P5 OOC SOS, the AAC, and the Grits Diet

[DobbsRover2]

With its famously weak lineup of teams this year, the AAC is still making its mark in the Sagarin ratings, with not only the top rated team in UConn but three other teams -- USF, East Carolina, and Tulane -- rated in the top 40. UConn and USF also have played the top two SOS so far, and the 11 conference team's average SOS ranking is 142.2. That's not spectacular, but it should be enough to keep the AAC as the 6th rated conference through the regular season. Besides the top two teams, Cincinnati has also played a top 100 schdule at 68, while UCF at 273 and Houston at 334 are in the bottom 100.

The rigors of both OOC and conference SOS are supposed to feature heavily in a conference's success in the Tourney. Currently of course the AAC has the top all-time Tourney winning percentage at 90%, but it is based on only one year's play and includes a now departed team. Among the P5 conferences in the 2010s, the PAC has by far the best Tourney winning percentage at 71.1%, followed by the SEC at 62.5%, the ACC at 60.1%, the B12 at 56.4% and the woeful B10 at a losing 46.7%. Though the B12 has the next to worst winning percentage in the Tourney during the last 5 years, it did claim two NCs with Baylor in 2012 and the now-departed Texas A&M in 2011, and it has sometimes been rated with the toughest conference SOS in recent years. Conference winning percentage during the period does not correlate very well with regular season ratings, but proponents of the "tough conference games make you ready" theory could likely reasonably cite stats to show that is so, especially if the focus is put on the B10.

The current P5 conference SOS rank averages probably reflect the concerns of two conferences whose in-house slates are generally rated pretty poorly and whose teams feel they need to get some OOC grit. The B10 has the top ranking average at 125.0, and the PAC is right behind at 125.7. Two conferences whose teams likely feel they get eaten up in conference play have terrible averages, with the SEC the worst at 203.1 and the B12 a fraction better at 202.8. The ACC teams generally play a more reasonable OOC and they are averaging a little worse than the AAC at 146.0.

A listing of P5 teams that have played OOC schedules in the bottom 100 of 349 teams so far include many of the usual suspects, but also at least two surprises UTenn and Texas A&M, plus an undefeated Georgia team that has the 101st worst SOS. The teams include: OK State (346), Miami (322), Wake Forest (321), Indiana (316), VA Tech (307), Arizona (302), OR State (300), Auburn (284), Florida (283), Tennessee (281), Texas A&M (276), undefeated Northwestern (275), Mississippi (271), MS State (265), KS State (256), TCU (251). Oregon State has not yet played a road game, and five of the teams have played just one game away.

Some of there teams like UTenn (hosting Stanford soon) still have a tester in OOC games, and once conference play starts they will presumably get their diet of grits. Maybe it will make them ready for Tourney play. Hard to say.

 

[ochoopsfan]

NCAA have RPI's out and the Nitty Gritty and Team Sheets
RPI
http://www.ncaa.com/rankings/basketball-women/d1/ncaa-womens-basketball-rpi

Nitty Gritty or Team Sheets(use the appropriate drop down box on the links page)
https://rpiarchive.ncaa.org/default.aspx

 

[ochoopsfan]

Neither of those were the sources for the OP....

On the team sheets page it does
https://rpiarchive.ncaa.org/Stats Library/DI WBB Team Sheets 12-14.pdf

scroll down to page 283 for Tennessee, it lists their opponents and their opponents RPI's

 

[Fightin Choke]

Neither of those were the sources for the OP....

Dobbs said it was Sagarin: LINK

But I think that Dobbs' methodology was flawed anyway, as one should not average rankings when the numbers from which those rankings originate (Sagarin's SOS calculation in this case) are also available. Thus I think that part of his argument is lacking, although it is unclear whether or not that would alter his conclusion. He is certainly correct that there are many teams from Power 5 conferences that play horrible OOC schedules. But there is still a significant drop-off in the Sagarin Ratings between the Power 5 and the rest of the conferences, even before conference play begins.

