The Bubble: 2022

[Bona03]

We had really good coversation in the SLU game thread so I wanted to kickstart a thread to talk about the rest of the season and the Bubble. The Bonnie are creeping up many lists out there into that 5th or 6th out area, which is good. They need to keep winning and winning by a lot. As we saw last night, KP and NET are stupid and closely linked. the VT loss is a huge anchor around our metric necks and that is really shitty. The good news is our KPI and SOR are in decent shape and are approaching at large territory:


Its really hard to know what exactly the committee will do but based on the reveal last weekend, they are valuing the result metrics a little more than the predictive, also good for us. However, most I have spoken to have said the NET and KP are just too low so we need to get those up. I thought blasting Rhody by 18 would help but it barely moved the needle. This is what I have:
Ignore the sites, I have not updated those since last week. The way I see it, all those teams at the bottom are on our hate list. Anybody that is a 10 seed or below is also someone that can slide down with some bad results.

Everything about our resume is a NCAAT worthy except for the predictive metrics. The reality is we have to run the table in the regular season and then probably make the A10 final beating one of the top 3 in getting there. However, alot of teams around the bubble are crumbling and we keep winning.

The bad news here is we are all Duke fans. They play UNC and Virginia and can do us a solid by not losing to them. Rutgers plays Michigan tonight in a loser leaves town match. Creighton plays St Johns. and here are other games of interest courtesy of Jive on twitter:

Duke - Virginia - Go Duke
Miss St. - So Car
St. Johns - Creighton
Rutgers - Mich
TCU - Texas
Vtech - Gtech
Mason - VCU

 

[Community Bank 3Pt Tee]

Yep. Win out and win 2 in the A10s is the only last ditch dream of an at large, and even then I'm not sure. I'd hope what would be 11 in a row at that point and 23 total wins would be enough to lift our metrics and resume but who the hell knows.

A10 as a whole, if Davidson wins out they can afford an early exit, MAYBE if Dayton wins out they can afford an early exit- who the hell knows how their Q4 losses will be weighed. For us and VCU we obviously play each other so only one could even potentially emerge, and SLU seems to officially be in title or bust territory.

It sucks because the A10 is like half a step away from being in such a different scenario. When you look at the OOC results of the top 5, the amount of close calls against good teams + the few close, but awful losses is insane . As a whole the conference is a handful of OOC possessions away from being solidly in the 4 bid conversation because of how those losses reverberate throughout the year.

 

[res]

Kenpom is stupid and yet KPI is not? How one's team is rated should not affect one's judgment of a particular metric. It's not surprising that the only year Bona fans weren't crapping all over kenpom was last year when, not so coincidentally, Bona finished 32nd.

Also, the "results" vs. "predictive" distinction is a false dichotomy. By this point of the season, kenpom and similar systems are entirely results based. The results of every single possession of every single game determines team ratings. Kenpom is capable of being "predictive" precisely because it is results based.

 

[Bona03]

Kenpom is stupid and yet KPI is not? How one's team is rated should not affect one's judgment of a particular metric. It's not surprising that the only year Bona fans weren't crapping all over kenpom was last year when, not so coincidentally, Bona finished 32nd.

Also, the "results" vs. "predictive" distinction is a false dichotomy. By this point of the season, kenpom and similar systems are entirely results based. The results of every single possession of every single game determines team ratings. Kenpom is capable of being "predictive" precisely because it is results based.

I know you love KP but the fact is that the VT game is weighing us down heavily. KP is just a metric that takes the difference between adj off and def (in a nutshell, I know there are other factors but they dont mean much) so if you have a bad stretch, you get punished and cant recover. I am fine using it to try and break ties but when our last 2 games we utterly dominate our opponents are are no better for it. Something isnt right. We could beat VCU by 10 points and move up maybe 4-5 spots and thats it. Meanwhile winning that same VCU game by 10 will have a greater effect on your results based metrics like KPI and SOR. Look at our team. Solid record, solid wins, no bad losses really. Many teams above us have muliple q3 and even some q4 losses yet here we are with this anchor around our neck because of one damn game. Perhaps Canisius and Coppin St games are hurting us too but the point remains. We dropped 15 spots after losing an ugly game against UConn. UConn is a top 20 team and there should be little penalty to losing to them. Most people get that but KP says you didnt play well, here is your parting gift. Then we drop 27 spots more after the VT debacle and we still have not made up that ground. We were at 66 prior to those two games. Does KP care that Lofton missed 3 games and really was not ready for VT? Or that we had a month long covid pause? I bet not.

