The top ten programs of the NCAA Tournament Era

[alexrgct]

Here's an open invitation to those who disagree with Alex's analysis. Do your own analysis with your own methodologies and share it with the rest of us here on the Boneyard. I'm sure it would provide for interesting conversation.

For one, thank you, and for another, I didn't create this methodology by myself. It was a collaborative, open discussion that started in January of 2012. Here is how it began:

http://the-boneyard.com/threads/program-of-the-decade.12120/#post-142664

 

[chrissaran]

Ten wins and 0 losses in the National Championship game. Four wins and 0 losses against the Vols. Five undefeated seasons. Winning streaks of 47, 70, 90 and now 37. Also, a home court winning streak of 99 games and no back-to-back losses since 1993. How to your calculations figure those statistics?

 

[rbny1]

Huh??? It's 30 for being a #1 seed and 70 for winning a national championinship. There are plenty of points for anything else. A #1 seed who was a national runner up gets 70 points total. Louisville, meanwhile, got 40 points for 2013 because they are a five seed (so no credit for their regular season). UConn has had several 60-point seasons for being a #1 seed who got as far as the national semifinals.

I'm not disputing your analysis or the work you put into it. Thank you for starting this thread. I'm just saying there are other ways to look at the numbers and assign weightings to various factors. I took the extreme of assigning all the weighting to one factor, which is legitimate (even if very narrowly focused) in its own right -- at least I try to think it is.

 

[JS]

Here's an open invitation to those who disagree with Alex's analysis. Do your own analysis with your own methodologies and share it with the rest of us here on the Boneyard. I'm sure it would provide for interesting conversation.

Good suggestion.

A bit labor-intensive as compared with taking partisan potshots or (more constructively) suggesting tweaks to the present methodology.

But good suggestion, even if meant partly as rhetorical rebuke.

It isn't easy to put these things together. I'm reminded of how vowelguy developed a system for ranking recruiting classes and had to endure a lot of what he may have regarded as ungrateful sniping. He's been quiet for a couple of months now, hopefully preparing a splashy re-entrance with new multi-colored charts.

 

[hussybulldog]

Name one system that's a better proxy for awarding a regular season than seeding in the tourney. Teams get high seeds for compelling records and SOS. It's how the overwhelming majority of the wcbb season is played. A disappointing result in the tourney doesn't erase everything the team in question had accomplished.

I am not against your methodology, other than question you the weight....if you do want to weigh in the regular season record, it comes to how much weight one ought to give and how one arrives to a conclusion of regular season greatness. Maybe an average ranking through out the season makes more sense than a one shot tournament seeding done by the selection committee since it involves geographic considerations limited perception of the few involved in the later? (And my argument with the #1 infamous seeding awarded to Stanford self exemplifies the error in that). Good work! But it can be always be improved.

 

[Outside Shot]

Terrific stuff, alex. Appreciate you taking the time to put it together. Thanks. I had the thought that a decade ago, Virginia might have made the list. Interesting how times change and how hard it is to sustain. Kudos to all the programs on the list.

 

[Gate81]

A Statistical Department is hiring mathematicians. Three recent graduates are invited for an interview: one has a degree in pure mathematics, another one in applied math, and the third one obtained his degree in statistics.
All three are asked the same question: "What is one third plus two thirds?"
The pure mathematician: "It's one."
The applied mathematician takes out his pocket calculator, punches in the numbers, and replies: "It's 0.999999999."
The statistician: "What do you want it to be?"
--------
Three statisticians go hunting. When they see a rabbit, the first one shoots, missing it on the left. The second one shoots and misses it on the right.
The third one shouts: "We've hit it!"

in my experience the only people who love these jokes are the ones who are deficient in math skills. I think that the work done was great. Not perfect, but neither is life.

 

[Waquoit]

in my experience the only people who love these jokes are the ones who are deficient in math skills. I think that the work done was great. Not perfect, but neither is life.

What did the mathematician do when he was constipated?
He worked it out with a pencil!

Now that's a joke! And so is any analysis that puts UT over UConn.

 

[BroadwayVa]

A Statistical Department is hiring mathematicians. Three recent graduates are invited for an interview: one has a degree in pure mathematics, another one in applied math, and the third one obtained his degree in statistics.
All three are asked the same question: "What is one third plus two thirds?"
The pure mathematician: "It's one."
The applied mathematician takes out his pocket calculator, punches in the numbers, and replies: "It's 0.999999999."
The statistician: "What do you want it to be?"
--------
Three statisticians go hunting. When they see a rabbit, the first one shoots, missing it on the left. The second one shoots and misses it on the right.
The third one shouts: "We've hit it!"

The National Collegiate Calculus Championships were held in Los Vegas April 8th thru 12th. Does anyone know who the individual and collegiate winners are??
MIT had 20 male and female cheerleaders dressed on cowboy outfits
Cal Tech had 10 male cheerleaders dress in combat engineers uniforms
Question does anyone know who won that tourney???
I thought since you are a Mathematic type you'd know--no joke who won?

 

[BroadwayVa]

What did the mathematician do when he was constipated?
He worked it out with a pencil!

