-50 regardless of who they are to be "good wins", then there must not be any difference between them.

Meanwhile, the very same committee treats everyone 101 to 345 as a bad loss.  In the eyes of the committee there is no difference between the #101 and #345 team.  At least no more so than between #1 and #50.

...And you fail to address any of the three points I bring up.

[Marquette84]

...And you fail to address any of the three points I bring up.

Well, you don't seem to like it when I adddress them point by point . . .but fine.

1) It is much easier to beat the #50 RPI team than it is to beat the #1 RPI team despite both being considered "quality wins"

Granted.  Its also much more likely to lose to the #101 team as the #345 team, despite both being considered "bad losses".

As I said, you're using NCAA's definition of "quality win" to conclude that they 1 through 50 are equal.  And that's fine--as long as you're consistent and declare that there is also no difference  between 101 and 345.

2) There is a smaller difference in RPI percentage between the #1 and the #50 teams than there is between the #150 and #340 teams

Not as much as you assume.  You make the mistake of assuming that there is a constant gap between teams, so the rank is a relative indication.

The difference between #1 KU (.6793) and #50 Colorado State (.5758) was .1035.

That's about the same as the difference between #150 San Jose State (.5051) and #318 Longwood (.4017).

Actually, because you're a stats guy and would appreciate this fact, even though Longwood is ranked 168 places behind San Jose State,  there is actually slightly less difference between Longwood and San Jose State than there is between KU and CSU--even though only 50 teams occupy that spread.

3) An at-large bid quality team should have relatively little difficulty beating either the #150 or #340 team at home (margin of victory is irrelevant)

Yes.  Thats why there is no difference between 150 (actually 100) and 345 in the eyes of the NCAA.  They're both equally bad losses.

[brewcity77]

Well, you don't seem to like it when I adddress them point by point . . .but fine.

Don't see you do it often enough to really have a comment on that.

Granted.  Its also much more likely to lose to the #101 team as the #345 team, despite both being considered "bad losses".

As I said, you're using NCAA's definition of "quality win" to conclude that they 1 through 50 are equal.  And that's fine--as long as you're consistent and declare that there is also no difference  between 101 and 345.

Yes and no. Both will be considered quality wins. Both will be considered bad losses. But it's the difference in degree of how much they effect us negatively. Duke, despite the loss, gave us a positive RPI boost of 1.12. Gonzaga, also a loss, gave us a positive boost of 0.64. That's as close to 1 and 50 as we'll get in the non-conference in terms of games on a comparable court. So the difference is 0.58, significant, but both notably positive, and less than double the positive boost for Duke.

Then look at the difference of UW-Green Bay (-0.21) and Centenary (-1.36); both home wins. A whopping 1.15 difference and over 600% the negative impact, so playing those crapcakes hurts us far more than playing a team about 50 spots away at the top of the RPI. If we're playing teams closer to 100 than the 171 of Green Bay, that difference would be even steeper. You'll take negative hits for the lower teams, but when they're that low, the hits are bigger.

And most important, we should be able to routinely beat UW-Green Bay or Centenary. Yes, upsets happen, but I think the odds of us beating a Gonzaga are much better than the odds of us losing to a Northern Colorado. So shoot for 30-50 ranked quality wins (the easier to get QWs) and aim for 101-200 cupcakes (the more valuable ones). My guess is had we beat Gonzaga, it would have been worth about as much as the loss to Duke. And if we had played 2 teams at Gonzaga's level instead of Gonzaga and Duke, we probably would have won one and been able to add another quality win to the resume in addition to the RPI boost.

Not as much as you assume.  You make the mistake of assuming that there is a constant gap between teams, so the rank is a relative indication.

The difference between #1 KU (.6793) and #50 Colorado State (.5758) was .1035.

That's about the same as the difference between #150 San Jose State (.5051) and #318 Longwood (.4017).

I'm not assuming anything. I looked up the numbers. And the difference between #150 SJ State (.5051) and #340 Central Arkansas (0.3584) is a bigger chunk by close to 50%. I saw what you did there, picking the Longwood team that was 22 spots off the number I gave. Hmm...why didn't you pick the much closer #342 Centenary that we also played? Because it would damage your argument?

Actually, because you're a stats guy and would appreciate this fact, even though Longwood is ranked 168 places behind San Jose State,  there is actually slightly less difference between Longwood and San Jose State than there is between KU and CSU--even though only 50 teams occupy that spread.

And? Where did I say 318? I understand the differences. I also understand how the Selection Committee responded to the number of sub-200 teams that were played. You constantly downplay this. Let's look at the direct numbers relevant to Marquette. Here is a list of teams with .500 BCS conference records or better, first their seed, then the team, then the number of sub-200s they played.

4) UConn, 2
8) Michigan, 3
9) Villanova, 3
9) Tennessee, 3
9) Illinois, 4
10) Penn State, 4
10) Michigan State, 4
11) Missouri, 5
11) Marquette, 6
12) USC, 6
12) Clemson, 6
OUT) Virginia Tech, 6
OUT) Alabama, 7
OUT) Colorado, 8

Maybe it's just coincidence that those numbers line up perfectly...yeah, that's probably it.

Yes.  Thats why there is no difference between 150 (actually 100) and 345 in the eyes of the NCAA.  They're both equally bad losses.

They're both bad losses, yes. But how they effect our RPI and SOS is light years away from equal.

And bottom line, last I checked, you seemed to care about getting wins. So do you think we should schedule Duke, Kentucky, Ohio State, and Wisconsin as our non-con heavy hitters every year and nine teams projected as sub-300 for the other games? That'll give us 9 guaranteed wins and games against some top opposition. Probably 3-4 more losses most seasons, but still, games against top opposition.

I'd rather see us play Washington, Missouri State, Georgia, and Wisconsin, where we have a better chance of getting 2-3 quality wins (and possibly all 4) while also raising the level of our cupcakes so they aren't all Centenary-level hits. And I realize PVAMU, TAMU-CC, and Longwood didn't hurt us that much, but still, they hurt us more than three times as much as UWGB did. And if things go well, those wins could be notable positive boosts, like we got from UW-Milwaukee and Bucknell.

Maybe it's just me, but I think that having a shot at 12-1 or 13-0 with 3-4 quality wins and low RPI drags is better than 9-4 and no quality wins and 9 heavy RPI drags. But like I said, that might just be me.