Blind Resume Wednesday Edition

Submitted by twotrueblue on March 13th, 2019 at 12:15 PM

In one of the posts earlier, someone mentioned they'd like to see a blind resume test. So here's one for the board. These are all teams with similarly balanced strength of schedules (according to the quartile system). A few of these teams are probably obvious, but more importantly, how would you rate them?

I'm thinking of doing this a few times before Selection Sunday. What other criteria would you like me to add to this matrix?



March 13th, 2019 at 12:17 PM ^

If nobody comes up with the five teams by tomorrow morning, I'll update this comment to reveal them. But I'm pretty sure somebody will come up with all five.


March 13th, 2019 at 6:56 PM ^

I like this exercise, but crashing the dance still uses RPI for their quadrants so a lot of this is wrong. 

Texas Tech doesn't have a Q2 loss, which was really throwing me off.  Iowa State is 22nd in NET so that home loss was still Q1.  They're 37th in RPI so crashing the dance has that a a Q2 loss, but that's incorrect.

Duke's actual resume lines up on a quadrant level almost exactly the way you display Texas Tech here (Q2 loss at home to Syracuse).  They have much better wins than Texas Tech's best win though.

So...can't use crashing the dance for these. EDIT: Sorry, didn't see this pointed out below.


March 13th, 2019 at 12:20 PM ^

E or C > D > B >A

And yes, Gonzaga(D) doesn't have that impressive of a resume. 

I wasn't aware Houston(C) had that good of a resume.

Texas Tech (E) plays a crappy conference 

Va Tech (A) has a more up and down offense than Michigan

Wofford (B)??? really?


March 13th, 2019 at 5:39 PM ^

I don't think the committee considers P5.  Conferences in general are irrelevant to them. 

I am surprised how good their resume looks in the quartile system as well.  Sounds like none of those wins would be considered 1A though so if that's something the committee looks at, it'll knock them down.

They're sitting at 13th in the NET.  I wouldn't be surprised if they're a 5 seed.  I would be shocked if they get a 4 seed though.  So they're not in the ballpark of these other teams.  While their 5-4 record against Q1 looks good on the surface, having only played 6 games against Q2 and Q3 combined and 15 games against Q4 limits them.


March 13th, 2019 at 11:23 PM ^

This is why NET is so much better than RPI, especially for mid majors. Under the old system, a road loss by a mid-major to a mediocre conference opponent killed them in seeding/at-large consideration, to say nothing of playing the dregs of a mid-major conference, who were usually major RPI anchors. Wofford obviously didn't have such a loss, but of course under the old system they wouldn't have gotten really any credit for winning any of those games either, and would probably be very much on or just inside the bubble had they lost in the conference tournament.


March 13th, 2019 at 5:47 PM ^

Houston will be an interesting case.  They have a shiny #4 NET and their quadrants look really good (as is demonstrated here).

But they're pretty bad in the three predictive rankings used on the team sheets: kenpom 13, sagarin 16th and BPI 14th.

The resume ranks say borderline 1 seed, the predictive ranks say 4 seed.  Will be interesting to see which way the committee goes.  Bracket Matrix says they're a three but I feel like they're probably a two considering how closely the committee stuck to the NET ranking in the initial seedings released a couple months ago.


March 13th, 2019 at 12:30 PM ^

Michigan is E, Houston is C, I think Tennessee may be D, Gonzaga is B (15 quad 4 games is a joke), I’m having a tough time with A. Kansas maybe? Texas Tech? Kansas St? Purdue?


March 13th, 2019 at 12:50 PM ^

One of the issues with this is that Crashingthedance is still using RPI instead of NET.  If you used RPI, NC State wouldn't even sniff the NIT, with an RPI over 100 and a 1-8 record against Q1, but they are a bubble team using NET.  As for Houston, they look way better using RPI (they get an additional Q1 win over BYU, for example, and they are in danger of one or both of their home games against UCF and Cincinnati dropping to Q2.  Houston has a ton of opponents with a much better RPI than NET ranking.  