Also, Tennessee has Notre Dame, Stanford, and Oregon State yet to play in their nonconference schedule, so their OOC SOS will certainly rise.

 

[DobbsRover2]

Dobbs said it was Sagarin: LINK

But I think that Dobbs' methodology was flawed anyway, as one should not average rankings when the numbers from which those rankings originate (Sagarin's SOS calculation in this case) are also available. Thus I think that part of his argument is lacking, although it is unclear whether or not that would alter his conclusion. He is certainly correct that there are many teams from Power 5 conferences that play horrible OOC schedules. But there is still a significant drop-off in the Sagarin Ratings between the Power 5 and the rest of the conferences, even before conference play begins.

Also, Tennessee has Notre Dame, Stanford, and Oregon State yet to play in their nonconference schedule, so their OOC SOS will certainly rise.

Yes it is Sagarin (type "Sagarin women's basketball" in Google to find it), and not sure where Sags has a conference SOS ranking, but if you could point out where it is it would be helpful, as I may have overlooked it. Sags does include three power ratings for central mean, simple average, and win 50%, but those are not SOS averages.

The SOS averages are fairly general, and an issue at this time of year for Sags is that recomputations can quickly change the rankings, as UConn and USF have already slid back from the early morning 1 and 2 ranking to 2 and 14, but teams stay within a general range. As to whether simply adding up the rankings of each team's SOS in a conference and dividing by the number of teams to get the average is a flawed or not is open to debate, but it should be close to actually adding the SOS power number such as 79.02 and 68.48 etc. True the gaps between two teams at the top may be a little more than for those in the middle, but at this moment there is only an 0.02 difference between SOS #1 Northwestern St. and #2 UConn. For those who are wondering how in Pluto that Northwestern St. (not Northwestern, which is at a miserable #284) could have the best SOS, they have played only 5 ratable D1 games that have included Texas, Texas A&M, Arkansas, and LA Tech, so they're a bit like Oakland.

I too was a bit surprised about UTenn's SOS since they have played Texas and an okay Rutgers, but when you play horribly rated teams like St. Francis (#329) and Lipscomb (#313), and also Tennessee St (#270) and some teams like Winthrop and Oral Roberts around the #200 power rating, your SOS is going to be poor.

Whether Sagarin's SOS system is the best is also open to debate, as some prefer Massey or Nolan or others. I personally do not even bother commenting on statements made about the brain-dead RPI system anymore, but others on the BY will wallow in it of course.

And stating that teams have played almost no road games or that a conference's average OOC SOS is above the 200 average is not meant to denigrate them. Each team needs to figure out what is the best method for them to get ready for conference and tournament play within their player and financial resources and their traditional scheduling. If a team like Stanford was to some year decide to play a #330 level OOC but use it to ride into an FF or NC game, I would say they probably had done things right.

But I do get a little ticked when I see some SEC team ride a #300 SOS with all home games to a 14-0 record and high ranking in December, and then watch them go 5-9 in conference play and finish the year without a Tourney bid. They shouldn't be rewarded in the early rankings if they do decide to go that route.

 

[DobbsRover2]

That's interesting regarding UT: they play top-10 Texas, top-20 Rutgers, top-10 Stanford, top-20 Oregon State, and top-5 Notre Dame. Those other OOC opponents on their schedule must be ranked in the 500s.....

Also, do you have a source link for your ranking data?

You just need to go to the Sagarin women's college basketball ratings page, which you can find in any search, or click here.
And no, a game that is on your upcoming schedule does not get counted in your SOS until you actually play them, so UTenn's OOC SOS will not get a boost until they really do play Stanford and ND, so no hypothetical "500" rated teams are needed to make their current weak SOS. Some teams will get their OOC SOS augmented an amount before conference play, while for others the die is pretty well cast. And for teams like UConn there will be a final kick with games like the one against USCar in the midst of conference play.

