Latest (2/12) Bracketology

[LoTrader]

And as the saying goes, "if a frog had wings..."

Agree, a stretch, but UCLA is a lot better than I originally thought. If they get through the Oregon schools this weekend with victories in both, that may be enough! Of course, being an OSU graduate and superfan, I hope we wreck that plan Friday night in Corvallis.

 

[FridaFriend]

I have several objections to the way in which the bracket is drawn. First, I think that a conference should not be given more than one #1 seed. The ACC champion should get the #1 and Baylor should get the fourth #1 seed. Either Louisville or Notre Dame should get a #2 along with Oregon, Texas, and South Carolina. I would go with UCLA and Maryland as #3 seeds, but I think Tennessee and Florida State are weak choices. Tennessee gave up 99 points to Marquette, lost to LSU, and only beat Wichita State by 12. Florida State defeated such powerhouse teams as Florida, Iowa, ASU, and Creighton and lost to Texas. I think that none of the #4 seeds have earned the seeding. The non-conference records for the four teams are not impressive. NC State defeated Alabama, Tulane, Georgetown, and Vanderbilt, but lost to Rutgers. Stanford has lost 8 games including twice to Ohio State. Georgia defeated Virginia and Georgia Tech (barely) and lost big to Texas. Missouri defeated Cal, Kansas State, Indiana, Xavier, and Illinois.

I have focused only on the #1 through #4 seeds because they have home court advantage through the first two rounds. This leads to my second objection, that these 16 spots are owned by UCONN (justifiably) and 15 P5 conference teams. I would prefer to give each P5 conference 2 slots and for one for UCONN. That leaves five for teams like DePaul, Green Bay, FGCU, Central Michigan, and a South Dakota team. I know it won't happen, because the P5 conferences have a lock on the home court, but it could make a few upsets more likely. I also think that it would be much more fair to spread the home court around.

Finally, I wanted to compare several conferences based on the composite non-conference records of conference members. I looked at the P5 conferences, the Big East, and the AAC only considering games between those seven conferences. Not surprisingly, the ACC and the SEC had the best win-lost records: 41-29 and 39-29 respectively. The Big 10 and Big 12 were next at 31-32 and 21-19. The Pac 12 and Big East were weak at 16-24 and 17-21. The Pac 12 teams scheduled the fewest games against other P5 conferences, the Big East and the AAC. Oregon State, Utah, USC, and Colorado only scheduled two games each and Arizona only one game. The AAC was 18-29 or 7-29 without UCONN wins. Based on this I would say that the Big 12 deserves more recognition and the Pac 12 much less.

 

[Plebe]

I have several objections to the way in which the bracket is drawn. First, I think that a conference should not be given more than one #1 seed.

Really? So you think that UConn in 2013 should have been a #2 seed?

I would prefer to give each P5 conference 2 slots and for one for UCONN.

Teams are selected and seeded on an individual basis. I'm not sure why people think conferences should be apportioned seats as if it were the House of Representatives.

The Pac 12 teams scheduled the fewest games against other P5 conferences, the Big East and the AAC. Oregon State, Utah, USC, and Colorado only scheduled two games each and Arizona only one game. The AAC was 18-29 or 7-29 without UCONN wins. Based on this I would say that the Big 12 deserves more recognition and the Pac 12 much less.

The Pac-12 teams are geographically isolated from all of the other conferences you included, and it's not necessarily easy for them to find East Coast teams who want to travel west as part of a home-and-home. The per-conference game totals you cite are also skewed by two other facts: (a) the other four P5 conferences are forced to play in the conference "challenges," thus inflating the number of inter-P5 games they play; (b) the Pac-12 has fewer teams than the other P5 conferences except the Big 12.

 

[nwhoopfan]

Baylor had Stanford, UCLA and Kentucky, usually three consistent top teams, on the schedule. So there are your three ranked teams. UCLA is currently 7 and Stanford is currently ranked 14th.

Since you like to point out constantly that UCLA beat Baylor w/out Mulkey and Cox and thus it doesn't really count, Baylor's victory over Stanford also doesn't really count because their leading scorer McPhee was out w/ an injury. I'm fairly sure Baylor still would've beaten Stanford if McPhee had played, but she didn't and I'm just using your rules.