I'm sure there is a site out there that has all the quadrant numbers using NET, but I don't know where it is.  I really liked the way crashingthedance did it, but I hope they shift to NET next year.  I assume the reason they didn't is that the NCAA wasn't publishing it every day.

EDIT: Houston's record against Q1 is still 5-2.  They lose the Q1 win over BYU but gain one over Utah St., who is #30 in NET but #33 in RPI.  However, that means that 2 of their Q1 wins and one Q1 loss are just barely in the Q1 range.  By comparison, the closest team on Michigan's schedule to the Q1/Q2 cutoff is the home game against #26 Maryland, and South Carolina is just as close on the other side of the cutoff.  


March 13th, 2019 at 12:54 PM ^

Good point there. I did notice the RPI being listed there, but it is such an easy website to gather the information. Can anyone attest to the accuracy of this site?

Given that the NCAA doesn't release the NET rankings daily, is this website tried to predict what the NET rankings should be or just using the last issued rankings?


March 13th, 2019 at 1:28 PM ^

That site looks right as of right now.  I don't know how often it (or NET, for that matter) is updated, though the NCAA has been updating NET on a daily basis as the tournament approaches.

As much as RPI sucked as a metric, it was easy enough to calculate that there were plenty of sources on the web to find updated RPI values within an hour after a game ending.  I hope the NCAA is willing to release their formula so that numbers nerds like me won't have to wait days for the next update.  For all the criticism it initially received (partly because they released the rankings too early), NET is an imperfect but significant improvement on RPI, and if the formula is good enough to use in the selection process, it should be good enough to be made public.


March 13th, 2019 at 1:52 PM ^

One other thing to mention on this point: Wofford (team B) is only 3-4 in Quad 1 when using NET rankings.  They lose both their home and neutral wins against UNC Greensboro, who is #30 in RPI but in the 50s in NET.  I liked your inclusion of them in this grouping, though, because I think they do deserve a good seeding according to metrics.  If you average their rankings in all of the metrics that are on the committee's team sheets, they end up somewhere in the 5-6 range.  I'm not sure I'd be looking forward to playing them in a hypothetical second round game.  

Leaders And Best

March 13th, 2019 at 1:38 PM ^

Agreed. There is no point looking at the arbitrary Quadrant cutoffs unless you are using the NET rankings used by NCAA. I was wondering why the OP's quadrant records were off.

The best site I have found for this is Warren Nolan. The layout is well done, and it has always had the most recent NCAA NET updates.


March 13th, 2019 at 12:53 PM ^

Pretty simple to figure out on KenPom. I was going to post them, but didn't want to ruin anyone's guessing so I'll leave some hints for those that want them:

B is not Gonzaga, as someone above suggested (but they are one of the teams). Should be obvious the UM is not one of the five - only team with five losses is E and our wins are better than #17 on Kenpom. 

Another hint is that none of the teams are SEC.


Leaders And Best

March 13th, 2019 at 1:50 PM ^

Gonzaga and Houston are the easiest to spot. They are the only two mid-major conference teams that have lost fewer than 4-5 games this year. Gonzaga is Team D (30 wins with the conference tournament completed), and Houston is Team C.

And any team that has played more than 30-ish games at this point has most likely completed their conference tournament. So Wofford is most likely Team B.

4th phase

March 13th, 2019 at 1:00 PM ^

1. D

2. C

3. E 

4. B

5. A

D has no bad losses and good record against Q1, and the best win. A has the most losses and is under .500 against Q1. B is 5-4 against Q1 with no good wins (highest is #55). 

I guess I'm really valuing record against Q1 and also your best wins and worst losses. Not really that concerned with overall record. 


March 13th, 2019 at 1:01 PM ^

1. D -- Most wins. Best win. Best Loss. 

2. E -- Most Q1 wins. Don't like that Q2 Loss.

3. C -- Fewest losses, but nothing exciting about this resume to put it above D or E. 

4. B? -- Good job not losing to bad teams. Though that is quite the trick to have your best win as #55, but still have 5 Q1 Wins -- Lots of road games vs 50-75?

5. A? -- Could interchange with B above, but that Q1 record is unimpressive. Plus a Q2 loss. That win over #3 looks more like a fluke than an indicator of success.