When you look at our resume, its solid. There are good wins and not many losses. At some point winning and losing has to matter more than your per possession numbers. Playing well and losing is better than playing ok and winning. That is fucked up. That is why the committee uses the results based metrics to make more decisions. Its a better indicator of if you can win games. Should we just let KP simulate the tournament and crown the champion that way?

KP is one tool. We didnt complain last year because there wasnt anything weird going on. What we saw in the metrics and on the court jived with one another. This year is far different. NET and KP are very closely related. Last year we had Colgate as a top 20 team. That was a running joke in CBB circles. When you have anomalies that are that far off, something isnt right. We all knew the NET was fucked up last year, it just didnt matter to us. Now it does. You can argue that KP is results based at this point in the season but my counter is that it still has baked in data from the whole season that thinks one bad game matters. To me that is not results based.

Love ya res but we will disagree hard on this.

 

[res]

Your last sentence, at least, is correct.

 

[Community Bank 3Pt Tee]

I cant hate on KP. We played well in only 6 of our first 19 games, even when we won, and the metric's purpose is to reflect that.

NET still has me puzzled as it does others and that issue extends far beyond us; it's gotten a fair amount of attention. We also shit the bed with a very rough 4-6 stretch at our most vulnerable time, when simply winning or losing dictated shifts in the NET by orders of dozens. I think we were as low as 120's in mid January? Certainly in 1-Teens. Considering that, we've climbed a good deal in a short amount of time. Its easy to lose sight of just how far we fell.

 

[res]

I cant hate on KP. We played well in only 6 of our first 19 games, even when we won, and the metric's purpose is to reflect that.

NET still has me puzzled as it does others and that issue extends far beyond us; it's gotten a fair amount of attention. We also shit the bed with a very rough 4-6 stretch at our most vulnerable time, when simply winning or losing dictated shifts in the NET by orders of dozens. I think we were as low as 120's in mid January? Certainly in 1-Teens. Considering that, we've climbed a good deal in a short amount of time. Its easy to lose sight of just how far we fell.

That's just a matter of timing, though. Reverse the order of games and the current rating would not change. I don't think most people understand that.

 

[3rdShayCripSet]

People that look at NET every single day after every game are truly nuts. People were complaining we dropped when beat Rhody last night, but didn't say much when we moved up while not playing.

Here's what I think our NET can do should we win the next three. As we've seen so far, road games move the needle more than home contests.

We are 84 right now, but hopefully we can be 81 or 80 by Saturday's game. Win that by more than the spread and perhaps we are at 76/77.

By the VCU game we could be top 75. Win there and we likely jump to 65...By that Friday's UR game, we are 63. If we beat UR, maybe we get to 60.

If we are top 65 by A10s, we will be in solid shape. If we are a 4 seed, we should likely have a Q2 game on Friday, and maybe a Q1 on Saturday. Like it or not, I think we need a trip to the title game to get an at large.

Much like 2018, it's great that we're winning all of these games, but besides SLU (X2), no win has enhanced our resume since we beat Marquette in November. All knockout games with minimal Q1/2 wins.

These past wins (for the most part) are keeping us in the picture, but VCU is the only one that's going to really help us. We saw what happened after we won at SLU. Big NET jump and we started creeping back in the picture.

If you lose down south, I am not sure there's a ton we can do to get back on the bubble. We'd need a perfect break. I think we'd need help and without question a trip to the finals with wins over two top 50 clubs while losing to no A10 bid stealer. If we got a weird 4/12 matchup or an 8 seed knocking out the No.1, we'd have to win the whole thing.