Now that's a joke! And so is any analysis that puts UT over UConn.

YUK---this wasn't funny in 1939 when I first hear it.

 

[BroadwayVa]

I am not against your methodology, other than question you the weight....if you do want to weigh in the regular season record, it comes to how much weight one ought to give and how one arrives to a conclusion of regular season greatness. Maybe an average ranking through out the season makes more sense than a one shot tournament seeding done by the selection committee since it involves geographic considerations limited perception of the few involved in the later? (And my argument with the #1 infamous seeding awarded to Stanford self exemplifies the error in that). Good work! But it can be always be improved.

the number of permutations that could be used to calculate this 10 or any other collegiate sports ranking is astronomical---so I ask the stars. J. Low had no idea.

 

[BroadwayVa]

I'm not disputing your analysis or the work you put into it. Thank you for starting this thread. I'm just saying there are other ways to look at the numbers and assign weightings to various factors. I took the extreme of assigning all the weighting to one factor, which is legitimate (even if very narrowly focused) in its own right -- at least I try to think it is.

I didn't start this thread---but the person who did put in some work, which I appreciate. Do I have some issue with it --for sure! But for every list of anything you can make up millions of versions using different inputs to be mind boggling. Again, not being nasty but if you have another way that makes you feel good--then put you list on Boneyard as a thread-- It's so easy to point out other methods--so difficult to put a list together.

 

[Bayareahusky]

Wow this is really stirred up. I personally do not mind different opinions..... and as I have said congrats to the Lady Vols for their program history as evidenced by Alex's figures.

I certainly do not think UConn needs any words to support their body of work. Tennessee has fallen a bit as we all know. They can still hold on to Pat's 1098 wins... and little else... even the attendance supremacy was taken away by South Carolina this past year.

We will always hold ourselves to a higher standard..... that is so clear. Tennessee has a great history. The formula cited by Alex doesn't shake me at all. UConn still rocks.

 

[BroadwayVa]

Good suggestion.

A bit labor-intensive as compared with taking partisan potshots or (more constructively) suggesting tweaks to the present methodology.

But good suggestion, even if meant partly as rhetorical rebuke.

It isn't easy to put these things together. I'm reminded of how vowelguy developed a system for ranking recruiting classes and had to endure a lot of what he may have regarded as ungrateful sniping. He's been quiet for a couple of months now, hopefully preparing a splashy re-entrance with new multi-colored charts.

I agree!!!! Since we are not charged for these charts and lists we have to know that the people who put these together put some thought and effort into compiling them. And like ALL of the PRo lists these lists are based on the builders opinions and feelings plus maybe some mathematical skills. I liked this list--I learned some things I forgot or didn't know--is this the ultimate list--most likely not. I thank the author for his/her work.

 

[BroadwayVa]

My statistical analysis: 100 points for winning a national championship, 0 points for anything else. Guess who comes out on top? Just saying.

0How about 90 points for any team with DT, Maya, Stewie, Anne, BT, Sue, Swin, Asyia or Uconn across it's chest and 10 points for everyone else.
But please don't ask me to compile that list!!!!

 

[Wally East]

Tennessee has been very, very good since the beginning of the NCAA period, 1982. UConn started being very, very good much later -- you could pick 1991 or 1994 or 2000. That is what's demonstrated in this statistical analysis.

Yes, obviously UConn has 10 titles. While UConn was collecting those titles, Tennessee was still performing at a very high level, including winning additional titles of its own. Even now, they're still performing at a pretty high level -- better than UConn performed in the 1980s, for example.

A season where you don't win a title isn't a total loss. Everyone was proud of UConn's 1991 trip to the Final Four, likewise in 1996. Those seasons had value. Wasn't everyone excited about the return to the Final Four in 2008?

Saying that only seasons that result in a title have value isn't a great way to look at things (which is different than saying a particular team going into a season is a favorite to win a title).

 

[UcMiami]

It's always fun to play with numbers. What logic did you use to come up with your point system and if things were worth different amounts would that change the rankings? I know in your system uconn would lose points if elite 8 was 30 final 4 40 and title 50 instead of 70. Is there also away to reward consistency? For example 5 consecutive sweet 16's would be worth something. Uconn's 8 straight final 4's or 3 straight titles have value. Just a thought.

Again - the issue is that Uconn was a black hole until 1991 - two tournament appearances and two tournament losses with zero wins. So for the first 9 years of the NCAA Uconn's points accumulated were basically zero - they had one top 4 seed.
During that same time period TN accumulated:
2 NC titles
1 NC runner-up
3 FF
2 Elite eight
1 Sweet sixteen
That is a lot of points to catch up on - and does not included points for regular season (seeding in the tournament) which they accumulated in each of those years.
They currently still own:
The longest string of continuous sweet sixteen appearances and the most in total (33 out of a total of 34)
The most NC game appearances (13)
The most FF appearances (18)

It is only in the last 8 years that Uconn has started to close the gap - 8 FF to zero, 5 NCs to zero. And more #1 seeds (8-1?)

