The gap between the BE and the Big 12

[handdownmandown]

I was bored and jonesing for tonight's hoop action, so I did some number crunching that you may or may not find interesting. There might be a bunch of stuff here that you already knew, but I figured it might be interesting, because it shows just how much little things can mean.

We argue on here all the time about what it's going to take to get 3/4/5/6/whatever amount of teams in, and where we are going to wind up in the conference RPI. Well, since the Big XII was the top dog last year and their conference and schedule structure is similar to ours, I figured they'd be a terrific place to start for the sake of comparison. And since we all know the die is cast during the non-conference, I decided to initially ignore all conference play.

First off: Nonconference record. The disparity is not as big as you might imagine: the Big 12 was 99-26 in D1 contests, the BE 93-31. The Big 12 had only one inexcusable loss: TCU to Longwood. We had at least four that I can think of: the FDU, St. Peter's, and Mercer losses by the Hall, and GTown losing to Northeastern. Adjust there, and we almost have the same exact non-conference record. Not quite, but close, which underscores what a total disservice Seton Hall did to the conference last season. Not to piss in people's eyes here, but when you really dig in, you can see how every non-con game is that big.

Then there's the non-con schedule strength. They kick our ass there, right? Well, the answer there is pretty shocking: depending on how you look at it, it's either not really, or flat out no.

Here are the numbers for the Big 12. The first number is the NCSOS as determined by the RPI, which is of course what's most important. The second number is the NCSOS as determined by Pomeroy, which is naturally a more involved attempt to devise a truer number - whether you believe it is or not is up to you, but I think a differing opinion might be helpful.

Kansas 1/13
Baylor 34/167
Oklahoma 70/162
Iowa St 88/143
Okla St 113/254
Kansas St 143/303
Tx Tech 160/210
Texas 169/246
WV 213/262
TCU 234/331

Now here we are:

GTown 24/81
Creighton 43/150
Nova 56/89
X 89/163
DePaul 126/221
StJohns 140/261
Marq 166/226
Prov 195/149
Butler 292/283
SHU 314/344

I looked at them side by side. When you use NCSOS RPI, we lose in three spots: we don't have anything to match Kansas, and Butler/SHU were way lower. Other than that, we were equal to, or a little better than, the Big 12 team in spots 2-8. That's it. However, in a side by side matchup using KenPom's statistics, we whack them. Again other than Kansas, we lead in every spot other than 10th (344 for SHU to 331 for TCU), and in some places it's by many spots. Nova for us is 2nd in the conference at 89 in KenPom; Iowa State is 2nd over there at 141.

Look at the difference in the Big 12 teams between actual non-con schedule as devised by RPI, vis-a-vis their Kenpom ratings. Some of those gaps are enormous, and they're all going the right way - in the favor of the conference. Other than Kansas, every single one is tremendous. On our side, it's all over the place. DePaul gets crapped on around here on a minute by minute basis, but they did the entire league a solid last season by going 8-5 with a decent NCSOS; West Virginia and Texas Tech each contributed an 8-5 nonconference record with a significantly worse non-con RPI. We say that, well, we can't have DePauls and still do well, but that's not true, the Big 12 had 2. On the other hand, Providence wound up with a schedule that, if you buy Pomeroy, was ridiculous considering the bang for their buck that they got out of it in the actual RPI. Was that their fault? Seems hard to blame them.

So, one of the two biggest things I think that got the Big 12 7 teams in and us 4 is borne out here: if you believe that Pomeroy's additional analysis is valid, they either a)got luckier than hell that the really crappy teams they scheduled wound up having a W/L record that is favorable for a higher RPI, b)specifically organized their schedules to achieve that gain, or c)both. I'm taking c), however I think that to attain such an enormous benefit across the board can't be done even if you organize well, because it's too damn hard to figure out if North Carolina A&T is going to do better on the whole than Northwestern State; you can do that to some level of certainty, but not entirely. At the same time, I suspect that our RPI/Kenpom split is weighted against us to a level that would be tough to duplicate. No data here, but based on what the Big 12 managed, we should be able to get better results going forward than we did last season.

