Oklahoma: Worst W-L record ever for an at-large team?

[Plebe]

Oklahoma's overtime win today (79-77) at West Virginia is monumental for their postseason hopes.

The Sooners have struggled to maintain a .500 record this season ever since a rough four-game stretch, early in the OOC, which included losses to Little Rock and Florida.

To their credit, they have played a very strong schedule: #3 nonconference SOS and #2 overall according to the the RPI. Their sky-high strength of schedule has kept their RPI ranking high — now at #25 after today's win — even while their W-L record has hovered around .500.

Today's win puts Oklahoma at 14-12, with 2 of their remaining 3 games at home against Iowa State and Texas Tech (and the regular season finale at Texas). It appears likely that the Sooners will be 16-13 heading into the Big 12 tournament, and they could be 16-14 or 17-14 on selection day.

17-14 is the worst W-L record to ever get an at-large bid. Oklahoma looks to be on track to at least tie that record.

 

[JordyG]

NCAA women need a play-in game. Simple as that.

 

[trisailin]

NCAA women need a play-in game. Simple as that.

Though I don't disagree, what they ?need? more is to quit using a selection committee with all of it's biases and "feelings". If I were doing it, I'd suggest averaging Massaey, Sagarin and maybe a few others. Conference champs plus the next in line to fill the 64 team field.
The way it is now, it's ALL feelings.

 

[Centerstream]

I am totally against computers (I am guessing that Massey and Saragin are computer based) deciding the tournament in any sport. It's not because I am living in the 20th century still, it's just that software can be manipulated or hacked. They also might not consider how teams actually have performed. A team may have lost games against ranked opponents but if the software just considers the loss and not the fact that it was a competitive game until the final couple of minutes that might not be fair. And vice versa if a game is not competitive until the losing team's starters outscored the winning team's bench to make it look like it was a competitive game.

 

[jdubct]

Oklahoma's overtime win today (79-77) at West Virginia is monumental for their postseason hopes.

The Sooners have struggled to maintain a .500 record this season ever since a rough four-game stretch, early in the OOC, which included losses to Little Rock and Florida.

To their credit, they have played a very strong schedule: #3 nonconference SOS and #2 overall according to the the RPI. Their sky-high strength of schedule has kept their RPI ranking high — now at #25 after today's win — even while their W-L record has hovered around .500.

Today's win puts Oklahoma at 14-12, with 2 of their remaining 3 games at home against Iowa State and Texas Tech (and the regular season finale at Texas). It appears likely that the Sooners will be 16-13 heading into the Big 12 tournament, and they could be 16-14 or 17-14 on selection day.

17-14 is the worst W-L record to ever get an at-large bid. Oklahoma looks to be on track to at least tie that record.

Good question They would be neck and neck with Oklahoma State which is currently tied with Oklahoma for conference record. Oklahoma State did win in their single regular season game. I think it would go down to how well Oklahoma does in the conference tournament.

 

[trisailin]

I am totally against computers (I am guessing that Massey and Saragin are computer based) deciding the tournament in any sport. It's not because I am living in the 20th century still, it's just that software can be manipulated or hacked. They also might not consider how teams actually have performed. A team may have lost games against ranked opponents but if the software just considers the loss and not the fact that it was a competitive game until the final couple of minutes that might not be fair. And vice versa if a game is not competitive until the losing team's starters outscored the winning team's bench to make it look like it was a competitive game.

And you think the committee actually watched every single game and can remember every single detail? I would suggest that at least some committee members select teams which they have not even seen.
Further, your comments suggest a limited understanding of how the algorithms in either Massey or Sagarin operate. ( I do agree that hacking is possible but so is bribery of committee members)

 

[Centerstream]

And you think the committee actually watched every single game and can remember every single detail? I would suggest that at least some committee members select teams which they have not even seen.
Further, your comments suggest a limited understanding of how the algorithms in either Massey or Sagarin operate. ( I do agree that hacking is possible but so is bribery of committee members)

Good points and I admit that I don't know what the algorithms take into account. And I too wonder how each committee member knows who to pick but I am guessing (or hoping) that they have people working with them that have done the proper research. Plus Committee members are assigned “primary” and “secondary” conferences to monitor throughout the season so they should be doing their due diligence.