 

[nwhoopfan]

Agree, a stretch, but UCLA is a lot better than I originally thought. If they get through the Oregon schools this weekend with victories in both, that may be enough! Of course, being an OSU graduate and superfan, I hope we wreck that plan Friday night in Corvallis.

I was fairly stunned when Stanford won in Corvallis and Eugene. I will be even more stunned if UCLA pulls off the road sweep in Oregon.

 

[DallasBUFan]

Since you like to point out constantly that UCLA beat Baylor w/out Mulkey and Cox and thus it doesn't really count, Baylor's victory over Stanford also doesn't really count because their leading scorer McPhee was out w/ an injury. I'm fairly sure Baylor still would've beaten Stanford if McPhee had played, but she didn't and I'm just using your rules.

They still had their head coach though....

 

[PAHuskeyFan]

Does the committee look at the games and who played or didn't, or that Coach was away? I really don't think so, I think they look at wins and losses.... and go from there.

 

[Plebe]

Does the committee look at the games and who played or didn't, or that Coach was away? I really don't think so, I think they look at wins and losses.... and go from there.

Actually they do (or at least can) consider availability of personnel.

The following page from the NCAA website lists "availability of talent" as one of the criteria considered by the committee:
Women's Basketball Selections 101 - Selections

 

[PAHuskeyFan]

Actually they do (or at least can) consider availability of personnel.

The following page from the NCAA website lists "availability of talent" as one of the criteria considered by the committee:
Women's Basketball Selections 101 - Selections

Wow... very interesting.

 

[DallasBUFan]

Wow... very interesting.

I think it's quite fair and nice that they do.

Sometimes cards are stacked against these kids and they some way find a way to make it happen.

 

[Fightin Choke]

I just looked over the whole bracket, and my general feelings are, should it hold up, I like the way it unfolds for UConn. MS St, Baylor, ND & Oregon all are placed on the opposite side of the bracket, so UConn would potentially only face one of them in the National Championship. Louisville would be the top potential opponent in the national semi-finals, and I'm OK with that. If Louisville stumbles, TX or TN are the most likely suspects to replace the Cardinals.

After UConn takes care of whomever during the 1st two rounds in Storrs, they travel to the friendly confines of the TUC in Albany, where UConn fans have pretty much purchased all of the tickets available for sale and would likely face either NC St or TX A&M in the Sweet 16 game and UCLA or SC in the Elite 8 game. I don't see any of those teams presenting a problem for the Huskies in Albany.

This is always confusing to people but at the top of Charlie's bracket, it reads "Note: Regions not reflective of Final Four pairings".

That means that he is not attempting to assign which bracket winners face other bracket winners. Thus you cannot assume that UConn would only have to worry about Louisville being in UConn's bracket. (Maybe Louisville from the Lexington bracket would face Kansas City winner and UConn would face the Spokane winner, for example.)

 

[Fightin Choke]

I'd like to see Texas in the Miss State bracket. See if their post depth can help get McCowan in foul trouble and on the bench. Also, put Higgs/Atkins on Vivians and McCarty on William.

I just think it would be a great game and UT will have already 6'7 center 2-3 times, so it wouldn't be something knew for them. If Holmes and White are playing well they could cause some trouble for Miss State.

I kind of want BU as a 1 or 2 vs ND. I don't know why but I do. However, I would like to maybe travel and get out of KC region.

I think a lot of Baylor folks feel that way, but how well has Mulkey done against Muffet since Griner graduated? I know Baylor has height on Notre Dame (and every team except the one in Starkville), but Muffet is a great strategist. It's the same reason why I hope Notre Dame avoids Stanford this year.

 

[oldude]

This is always confusing to people but at the top of Charlie's bracket, it reads "Note: Regions not reflective of Final Four pairings".

That means that he is not attempting to assign which bracket winners face other bracket winners. Thus you cannot assume that UConn would only have to worry about Louisville being in UConn's bracket. (Maybe Louisville from the Lexington bracket would face Kansas City winner and UConn would face the Spokane winner, for example.)

That’s correct. The committee tries to keep the four #1 regional seeds as close to home as possible. Hence UConn is a lock for Albany. But with UConn the overall #1 seed, if the favorites all advanced, the Huskies would face the overall #4 seed, whichever region that seed came from.

 

[Fightin Choke]

If two teams have played three times, they should never have to face each other till the championship game. We've seen a couple of times in the not-too-distant past (TAMU and Baylor; UConn and Notre Dame) that it's really hard to beat a team a fourth time in one season. I'm pretty sure those were both finals, but not certain.