Before this current string Uconn and TN from 1995-2008 were running pretty closely TN 5 NCs Uconn 5 NCs, etc. Prior to 1995 Uconn just wasn't in the same league.

 

[UcMiami]

In thinking about pre NCAA season, one possible valuation or added valuation would be 20 win seasons, 25 win seasons, 30 win seasons, zero loss seasons. But ... then you get into all sorts of issues about strength of schedule and strength of conference. I actually think 'seeding' is a pretty good evaluation - not perfect, but a subjective valuation based on a committee looking at all those values - SOS, Wins/Losses, Good Wins/Bad Losses, etc. Is it perfect, no. But in the top four seeds, I can't think of any real disputes that were more than one seed line off - 'should have been a two seed, not a three seed' kind of issues. And for the most part individual teams have not been consistently the source of those complaints over multiple years.

Uconn since 2000 has been brilliant - 9 NC, 4 Undefeated seasons, two mammoth winning streaks, 5 and 8 year runs in the FF. No other team comes close nor has another team had a run of 16 years that in anyway compares. I am really really happy about that! 1995 was a super turning point for the program and the origin of that brilliant run. But I also can easily recognize that TN has been exceptionally good for a much longer period of time - never as dominant - but consistent. And that excellence didn't start in 1982, but stretches back into the AIAW era where they were in the FF in 77, 79, 80, and 81. And 33 of 34 NCAA sweet sixteens is an impressive run of consistency.

 

[KnightBridgeAZ]

You miss one of the great early knocks on statisticians that I think goes back to Ambrose Bierce. It was noted that if one stands with one foot on the stove and one foot in the freezer, the statistician would say that, on average, you were comfortable.

 

[01Neaggrad02]

I'm not sure how you say that the team that has won the tournament the most times isn't the best team during the tournament era. Especially when that team is 4-0 in the championship game against the team you have 1st.

Add to that 5 undefeated seasons(2 back to back)
90 game win streak
70 game win streak
3-peat... twice
8 consecutive final fours(and counting)
etc...

This is what I am talking about. However I am delusional? My point is valid that data and statistics are often skewed. Can anyone prove that differently?

 

[Waquoit]

YUK---this wasn't funny in 1939 when I first hear it.

Yeah, but did you know that old math teachers never die? They just reduce to lowest terms.

I got a million of 'em!

 

[Bestiarius]

in my experience the only people who love these jokes are the ones who are deficient in math skills. I think that the work done was great. Not perfect, but neither is life.

My purpose in providing those math jokes was not to denigrate Alex's work. Just providing a bit of levity to what was becoming contentious.

What I did have in mind was happiness to see that that those results were allowed to appear and stay on this site. If the numbers were reversed and UConn were shown to be #1, they would not be allowed to appear on sites of orange persuasion, let alone be supported by members.

The value of Alex's work is that it gives UConn a goal to shoot for. There is so little else available to keep up the motivation.

 

[KnightBridgeAZ]

This is what I am talking about. However I am delusional? My point is valid that data and statistics are often skewed. Can anyone prove that differently?

You are not wrong in principle. But the problem that most often occurs, and of which I thought you were guilty, is trying to take a study and make it mean something that it doesn't. It is done all the time, especially in politics.

In this particular case, over the history of the NCAA tournament, measured by the parameters given, Tennessee has been more successful than any other team. Each decade taken separately with the same parameters, would have different results, for example. Later decades would clearly favor UConn; the first decade taken by itself wouldn't mention UConn in the top 10.

It has little to do with the fact that UConn may have had the 4 best seasons ever; may have been more dominant in a given period than Tennessee ever was, may pass every test to be the most notable program in the history of the sport. None of that makes the study skewed or wrong - it isn't a study to show any of those things, in fact I assume when Alex started it he didn't know (exactly) how it would come out. When you want to show a preconceived result, that does often skew the data, because you choose the terms and conditions that will produce the result you want. Kudos to Alex for not doing so!!!

 

[01Neaggrad02]

Should schools receive points for the number of WNBA they have produced if we are talking about the success of the overall program? UCONN has had 30 different players play in the WNBA. This does not count Shea who was drafted but unable to ever play because of her ACL tears. Also that they developed the Liberty around Lobo and the Miracle around Sales is worth noting.
We also have at least 12 Olympians... and some of those have played in multiple games. DT and Sue are going for their forth.

Should some points be added there?

 

[Suebird10fan]

What it comes down to is you can massage numbers any way you want, but if you are looking at the full history of the NCAA's the approximately 1/3 of that period where Uconn was nonexistent by any measure is a currently insurmountable 'penalty' when comparing to a team that was very good for the whole time period. The only way to come to a different result would be to so overweight NCs as to make the rest of the 'quality' numbers meaningless - in which case, there is no need for analysis - you just count NCs.
The fact is TN has played in more NC games, more final four games, more elite eight games, etc. They just have not won more NC games.

Shorten the time frame to 1990 forward and you will get a different result, but then you will not be analyzing 'the NCAA period' but an arbitrary time frame based on 'Uconn becoming relevant'.

It's not arbitrary if you're interested in the Geno era. It's just a different question.