The other thing that I think is big for them is, quite simply, Kansas. Just having them makes everyone else look better. They do some serious heavy lifting numberswise and keep the conference constantly in the conversation. And as we know, perception can become reality, otherwise ESPN wouldn't be fighting so hard to make it so.



After crunching all this, I decided to post it just to be a ray of hope into the forum. It could easily be that we're undershooting, and that our season is not going to be defined by the lower number of teams that we get into the tournament but the higher number. The bottom line here is, we shot ourselves in the foot with some ridiculous losses and, if I'm theorizing correctly, we wound up with a significantly worse bang for our buck in the non-con SOS for the RPI that we should have, or could have, had.

Fix the two of those and everything looks better. EVERYTHING. And on that note, go everybody, especially Depaul at home against Illinois-Chicago. Who knows, a win there instead of a loss might get a bubble team in.

Yes, Stever, it might.




**Edited to fix one statistical and one grammatical error

[R Jay]

[Bill Marsh]

Thanks.

[billyjack]

Wow, great work!

[stever20]

Part of the Big 12 was they had the right teams at 6 and 7- Baylor was 9-9 and in 6th, but had the 10th best OOC RPI. Oklahoma St was 8-10 and in 8th, but had the 12th best OOC RPI. Both of them lost 1 game OOC all year long. individual teams matter.
Oklahoma St 12-1
Baylor 12-1
Kansas St 10-3
Texas 11-2
Iowa St 12-0
Oklahoma 11-2
Kansas 9-4

Kansas only one that had more than 3 losses, and they had 3 top 20 losses. They then went 14-4 in conference play. Going into conference play with 3 teams with 0-1 losses and 5 with 0-2 losses just was huge for them. Really only Texas and to a lesser degree Kansas St didn't have a great OOC SOS.

Meanwhile, the Big East had a 3 way tie at 10-8 for 3rd place. all 3 teams had 3 OOC losses. Only X with a top 100 SOS in that group. The 9-9 team had 5 OOC losses. The 8-10 team had 3 OOC losses, then a killer loss in the conference tourney 1st rd.

Really don't know how much if Seton Hall would have made things better if they had won. Xavier would have still had 12 losses. St John's still would have had 12 losses. PC would have had 11 losses. That's the key number.

In short, the Big 12 got "lucky" because the right teams finished in the borderline positions. If Kansas St had finished 9-9 or 8-10, they don't make the tourney. If you do well in OOC you have a bigger leash- just look at Oklahoma St and Baylor last year.

[handdownmandown]

Seton Hall would have improved everyone's SOS by six wins had they won those, and Nova and Providence would have been aided by 9. Everyone's RPI would have gone up, making the numbers look better on a standalone level, as well as improving every league win incrementally. They would have also become one of those teams with one non-conference loss that you praised the Big 12 for having. And since two of those losses were at home, they had a disproportionate effect on Seton Hall's RPI. Seton Hall would have easily been an additional top 100 RPI team and probably better than 75. Wins over them would have been useful, losses less damaging.

There are a lot of little things there, but stuff like that can be significant when the entire impact is considered. if GTown is the first team out, are they now the last team in? Are X's numbers improved enough to get out of the PIG? Would it have been enough for Providence to get in before beating CU?

We would be singing a little better tune if GTown had squeezed in and X out of the play-in games.

[WaitingPatiently]

GTown losing to Northeastern was the biggest hit to their chances. And that was self inflicted.

[handdownmandown]

And DePaul.

[R Jay]

Those two bad losses really hurt Georgetown's NCAA Chances. If they won both of them they are much closer to, if not squarely on, the bubble on Selection Sunday.

[billyjack]

In addition to their bad home losses, Seton Hall also should've won their games vs Oklahoma (had big lead under 30 seconds, neutral court) and Mercer (blew double-digit lead, lost in OT, true road game I think). Would be curious to see how much we all would've improved our numbers if those two game also fell their way. During ooc, I think they were feeling their way thru an injury bug (Auda, Edwin).