I agree with your point, but none of the games you mentioned were finals. In 2011, Texas A&M beat Baylor in the Elite 8 after losing the first 3 meetings. In the same year, Notre Dame beat UConn in the Final Four after losing the first 3 meetings. In 2013, UConn returned the favor; in the Final Four, UConn beat Notre Dame after falling to them the first 3 times they met that season.

 

[Fightin Choke]

That’s correct. The committee tries to keep the four #1 regional seeds as close to home as possible. Hence UConn is a lock for Albany. But with UConn the overall #1 seed, if the favorites all advanced, the Huskies would face the overall #4 seed, whichever region that seed came from.

But at the time of your post, wouldn't that have been Notre Dame (Spokane) and not Louisville who faced UConn in the Final Four? I thought you interpreted the bracket as if the brackets on the left met in the FF (and the teams that won the right 2 brackets met in the other FF game). That is, of course, the way it works when the actual bracket is published, but Charlie is intentionally not predicting that (for whatever reason).

 

[Bigboote]

I agree with your point, but none of the games you mentioned were finals. In 2011, Texas A&M beat Baylor in the Elite 8 after losing the first 3 meetings. In the same year, Notre Dame beat UConn in the Final Four after losing the first 3 meetings. In 2013, UConn returned the favor; in the Final Four, UConn beat Notre Dame after falling to them the first 3 times they met that season.

Thanks, mon.

 

[Fightin Choke]

I am a little skeptical that ND will garner a #1....But we will see.

Why are you skeptical? Is it that you do not believe that their resume is presently sufficient for a 1-seed? Or that you think Louisville will beat them in the ACC tournament? The tournament committee has downgraded teams because of injuries (which is why sometimes teams hide injuries or lie about their severity as we approach March), but Notre Dame has arguably played their best ball of the season after they adjusted from their last personnel loss (Lili Thompson). Sure the adjustment came after getting stomped at Louisville, but I'm sure that ND is excited for that rematch.

 

[oldude]

But at the time of your post, wouldn't that have been Notre Dame (Spokane) and not Louisville who faced UConn in the Final Four? I thought you interpreted the bracket as if the brackets on the left met in the FF (and the teams that won the right 2 brackets met in the other FF game). That is, of course, the way it works when the actual bracket is published, but Charlie is intentionally not predicting that (for whatever reason).

I probably did not look very closely at the brackets. My working assumption is that the overall seeding will eventually have UConn #1, MS St #2, Baylor #3 and either ND or Louisville #4. Of course the ACC champion could well displace Baylor, but again my working assumption is that if Baylor wins out they will get the #3 overall seed.

 

[Fightin Choke]

It really is a shame that RPI is still emphasized by the committee in any form. It is a horribly misguided system, especially given the nature of women's basketball. At the level we are talking about here, only a team in the top 40 or so has any chance to upset a top 5 team. After that, the result is essentially predetermined. It doesn't matter if your opponent is number 100 or number 300, it will be a blowout either way. The good computer ranking systems take this into account, and don't penalize a team for a blowout over a bad team. The only games which truly matter in nonconference for the elite teams are the few games against other elite teams. All the other games are just there to fill out the schedule, and can essentially be disregarded. Unfortunately, the RPI weights all games equally, so a blowout win over a team with a bad record is tremendously damaging. This makes it possible to "game" the RPI, by playing a lot of middling teams that will finish the season with strong records, but aren't major threats to actually beat you. Given the computer power available to us today, it is said that a pen-and-paper method like RPI is still given any weight at all.

Sorry for the rant, I am a big fan of advanced metrics and can't stand the dinosaur that is the RPI.

There are a few points that I believe require consideration here. First, the reason they use the RPI and not Massey or Sagarin is because they do not want margin of victory (MOV) included in the ranking they use, as it would encourage teams to run up scores. Thus we need to use rankings such as RPI or ELO Chess that ignore MOV. I guess you could argue whether or not that's a sound reason, but as of now, that is the methodology employed. Baylor is currently 5th in RPI and 4th in ELO Chess while MOV ratings hold them in the highest regard (Massey 2nd, Sagarin 2nd).

