North v. South (Kinda Powertardish)

[Bullfn33]

So I expect we'll be hearing that Texas has a good league record but they played an unbalanced schedule, getting to play Texas Tech and Baylor twice therefore should still be considered a bubble team, right?

 

[sys]

I suppose I'm saying that those data are flawed them. 

If everyone played the same number of teams that are ranked 200+, 100+, 50+, etc. then it may make sense, but they don't. 

Have to agree to disagree here I suppose.

if i can presume to speak for you, i think what you are saying is that those data (games between 25-45 teams and 200+ teams) are not very informative in evaluating distinctions between the former group.  i would agree with that statement.

however, that doesn't mean you should discard those data, merely that those data are not extremely useful in solving the problem with which you are confronted.

to a certain extent this is semantics.  however, a) words matter, and b) the data matter, to the limited extent that differences exist in how the 25-45 teams fared against 200+ teams, those differences are relevant.

[ksu_FAN]

I suppose I'm saying that those data are flawed them. 

If everyone played the same number of teams that are ranked 200+, 100+, 50+, etc. then it may make sense, but they don't. 

Have to agree to disagree here I suppose.

if i can presume to speak for you, i think what you are saying is that those data (games between 25-45 teams and 200+ teams) are not very informative in evaluating distinctions between the former group.  i would agree with that statement.

however, that doesn't mean you should discard those data, merely that those data are not extremely useful in solving the problem with which you are confronted.

to a certain extent this is semantics.  however, a) words matter, and b) the data matter, to the limited extent that differences exist in how the 25-45 teams fared against 200+ teams, those differences are relevant.

Quite clear that your knowledge of statistics is more than I know (or I care to know).

IMO this very point is the reason quality wins, bad losses, overall SOS, OOC SOS, last 12 games, etc. are points to be emphasized (or those data points emphasized), especially when looking at teams 25-45 and the reason they have been for years.  I don't disagree with the selection committee on that, just that a couple seem to be emphasized differently based on the team they are looking at and the make-up of the committee in any given year.

I think we all agree at the end of the day the solution is to make sure you are a clear cut pick by the committee by a) scheduling a quality OOC and winning some of those games and b) having success in your conference by winning a few of the tough games (OU for us this year) and not losing games you shouldn't (Baylor for us this year).  Clearly, if you can win half your road games in your league and 80% of your home games in your league you are most likely going to be selected most years.  The reason we are probably in trouble this year is b/c we didn't follow through with _FAN criteria a) (see above). 

[catzacker]

this would really just be easier if the conferences sent their regular season champ. the real leagues lose about 6-7 spots every year because Butler or Davidson or Xavier decide not to show up in the Conf Tourney.  

[Skycat]

3) end the overemphasis on "quality wins", which inherently rewards schizo teams rather than consistent, lower-ceiling teams, and also reduces the evaluative environment from a 30+ game schedule to a much lower-game environment.

I think in this the committee is doing exactly what it wants to do on this point. 

It is their belief that a schizo team has a better chance of collecting more wins in the tournament than the team whose performance is more consistent.  And since the committee thinks that the schizo team has a higher probability of making a later round, they are therefore more deserving than the lower-ceiling team, and also a potentially more exciting addition to the field.

I'm not sure that I disagree.

[sys]

i can see all your points, skycat.  just don't think it is "fair".

[Skycat]

i can see all your points, skycat.  just don't think it is "fair".

I can see that.  I guess I'm just pointing out that the committee is unlikely to change a selection parameter when it is getting what it wants.



Also, on the broader point of a systematic selection criteria:  the problem with establishing such a thing is that the jokers that run the system would constantly be monkeying with it.  Over and over with the BCS the results of an established system have been changed because "they" decided that the wrong teams were selected by that system. 

I don't see much benefit to a systematic aproach when I have the expectation that that approach will change after every season or two.