 

[Bear11]

While I agree that Sagarin/Massey should not be used exclusivley by the committee, they could certainly replace the RPI as the metric of choice. That would be a good first step for improving the bracket. The RPI should have no place in the committee discussions.

I also think that if Oklahoma wins 2 out of their remaining 3 they should be in the field. They are a top 25 team in the computer rankings and have certainly been playing like it after a slow start.

 

[Plebe]

While I agree that Sagarin/Massey should not be used exclusivley by the committee, they could certainly replace the RPI as the metric of choice. That would be a good first step for improving the bracket. The RPI should have no place in the committee discussions.

The committee is rightfully reluctant to adopt a metric that considers margin of victory because that would incentivize teams to game the system by running up the score.

 

[cbraarj]

Maybe Men and Womens basketball should go to a playoff like football! 4 teams only,lol

 

[Bear11]

The committee is rightfully reluctant to adopt a metric that considers margin of victory because that would incentivize teams to game the system by running up the score.

The creators of these systems write them in a way so as to minimize the potential impact of running up the score. Once the game reaches a certain point, slight differences in final margin are inconsequential and I believe the ratings reflect that.There is also something to be said for being in a position to run up the score in the first place.

If they have their heart set on not using margin of victory, there are methods that disregard it. Jeff Sagarin releases Elo ratings that are based on W-L only. This would be better than the RPI, which is arbitrary and has no basis in mathematical principles. The RPI is also subject to manipulation by teams in the form of scheduling, which can artificially inflate the RPI without truly playing a more difficult schedule.

I understand questioning the fairness of using margin of victory, but the results back it up. I think most everyone on this board would agree that the Massey/Sagarin systems do a much better job of ranking the teams than the RPI or polls. A quick comparison of their top 25 against your own eye test should confirm that.

 

[CTBballfan243]

NCAA women need a play-in game. Simple as that.

What they need to do is cut the field in half, not add a play-in game. There are not 64 Women’s teams that are good enough to play in the tournament. There are just a lot of really bad teams out there.

 

[Plebe]

The creators of these systems write them in a way so as to minimize the potential impact of running up the score. Once the game reaches a certain point, slight differences in final margin are inconsequential and I believe the ratings reflect that.There is also something to be said for being in a position to run up the score in the first place.

This is the first time I've heard that Sagarin or Massey consider the point differential during the game to mitigate against inconsequential scoring in garbage time.

I'd like to read more about this particular feature, if you or someone else can point me to a good explanation.

 

[Bear11]

This is the first time I've heard that Sagarin or Massey consider the point differential during the game to mitigate against inconsequential scoring in garbage time.

I'd like to read more about this particular feature, if you or someone else can point me to a good explanation.

The effect of point differentials is minimized as the differential gets larger. For example, the difference between winning by 20 or by 30 is not significant in these systems. On the other hand, winning by 12 instead of 2 would have a much larger effect. In addition, by taking all 30 games into account, the effect of running up the score in any particular game is minimal. A team would have to actively work all season to run up scores to make a major impact on their ranking. Any team able to do that is probably a much superior team anyway. When Baylor beats Kansas by 35, it isn't running up the score, that's just how much better Baylor is. If they had pressed hard all game and won by 50, it would've had almost no effect on their computer rating. On the other hand, if the final score had been close, within 15 points or so, Baylor would've been penalized even with the win.

An example of this is TCU this season. They had a phenomenal shooting night and beat Texas, and won a couple other close games. This propelled them into the human polls, but they never cracked even the top 40 of the computer ranking, because they recognized that luck was involved and sometimes teams win games they should've lost. Sure enough, TCU got crunched the second time through the conference. However, throughout their entire 7 game win streak and 4 game losing streak, the computer rankings recognized them for what they were; a borderline top 40 team.