Second, Mulkey KNOWS that they use the RPI and that they will penalize you if you play too many hopelessly overmatched teams (cf Maryland 206-17), but she scheduled her murders' row of Lilliputians anyway. Why? Possibly because teams that were ranked a bit higher can request home-and-homes instead of just playing in Waco. Which leads to...

...Third, Baylor will have played 16/29 (55%) games at home this season (11 road 38%, 2 neutral 7%). Miss. St. is similarly bad: 55%, 31% 17%, as is Louisville: 55%, 45%, 0%. . In contrast UConn's numbers are 41%, 45%, 14%. Notre Dame is 45%, 45%, 10%. Creating a quick metric by adding the % home game to .5 x % neutral games, here are the "most homey" schedules of the top 16 teams in RPI (listed by how homey their schedule is):
61.7 Miss. St.
58.6 Baylor
58.6 Duke
58.1 Oregon
56.9 Tennessee
56.9 Mizzou
55.2 UCLA
55.2 Texas
55.0 Ohio St.
54.8 Louisville
53.4 Maryland
53.3 Stanford
50.0 Notre Dame
50.0 Green Bay
48.3 UConn
48.3 Florida St.

This is one list where you wanted to be ranked low, as it means that you play more games in hostile locations. So not only does Baylor have one of the most home dominant schedules of the top teams, but they also play a lot of very bad teams (see point 2 above). That's a double whammy.

Fourth, even if you argue that only games against top teams matter (as you do), Baylor has played fewer top 50 teams than the other top contenders.
Top 50 opponents (counts all games schedule in regular season)
16 Notre Dame
15 Stanford
15 Oregon
15 UCLA
14 UConn
13 Louisville
13 Tennessee
12 Ohio St.
12 Miss St.
12 Florida St.
11 Duke
11 Texas
10 Maryland
9 Mizzou
8 Baylor
5 Green Bay

The eye test suggests that Baylor is excellent, and woe is the team to have them as their 2-seed (if the Baylor fails to be awarded a 1-seed). But didn't many people feel that way about Maryland last season and that didn't work out well for Maryland. I would be disappointed to have Baylor as Notre Dame's 2-seed, but I still think that the Irish could beat them.

 

[Golden Husky]

I probably did not look very closely at the brackets. My working assumption is that the overall seeding will eventually have UConn #1, MS St #2, Baylor #3 and either ND or Louisville #4. Of course the ACC champion could well displace Baylor, but again my working assumption is that if Baylor wins out they will get the #3 overall seed.

This makes sense to me. ..which probably means we're both off our meds again, Dude.

 

[PAHuskeyFan]

Sure is a lot of discussion about a Bracket that means VERY LITTLE. Charlie Creame's Bracketology is accomplishing what it's designed to do I suppose...get people to talk about his "projection"... As we have seen year after year, Charlie's brackets do not really match up with what the NCAA Women's Basketball Tournament Committee will do in their final seedings. And, Charlie has not followed some of the basic "rules" the committee follows... such as Teams from the same conference are never (??) put in the same bracket unless there are more than 4 teams from a conference that make the tournament, thus it's impossible to not have more than one. For example...even if the "S" curve seeding would put USF in UConn's bracket, if they are the only 2, or 2 of 3 AAC teams in the tournament, the Committee would change USF with another similar seeded team (i.e., a 6 seed for a 6 seed) so that they wouldn't see each other until the final 4.

I don't get too excited about Charlie's brackets as they really don't mean much in the end.

 

[Fightin Choke]

Sure is a lot of discussion about a Bracket that means VERY LITTLE. Charlie Creame's Bracketology is accomplishing what it's designed to do I suppose...get people to talk about his "projection"... As we have seen year after year, Charlie's brackets do not really match up with what the NCAA Women's Basketball Tournament Committee will do in their final seedings. And, Charlie has not followed some of the basic "rules" the committee follows... such as Teams from the same conference are never (??) put in the same bracket unless there are more than 4 teams from a conference that make the tournament, thus it's impossible to not have more than one. For example...even if the "S" curve seeding would put USF in UConn's bracket, if they are the only 2, or 2 of 3 AAC teams in the tournament, the Committee would change USF with another similar seeded team (i.e., a 6 seed for a 6 seed) so that they wouldn't see each other until the final 4.

I don't get too excited about Charlie's brackets as they really don't mean much in the end.