Nonetheless, so much will happen (good and bad) during the next couple of weeks. Just WINN

 

[Community Bank 3Pt Tee]

That's just a matter of timing, though. Reverse the order of games and the current rating would not change. I don't think most people understand that.

Yeah that's more or less my point. There's an expectation that a win should bump us up as much as previous losses knocked us down, and that's not how it works. If you want to say our rating doesnt make sense compared to others with similar team sheets, that's a different discussion, but a lot of the chatter right now seems to be converging the two conversations.

 

[Bona03]

Color me nuts then because I look every day but not just at ours. Our 3 biggest NET movements were losing to UConn, losing to VT and beating VCU. That resulted in -22 total (-15,-27,+20). I am not sure what I was expecting after Duq and Rhody but I didnt think it would be so small given that we won both by 18 and 26 and most importantly only giving up 55 points each time. Its wildly unpredictable. My argument is they are no reflective of the results and we have seen examples of teams like Wash St, who we know are not great basketball teams but they are so much higher. Just win games.

 

[res]

Movements early in a season are going to be larger than later in the season for the same reason a win or loss early affects your winning % more than an equivalent win or loss later does.

 

[3rdShayCripSet]

Color me nuts then because I look every day but not just at ours. Our 3 biggest NET movements were losing to UConn, losing to VT and beating VCU. That resulted in -22 total (-15,-27,+20). I am not sure what I was expecting after Duq and Rhody but I didnt think it would be so small given that we won both by 18 and 26 and most importantly only giving up 55 points each time. Its wildly unpredictable. My argument is they are no reflective of the results and we have seen examples of teams like Wash St, who we know are not great basketball teams but they are so much higher. Just win games.

I think it's somewhat easy to see when we move.

The teams that had the biggest moves last night in the NET (Miami, Fresno and Toledo) all had 20+ wins on the road.

Winning at home against team rated much lower than you, granted by nice margins, doesn't appear to move the needle too much. We jumped big time after beating VCU by 20 because we were dogs and VCU was at least 20 spots higher than us in the NET.

What we've done against Joes, UMass, SLU at home, Duq and Rhody isn't all that incredible in terms of the numbers.

Everyone wants to blame the corrupt NCAA, but it's not their fault we've had ZERO Q1 home games all season and only 3-4 Q1 road games. It sucks being a mid major without doubt, but if our schedule had 2-3 Q1 home games and 6-8 Q2 road games we'd likely be in a better position.

You point to Washington State, but they have eight losses by 6 points or less and then a 12 point loss at UA and a 20 point L at UCLA. They aren't even close to the bubble, so don't worry about them.

I know this is an unpopular opinion, but if you have a bunch of losses to top 50 teams you shouldn't be punished unless you get pummeled. We complain year and year about the unfairness of the CBB landscape, but we refuse to acknowledge virtually all A10 team shit the bed year in and year out OOC making this conference what it is.

Look at what the MWC is doing this year. They could get 3 maybe 4 bids. They are a mid-major, so it's still possible to get 4 bids IF you perform.

 

[Bona03]

Yea, the A10 did not do its job this year. Dayton and Davidson were nice surprises but Richmond shit the bed and so did we to an extent. The middle and bottom of the conference are pretty bad. I also agree about being competitive in the Q1. That is why UNC is in a bad position. They have got their ass kicked every time they have played a good team.

Last night was a pretty good night all things considered. Status quo in conference. It looked like Mason might help us out but no dice. However outside the A10 the following teams lost:
TCU
Virginia
Miss St
Wake Forest - to Clemson so double bonus
Rutgers - flip side is Michigan won but they have a better resume so Rutgers losing is probably the better outcome

Northern Iowa won, moving them almost into top 100

FWIW, Joey Brackets has us as first four out now. This is what I was hoping would happen, a gradual climb during a winning streak.

Our shit list for today:
Temple @ Memphis
Maryland @ Indiana
UCLA @ Oregon
LMU @ BYU
Belmont @ Murray St

 

[Community Bank 3Pt Tee]

Finally, a chance to root for Fatts Russel! A moment I've been waiting on since 2017.

 

[brownindians85]

I know this is an unpopular opinion, but if you have a bunch of losses to top 50 teams you shouldn't be punished unless you get pummeled.