 

[DobbsRover2]

And why the specific Sagarin SOS numbers would be more useful than rankings is unclear because not only will the two methods be very close but using rankings would avoid certain skew factors from one or two really good or bad ratings numbers that pull a conference's average a little more in a direction that is not as representative for the other teams in the conference. Plus, those numbers are based on 5 games for some teams and 11 for others at this point, so you cannot get a representative average based on game played. But Choke it appears, sees it differently.

UTenn's schedule this year is very interesting, and may be a reasonable model for developing strength for the Tourney. The first 8 games (before Rutgers) the SOS is about 300, then the next 8 are a mix mainly of top OOC teams and one top SEC team (Texas A&M) inearly conference play, then the t

 

[KnightBridgeAZ]

For those who are wondering how in Pluto that Northwestern St. (not Northwestern, which is at a miserable #284) could have the best SOS, they have played only 5 ratable D1 games that have included Texas, Texas A&M, Arkansas, and LA Tech, so they're a bit like Oakland.

And stating that teams have played almost no road games or that a conference's average OOC SOS is above the 200 average is not meant to denigrate them. Each team needs to figure out what is the best method for them to get ready for conference and tournament play within their player and financial resources and their traditional scheduling. If a team like Stanford was to some year decide to play a #330 level OOC but use it to ride into an FF or NC game, I would say they probably had done things right.

But I do get a little ticked when I see some SEC team ride a #300 SOS with all home games to a 14-0 record and high ranking in December, and then watch them go 5-9 in conference play and finish the year without a Tourney bid. They shouldn't be rewarded in the early rankings if they do decide to go that route.

1 - There are a number of what I call the "ladies of the evening" to be polite, that make their living visiting better schools, collecting their payout, and usually losing. A few of them have spiced their schedule up with 2 or 3 DII or DIII teams which don't count. Grambling comes to mind as another school following that route this season, as I suppose NW St. is.

2. Road games are over-rated. For many reasons, it makes sense to play mostly at home. It doesn't make sense (at least to me) to combine that with a notorious weak schedule.

2b. Some teams are not getting ready for conference and tournament play. Or even trying too. Arizona's weak SOS is an effort to win a few games. Whether to keep the coach's job or just please some very angry fans I couldn't tell you, but . . .

3. Why get ticked - teams (often from the SEC) have a history of OOC success and in-conference failure, including Auburn many times when Fortner was there, Arkansas under Collen and Florida back in the day when Ross was there. And Ross had the nerve to complain when she didn't get a bid after a mediocre season and a very poor OOC SOS. No reason to get ticked.

And for all your RPI complaints, a bad schedule is a bad schedule - it isn't rocket science to know which teams schedule cupcakes, which schedule a pleasing mix and which actually play a competitive slate. Stats are only needed if you have a need to specifically rank them.

 

[DobbsRover2]

1 - There are a number of what I call the "ladies of the evening" to be polite, that make their living visiting better schools, collecting their payout, and usually losing. A few of them have spiced their schedule up with 2 or 3 DII or DIII teams which don't count. Grambling comes to mind as another school following that route this season, as I suppose NW St. is.

2. Road games are over-rated. For many reasons, it makes sense to play mostly at home. It doesn't make sense (at least to me) to combine that with a notorious weak schedule.

2b. Some teams are not getting ready for conference and tournament play. Or even trying too. Arizona's weak SOS is an effort to win a few games. Whether to keep the coach's job or just please some very angry fans I couldn't tell you, but . . .

3. Why get ticked - teams (often from the SEC) have a history of OOC success and in-conference failure, including Auburn many times when Fortner was there, Arkansas under Collen and Florida back in the day when Ross was there. And Ross had the nerve to complain when she didn't get a bid after a mediocre season and a very poor OOC SOS. No reason to get ticked.

And for all your RPI complaints, a bad schedule is a bad schedule - it isn't rocket science to know which teams schedule cupcakes, which schedule a pleasing mix and which actually play a competitive slate. Stats are only needed if you have a need to specifically rank them.