 

[Plebe]

The effect of point differentials is minimized as the differential gets larger. For example, the difference between winning by 20 or by 30 is not significant in these systems. On the other hand, winning by 12 instead of 2 would have a much larger effect. In addition, by taking all 30 games into account, the effect of running up the score in any particular game is minimal. A team would have to actively work all season to run up scores to make a major impact on their ranking. Any team able to do that is probably a much superior team anyway. When Baylor beats Kansas by 35, it isn't running up the score, that's just how much better Baylor is. If they had pressed hard all game and won by 50, it would've had almost no effect on their computer rating. On the other hand, if the final score had been close, within 15 points or so, Baylor would've been penalized even with the win.

An example of this is TCU this season. They had a phenomenal shooting night and beat Texas, and won a couple other close games. This propelled them into the human polls, but they never cracked even the top 40 of the computer ranking, because they recognized that luck was involved and sometimes teams win games they should've lost. Sure enough, TCU got crunched the second time through the conference. However, throughout their entire 7 game win streak and 4 game losing streak, the computer rankings recognized them for what they were; a borderline top 40 team.

What you're describing in this post is something very different than measuring "game control" during the game.

Sometimes a game is nearly tied with about a minute to go, but the final result ends up being around 10 points because the team that falls behind by a bucket is forced to foul in the final seconds. Then you have the UConn-Louisville game, where UConn was in firm control from a dominant first quarter but the margin of victory was still only about 10 points. Vast difference between those two games. But same margin of victory for the "power ratings."

 

[Bear11]

What you're describing in this post is something very different than measuring "game control" during the game.

Sometimes a game is nearly tied with about a minute to go, but the final result ends up being around 10 points because the team that falls behind by a bucket is forced to foul in the final seconds. Then you have the UConn-Louisville game, where UConn was in firm control from a dominant first quarter but the margin of victory was still only about 10 points. Vast difference between those two games. But same margin of victory for the "power ratings."

There are limits to what can be done with computer rankings without feeding play by play data for every game, which would obviously be impossible on this scale. However, I think they do a very good job with what they have, better than the RPI and human polls. I wish there was a system in place like KenPom for men's basketball, which does look at player stats and game trends. However, the final results between KenPom and Sagarin are largely comparable.

When bookies and gamblers in Vegas set lines and place bets, they use KenPom and Sagarin to do so, not the RPI or human polls. While the NCAA wouldn't and shouldn't have gamblers decide the bracket, it is telling that those with money on the line prefer to use those metrics.

There will never be a perfect ranking system, because there can always be upsets. It also depends on whether they want to select to best teams, or the most deserving. There are arguments both ways, both very reasonable. My greatest wish is for the committee to have the best information available, which to me includes watching the games, advanced computer rankings, and hopefully less of the RPI.

 

[shizam]

What they need to do is cut the field in half, not add a play-in game. There are not 64 Women’s teams that are good enough to play in the tournament. There are just a lot of really bad teams out there.

So true. There is not one Conference this year that is really good. Fewer teams this year that are really good.
PAC 12 is the most entertaining and have by far the most competitive games but they don’t have any elite teams. Oregon is not Elite.

 

[MSGRET]

The main reason why the Power 5 get from 5 to 8 teams in the Dance is because of MONEY. ESPN wants teams that have a following and most of those are from the Power 5. If it wasn't so then the AAC would have 3 teams in this year, UConn, USF, UCF, maybe even a 4th with Houston.

 

[triaddukefan]

The main reason why the Power 5 get from 5 to 8 teams in the Dance is because of MONEY. ESPN wants teams that have a following and most of those are from the Power 5. If it wasn't so then the AAC would have 3 teams in this year, UConn, USF, UCF, maybe even a 4th with Houston.

