Buzz concerned we did not play tOSU, is he right?

[Canned Goods n Ammo]

2) MU was without a key player during all of those losses you mentioned.  The committee will give extra weight to the post-UWGB schedule.

I don't think they factor in streaky reserve players.

[Benny B]

I don't think they factor in streaky reserve players.

To the extent that a "streaky reserve" player has a meaningful impact on the team's play and performance, the committee will take that into consideration.  Is that alone going to turn MU from a hypothetical 4-seed into a 2-seed ?  Is Mayo the difference between MU being a national contender and NIT-bound? Of course not.  But he could be just enough of an impact to bump MU up the s-curve a single line, and if you're already sitting at #9 or #13, that makes a difference on where you're seeded and, perhaps, where you play.

That said, maybe it's just my opinion, but I think MU is a better team with Mayo; if I am representing MU in committee, I'm absolutely going to point this out regardless.

[ChicosBailBonds]

Don't you remember Chicos giving us the valuable (and prescient) info that rpiforcast had us finishing the season with an RPI of 60 (or so) and I think a conference record of 9-9? He was positively giddy.

Actually, I was positively concerned and not giddy at all, but you're on a roll the last few days with your lying again so keep it going.

Also, that was in December, and as more games are played, the more data that comes in.  If you win games you are not predicted to do so, your outlook improves, that's why it is called a forecast.  If I believe Q1 will generate $6 billion but it comes in at $7 billion, guess what...my year forecast is likely to improve.  No different.

Keep them coming...giddy..yeah...maybe you out to re-read what I actually said.   

[tower912]

I don't think they factor in streaky reserve players.

Agreed that Mayo is a streaky reserve player.   They are 16-4 with him, 7-3 without him.    Now, that better record can be attributed to the team coming together, to STjr's continued improvement, Jamil stepping up, CO getting healthier.....lots of things.      MU has lots of positives for the committee to consider.    Last 12, RPI, SOS....MU's resume is very good.     And though MU's record with Mayo is better than without, I doubt that the committee will see his presence as the causality. 

[Canned Goods n Ammo]

To the extent that a "streaky reserve" player has a meaningful impact on the team's play and performance, the committee will take that into consideration.  Is that alone going to turn MU from a hypothetical 4-seed into a 2-seed ?  Is Mayo the difference between MU being a national contender and NIT-bound? Of course not.  But he could be just enough of an impact to bump MU up the s-curve a single line, and if you're already sitting at #9 or #13, that makes a difference on where you're seeded and, perhaps, where you play.

That said, maybe it's just my opinion, but I think MU is a better team with Mayo; if I am representing MU in committee, I'm absolutely going to point this out regardless.

I guess I just disagree. I don't think the committee knows these teams intimately enough to look at the roster and be impressed with Todd's stats.

I imagine they go through the normal criteria and add in any special circumstances. I don't see how a 6pt. 1reb. per game player would count as a "special circumstance".

To put it another way, if Todd suddenly can't play the rest of the year, do you think that moves them down a seed? I don't.

[The Equalizer]

Perhaps OSU would have won, but does that really matter now?  Remember that:

1) The committee doesn't care what might have happened then, it cares about what will happen 10 days from now.

Wrong.  The committe doesn't care at all what will happen 10 days from now.  100% of their attention is on what has occurred from November 11th through 7 days from now. 

There's no question that playing OSU would have potentially been beneficial to MU's RPI and SOS, particularly if MU won the game. 

Again, wrong.  Sorry, but I ran the numbers.  The game helps our RPI only if we won.  It hurts us if we lost.

But the committee is not going to run a simulation to determine which team might have won for purposes of a head-to-head discussion of the two teams in committee (i.e. who should be seeded higher).

They may not run a simulation, but if someone starts the discussion claming one of the teams should get special seeding consideration because this game was scheduled, the committee will have to find some way to decide which of the two teams should get that consideration. 

[Pakuni]

Wrong.  The committe doesn't care at all what will happen 10 days from now.  100% of their attention is on what has occurred from November 11th through 7 days from now.  

You know what didn't occur from November 11 through 7 days from now?
A game between Marquette and Ohio State.
They are not going to decide seeding based on something that didn't happen.

[The Equalizer]

You know what didn't occur from November 11 through 7 days from now?
A game between Marquette and Ohio State.
They are not going to decide seeding based on something that didn't happen.

Probably not.

Which is why its dumb to try to make the case that we deserve special consideration in seeding merely for scheduling the game.

And it doesn't excuse the incorrect comments that the game would have helped our RPI regardless of outcome.

[MU82]

MU was without a key player during all of those losses you mentioned.  The committee will give extra weight to the post-UWGB schedule.

If that proves to be true, Mayo will have been more valuable to us for the games he missed than the games he played!

[Benny B]

Wrong.  The committe doesn't care at all what will happen 10 days from now.  100% of their attention is on what has occurred from November 11th through 7 days from now. 


Again, wrong.  Sorry, but I ran the numbers.  The game helps our RPI only if we won.  It hurts us if we lost.


They may not run a simulation, but if someone starts the discussion claming one of the teams should get special seeding consideration because this game was scheduled, the committee will have to find some way to decide which of the two teams should get that consideration. 