Depends if you are also winning games. You should not get rewarded for going 0-11 against the top half of your conference in the B1G no matter how small the margin of victory, and then going 9-0 against PSU, Nebraska, Minn, NW and pointing to a lot of "good losses" as your resume highlight. If there's balance and you sprinkle in some good wins too, then sure, I'd believe a .500 in-conference P5 school is good enough to get selected.

 

[Citizen X]

I am back from overseas and looking forward to viewing some recordings of recent wins, which I did enjoy by listening to WPIG on the internet. As to Kenpom, it's pretty good in general, but Ken could enhance it with some application of the trimmed mean or Bayesian inference as some metrics do. However, I will also utter the blasphemy that the committee isn't the gaggle of idiots we often assume. I think our model for the coming selection should be their clairvoyance in choosing Dayton in 2014, VCU in 2011, or LaSalle in 2013. I know most feel we got boned in 2016, but to forgive is divine. Well, that's what the nuns told me anyway.
Note: A good example of what some smart guys are doing is the machine learning used by AMSTS, viewed at the following link:

Machine Learning Bracketology Methodology - All My Sports Teams Suck

Machine Learning Bracketology Methodology, say that five times fast. Anyway, the AMSTS Machine Learning Bracket Projection is exactly what it’s called, it leverages a number of Machine Learning models to try to replicate the NCAA Selection Committee’s decision process. We’ll hit on the process...

www.allmysportsteamssuck.com

 

[3rdShayCripSet]

Depends if you are also winning games. You should not get rewarded for going 0-11 against the top half of your conference in the B1G no matter how small the margin of victory, and then going 9-0 against PSU, Nebraska, Minn, NW and pointing to a lot of "good losses" as your resume highlight. If there's balance and you sprinkle in some good wins too, then sure, I'd believe a .500 in-conference P5 school is good enough to get selected.

Here's the thing. All of our fans are freaking out as if the bracket is set.

No team that goes 0-11 or 1-11 vs. Q1 is making it. The resume is much more than one quad despite what our fans think.

People really don't think the committee dives into this stuff? Sure, Q1 wins matter, but there's no doubt that a win over a team projected in the field is much more valuable than the same win over some team that just happens to be Q1.

Stop bitching about NET. IF we win the games we should, the NET will be fine and so will our resume.

 

[Bona03]

Last night didnt do much for us outside of Belmont falling, which was good. Tonight not much on the docket

SLU @ Richmond
SDSU @ San Jose St

SDSU should win easily. I think we want Richmond to cement the double bye. The loss wouldnt knock SLU down very far and would afford us a setback without worry.

 

[3rdShayCripSet]

Last night didnt do much for us outside of Belmont falling, which was good. Tonight not much on the docket

SLU @ Richmond
SDSU @ San Jose St

SDSU should win easily. I think we want Richmond to cement the double bye. The loss wouldnt knock SLU down very far and would afford us a setback without worry.

One of our best wins is a Q1 at SLU. Do we want them to potentially slip from a Q1 to Q2 for the road and Q2 to 3 for our home contest?

On the other hand, if Richmond wins a few games here, our game next Friday (LONG SHOT) could become a Q2. They're about 20 spots away, so it's not likely.

IF SLU wins tonight, I bet they win out. That means we have to win out otherwise we get the 5.

 

[Bona03]

One of our best wins is a Q1 at SLU. Do we want them to potentially slip from a Q1 to Q2 for the road and Q2 to 3 for our home contest?

On the other hand, if Richmond wins a few games here, our game next Friday (LONG SHOT) could become a Q2. They're about 20 spots away, so it's not likely.

IF SLU wins tonight, I bet they win out. That means we have to win out otherwise we get the 5.

I dont see them falling 15 and 16 spots to fall out of those quadrants with 1 road loss to a decent team. I guess if they lose by 40 it would happen. To me, cementing the bye is #1. Now of course Bonnies can win out and do that but being realistic, that game @VCU is not easy and we cant just expect to win it.

On the flip side, SLU and VCU play the last game of the season so maybe that works out anyway. That game is at SLU.