Id take the 5th-6th, 7th place team in the ACC and SEC over UCF and Houston any day of the week. By the way.... I doubt ESPN makes much money on the NCAA Women's tourney.

 

[Phil]

The committee is rightfully reluctant to adopt a metric that considers margin of victory because that would incentivize teams to game the system by running up the score.

No, they not rightfully reluctant. I understand they claim that's the reason, but it is trivial to overcome.

 

[southie]

Oklahoma is a very interesting story. I applaud Sherri Coale for scheduling a very tough non-conference schedule. And, they didn't just play in the comforts of home; not many teams with OU's success over the last 2 decades are gonna make road trips to Colorado State and South Dakota State.

In last week's bracketology, Crème didn't have OU in the field. Don't even think he had them on the bubble. Think it w as the first time he them out of the tourney.

That led me to wonder if there is a rule that an at-large team has to have a winning record (above .500)? There is in volleyball. Texas A&M wasn't eligible despite having the #1 SOS in the country.

OU is currently at #25 in RPI. But, playing #227 Texas Tech and #111 Iowa State will bring them down a bit (mid-to-high 30's, I predict) as it will hurt their overall SOS, even with two wins. Still good. Then, they close the season at Texas (which will raise their RPI, win or lose, IMO).

That OT win at WVU was huge for them yesterday.

 

[Plebe]

The main reason why the Power 5 get from 5 to 8 teams in the Dance is because of MONEY. ESPN wants teams that have a following and most of those are from the Power 5. If it wasn't so then the AAC would have 3 teams in this year, UConn, USF, UCF, maybe even a 4th with Houston.

There is simply no evidence for any of this.

UCF and Houston won't make the Dance because their resumes aren't good enough. A win over USF today might have gotten UCF into the conversation. But as it is, UCF has zero wins over the RPI top 50 (their best win is over UC Davis) and two losses to teams outside the top 100. Houston's resume is even weaker: no top-50 wins with four losses outside the top 100. Both of them have resumes perfectly suited for the WNIT.

 

[Plebe]

That led me to wonder if there is a rule that an at-large team has to have a winning record (above .500)? There is in volleyball. Texas A&M wasn't eligible despite having the #1 SOS in the country.

OU is currently at #25 in RPI. But, playing #227 Texas Tech and #111 Iowa State will bring them down a bit (mid-to-high 30's, I predict) as it will hurt their overall SOS, even with two wins. Still good. Then, they close the season at Texas (which will raise their RPI, win or lose, IMO).

That OT win at WVU was huge for them yesterday.

Yes, only teams with a .500 record or above are eligible for at-large bids.

As long as Oklahoma wins those two games, I don't think that playing Texas Tech and Iowa State will bring down their RPI too much. Yes, their SOS will take a hit, but that hit will be offset, at least partially, by the improvement in winning percentage.

I think the biggest danger to Oklahoma's postseason hopes at this point is no longer their W-L record, but the fact that their wins over West Virginia and TCU may not count as top-50 wins in the final resume. That would leave the Sooners with only one top-50 win, over South Florida, which may not be enough for the committee.

 

[Plebe]

No, they not rightfully reluctant. I understand they claim that's the reason, but it is trivial to overcome.

I've heard people claim that it would be easy to overcome, but since I haven't seen anyone actually do it, I'm withholding.

 

[shizam]

There is simply no evidence for any of this.

UCF and Houston won't make the Dance because their resumes aren't good enough. A win over USF today might have gotten UCF into the conversation. But as it is, UCF has zero wins over the RPI top 50 (their best win is over UC Davis) and two losses to teams outside the top 100. Houston's resume is even weaker: no top-50 wins with four losses outside the top 100. Both of them have resumes perfectly suited for the WNIT.

I was at the UCF-UC Davis game. It was by no means in an easy win for UCF and I expected UC Davis to win the game but UCF was the better team.