One of the biggest fallacies in the selection process is that it's an ex post-type of process where the committee is rewarding teams for what they've done over the past five months.  The foremost job of the committee is to create a bracket that is competitively balanced... the committee focuses more on what a team is likely to do in the tournament rather than what they did in November.  While teams who have performed well over the course of several months will inevitably be rewarded, this is an effect of the selection process, not the cause.  In other words, if it was all about results and performance, then there is a number of ranking systems and computer models that could come up with a 68-team bracket... but then you'd have a bracket with UNM as a #1.

And you need to doublecheck your RPI math.

[Jay Bee]

And you need to doublecheck your RPI math.

Agreed there are issues with some of his math, but the assertion is true - a Marquette loss to Ohio State would have lowered MU's RPI calculation. The win-loss percentage taking a (25%-weighted) hit of .0083 would sting it badly.

[bilsu]

I remember Thad Matta saying they could reschedule the game and play it a Ohio St. I am not sure he was serious, but if Buzz is going to bemoan not playing the game because it hurt our strength of schedule, then he should of jumped on rescheduling the game at Ohio St.

[Benny B]

Agreed there are issues with some of his math, but the assertion is true - a Marquette loss to Ohio State would have lowered MU's RPI calculation. The win-loss percentage taking a (25%-weighted) hit of .0083 would sting it badly.

In any event, it's too time consuming – without the right software – to calculate marginal RPI in hypothetical scenarios; however, there's an interesting case I found in three teams with relatively similar RPI/SOS numbers: Kansas, OSU and MU.  Kansas beat OSU on their home court earlier in the year, and according to RPIforecast.com, that win has an impact of +1.15 on KU's current RPI.  But when you look at the impact on OSU's RPI, it's also positive: +0.98.

So OSU loses - at home, no less - to Kansas yet the OWP & OOWP seem to be overcoming the effect of 1.4 losses as the game still has a positive impact on OSU's RPI.

[The Equalizer]

One of the biggest fallacies in the selection process is that it's an ex post-type of process where the committee is rewarding teams for what they've done over the past five months.  The foremost job of the committee is to create a bracket that is competitively balanced... the committee focuses more on what a team is likely to do in the tournament rather than what they did in November.  While teams who have performed well over the course of several months will inevitably be rewarded, this is an effect of the selection process, not the cause.  In other words, if it was all about results and performance, then there is a number of ranking systems and computer models that could come up with a 68-team bracket... but then you'd have a bracket with UNM as a #1.

And you need to doublecheck your RPI math.

The math isn't wrong--I transcribed the wrong number for the calculated opponents/opponents RPI. 

[Jay Bee]

In any event, it's too time consuming – without the right software – to calculate marginal RPI in hypothetical scenarios; however, there's an interesting case I found in three teams with relatively similar RPI/SOS numbers: Kansas, OSU and MU.  Kansas beat OSU on their home court earlier in the year, and according to RPIforecast.com, that win has an impact of +1.15 on KU's current RPI.  But when you look at the impact on OSU's RPI, it's also positive: +0.98.

So OSU loses - at home, no less - to Kansas yet the OWP & OOWP seem to be overcoming the effect of 1.4 losses as the game still has a positive impact on OSU's RPI.

I think you may be confused as to what the different numbers at rpiforecast.com mean.

[The Equalizer]

In any event, it's too time consuming – without the right software – to calculate marginal RPI in hypothetical scenarios; however, there's an interesting case I found in three teams with relatively similar RPI/SOS numbers: Kansas, OSU and MU.  Kansas beat OSU on their home court earlier in the year, and according to RPIforecast.com, that win has an impact of +1.15 on KU's current RPI.  But when you look at the impact on OSU's RPI, it's also positive: +0.98.

So OSU loses - at home, no less - to Kansas yet the OWP & OOWP seem to be overcoming the effect of 1.4 losses as the game still has a positive impact on OSU's RPI.

You're misreading the data.  

The +1.15 applies only applies to Factor 2 (Opponents WP) & Factor 3 (Opponents/Opponents WP).  But it doesn't reflect the impact of the outcome of the game!

Again, I'm excluding the opponents' opponents WP--but the other two factors are easy to calculate.

Without the Kansas game, tOSU's opponents average winning percentage was .5690.  
(30 games ranging from .8709 for Duke  to .1923 for Chicago State)

With Kansas, Factor 2 increased to .5780, an increase of .0090.  With a weight of 50%, the RPI due to Factor 2 increased .0045.

Now consider the impact of the loss on Ohio State's own W/L (using the .6/1.4 factors for wins and losses:

Without the Kansas loss, tOSU/s own winning percentage was .8087 (18.6 wins vs. 4.4 losses)
With the loss to Kansas, tOSU/s own winning percentage was .7522 (18.6 wins vs. 5.8 losses)
WIth the loss, the WP decreased by .0465.
Weighted at 25%, the RPI decreased .0116

Now put them together.
--Improvement by playing Kansas:  +.0045
--Decline by losing to Kansas:  (.0116)
--Net change to Ohio State's RPI by playing Kansas and losing:  (.0071)

Ohio State's RPI declined by .0071 relative to the rest of their schecule because they played and lost to Kansas at home.  

[Benny B]

I think you may be confused as to what the different numbers at rpiforecast.com mean.

Then please enlighten, because the blog explanation on the site didn't help at all.