Charlie is not wrong to place USF in UConn's bracket, as he is partially going off recent history. In both the 2015 and 2016 NCAA tournaments, USF has been placed in UConn's bracket (see 2015 and 2016).

Charlie doesn't know too much about women's basketball, but he really does understand NCAA bracketing principles. But he has trouble projecting the bracket because the committee changes which rules they follow. Sometimes it's because there are conflicting rules, and sometimes it's because they come up with a different S-curve than him. He analyzes their projections (the reveals) because it tells him how this particular seeding committee evaluates data, which should improve his brackets. He honestly doesn't do too bad, as his final bracket is usually fairly close. Now his early brackets are sort of ridiculous, but at least they get us talking.

 

[DallasBUFan]

I'm shocked we only have 1 loss. This was truly going to be a rebuilding year and the fact that we could be a top 8 seed, I'm cool with it. Young and lack of depth. These girls may very well be playing better than last years squad. They all seem to work together better.

I want ND and I don't care who is no.1 or no.2. Neither have depth, they history of beating each other in the finals, and I feel the best about beating them. However, I felt that way about Oregon State and we see how that turned out.

So what happens if Miss State loses? Their schedule is similar to BUs other than conference being stronger. Do they drop to a 4 seed?

 

[Bear11]

There are a few points that I believe require consideration here. First, the reason they use the RPI and not Massey or Sagarin is because they do not want margin of victory (MOV) included in the ranking they use, as it would encourage teams to run up scores. Thus we need to use rankings such as RPI or ELO Chess that ignore MOV. I guess you could argue whether or not that's a sound reason, but as of now, that is the methodology employed. Baylor is currently 5th in RPI and 4th in ELO Chess while MOV ratings hold them in the highest regard (Massey 2nd, Sagarin 2nd).

Second, Mulkey KNOWS that they use the RPI and that they will penalize you if you play too many hopelessly overmatched teams (cf Maryland 206-17), but she scheduled her murders' row of Lilliputians anyway. Why? Possibly because teams that were ranked a bit higher can request home-and-homes instead of just playing in Waco. Which leads to...

...Third, Baylor will have played 16/29 (55%) games at home this season (11 road 38%, 2 neutral 7%). Miss. St. is similarly bad: 55%, 31% 17%, as is Louisville: 55%, 45%, 0%. . In contrast UConn's numbers are 41%, 45%, 14%. Notre Dame is 45%, 45%, 10%. Creating a quick metric by adding the % home game to .5 x % neutral games, here are the "most homey" schedules of the top 16 teams in RPI (listed by how homey their schedule is):
61.7 Miss. St.
58.6 Baylor
58.6 Duke
58.1 Oregon
56.9 Tennessee
56.9 Mizzou
55.2 UCLA
55.2 Texas
55.0 Ohio St.
54.8 Louisville
53.4 Maryland
53.3 Stanford
50.0 Notre Dame
50.0 Green Bay
48.3 UConn
48.3 Florida St.

This is one list where you wanted to be ranked low, as it means that you play more games in hostile locations. So not only does Baylor have one of the most home dominant schedules of the top teams, but they also play a lot of very bad teams (see point 2 above). That's a double whammy.

Fourth, even if you argue that only games against top teams matter (as you do), Baylor has played fewer top 50 teams than the other top contenders.
Top 50 opponents (counts all games schedule in regular season)
16 Notre Dame
15 Stanford
15 Oregon
15 UCLA
14 UConn
13 Louisville
13 Tennessee
12 Ohio St.
12 Miss St.
12 Florida St.
11 Duke
11 Texas
10 Maryland
9 Mizzou
8 Baylor
5 Green Bay

The eye test suggests that Baylor is excellent, and woe is the team to have them as their 2-seed (if the Baylor fails to be awarded a 1-seed). But didn't many people feel that way about Maryland last season and that didn't work out well for Maryland. I would be disappointed to have Baylor as Notre Dame's 2-seed, but I still think that the Irish could beat them.

I realize that Baylor hasn't played quite as difficult a schedule as some of the top teams. My point is that the arbitrary and archaic RPI makes Baylor and the entire Big 12 look worse than they really are. I do think Baylor and Notre Dame would be an intriguing matchup that could go either way.

 

[Plebe]

So what happens if Miss State loses? Their schedule is similar to BUs other than conference being stronger. Do they drop to a 4 seed?

You think Mississippi St would drop from #2 overall to outside the top 12 after just one loss?