First, obviously a schedule that is 95% road games is factored into a team's poor OOC SOS, so it is not an issue of "having a problem with it" based on a team getting away with anything. They pay for it in their SOS.

But these weak P5s squads can't get a lot of decent teams to come visit them in the OOC, so no, a n0-travel sked that is almost all at home cannot work and does not make sense. It cannot make sense for a team to go into conference play having played 1 road game in the 7 weeks. Sure it will cut down on financial costs, which is no small matter for WCBB teams, but besides the lack of exposure beyond their campus, these teams cannot be as well prepared to hit the road against big time rivals when January hits, and they go 2-7 then. It is likely a reason that they do go 5-11 in conference play, and it is that which affects a coach's tenure more than 13 wins against OnTheRoad U and Number349AndProudofIt State. Sure these coaches can get some proud headlines in December on the PiggyBack website saying "13-0 and We're Looking Great," but then everything crashes down and everybody asks, "What Happened?" Well, reality set in for the unprepared.

Are any schools not trying to get ready for conference play, even if their goal is a humble 4-10 and maybe one more win than last year? Sure, expectations can vary widely, but I would never disparage even the coach I have least respect for and say she is not concerned about her team being ready to play in conference. That's just not right. They do care, but they may not be going about it in the most effective way.

And yeah, it's just me, but when I see that undefeated MS State team getting ranked off of a terrible but undefeated schedule that has included 9 of 11 at home, I do get a little ticked. Last year they were 13-1 in OOC off of only two road games and then promptly crashed to 5-11 in the SEC. Did they do anything last year to earn any kind looks this year? And yes, stupid to get ticked about it, and who knows, maybe they are the real deal this year. We'll see in January.

And if you compare the different ratings services, you will see some massive differences in how they rate schedules even in February, and one services #10 is another's #75. So no, there is not necessarily an agreement that a bad schedule is a bad schedule. Heck, even though UTenn's first seven to eight games were rated very poorly by Sags, their fans could say with a lot of justification just as UConn fans have sometimes said, "We played some tough teams, and sure there were a bunch of weakies in there that drag our SOS down, but the important thing is to test yourself against some strong top-level teams instead of getting a decent SOS by playing a slate of teams in the #75-100 level." Couldn't agree more if you are a top team.

 

[KnightBridgeAZ]

I may have been misleading about 2 things so let me correct them -

- I was referring to top level P5 teams when I said you should want to play at home as much as possible, maybe up to about 75% - if you are really playing some "tough" games they are going to be home and home and there are only so many top level tourneys. But most schools do want to play more than 50% at home, and if you can get a couple decent teams in the mix . . . Years ago RU had an in-season tourney and used to get some RPI boosters in for it (RPI does matter for tourney seed).

- I was being a bit catty about teams preparing for conference. Of course coaches are trying to develop their team for the best record they can achieve. Unfortunately, at least for Arizona that I follow, playing a seriously weak OOC schedule every year hasn't done so, we won one PAC game last year. And at some point, doing the same thing every year with more-or-less the same result . . . I will add that virtually annual transfers and injuries have had a major effect on the team, although I'm not sure how good they could have been without them.

 

[DobbsRover2]

Oh yeah, for the top teams the OOC issues aren't really a problem, since they are all pretty much invariably committed to a strong schedule with long-term home-and-away series, even if like UTenn this year the biggest parts of it are loaded on the end of it or in the midst of conference play. And sometimes teams like Georgia play a projected mediocre OOC schedule that heads even further south when a team like Ohio State goes bad. Almost all the teams on the "current weak OOC" list are those back of the P5 teams that have only a middling chance of making the Tourney even in a good year.

I suppose that some schools will never think seriously about getting the resources to make the Tourney, and that recruiting players with a "you get to play Stanford twice a year" line is enough for them. A bit sad though.