March Madness 2022

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 16th, 2022 at 7:33:35 AM permalink

Quote: lilredrooster
first, I would like to say, this was a really great post on your part.
link to original post

Thank you!

Here is another bet with a perceived big player advantage. Over/under total wins by #11 seed teams.

Number	Pays	Probability	EV
Over 3.5	100	0.242424	-0.515152
Under 3.5	-130	0.757576	0.340327

In the 36 seasons of the tournament thus far, 11 seeds have won 90 games, for an average of exactly 2.5 per year.

Using the Poisson function, the probability of 0 to 3 wins, assuming a mean of 2.5, is 75.76%. That makes laying -130 a great bet, with a 34.03% player advantage.

In case you're wondering, the 11 seeds are Virginia Tech, Michigan, Iowa State, and Notre Dame.

My question -- Is this a good bet or are the 11 seeds this year stronger than usual?

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

DRich

DRich

Threads: 86
Posts: 11709

Joined: Jul 6, 2012

March 16th, 2022 at 10:33:46 AM permalink

Quote: Wizard

In case you're wondering, the 11 seeds are Virginia Tech, Michigan, Iowa State, and Notre Dame.

My question -- Is this a good bet or are the 11 seeds this year stronger than usual?
link to original post

My opinion is that the 11 seeds may be a little bit stronger than average but also in general I believe there is more parity in the field this year.

At my age, a "Life In Prison" sentence is not much of a deterrent.

ThatDonGuy

Threads: 117
Posts: 6267

Joined: Jun 22, 2011

March 16th, 2022 at 10:55:10 AM permalink

Quote: Wizard

In case you're wondering, the 11 seeds are Virginia Tech, Michigan, Iowa State, and Notre Dame.

You forgot Rutgers - and Rutgers and Notre Dame play each other in Dayton two days before the winner plays Alabama in San Diego.

Wizard
Administrator

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 16th, 2022 at 11:37:52 AM permalink

Quote: ThatDonGuy
You forgot Rutgers - and Rutgers and Notre Dame play each other in Dayton two days before the winner plays Alabama in San Diego.
link to original post

Thanks. That was more of a typo than forgetting.

I just came from the Rampart, which uses South Point lines. The only prop that wasn't some kind of futures bet was on the seed to win the whole enchilada. Here they those bets. The probabilities are based on the 36 previous seasons.

Seed	Pays	Probability	EV
1	-150	0.638889	0.064815
2	200	0.138889	-0.583333
3	375	0.111111	-0.472222
4	900	0.027778	-0.722222
5	1200	0.000000	-1.000000
6 to 16	1000	0.083333	-0.083333

Only perceived advantage is on a #1 seed to win, but we've already covered that the #1 seeds are softer than usual this year.

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

Monkmister1

Threads: 0
Posts: 9

Joined: Mar 13, 2022

March 16th, 2022 at 12:47:38 PM permalink

Just saying the #1 seeds are weaker does not make it so.

SOOPOO

Threads: 122
Posts: 10992

Joined: Aug 8, 2010

March 16th, 2022 at 1:04:40 PM permalink

Quote: Monkmister1
Just saying the #1 seeds are weaker does not make it so.
link to original post

I think if you ask any CBB expert, they would tell you the difference between the top four teams and the next eight is much smaller than most years.

I�m guessing <5 seeds 11 and higher win in the round of 64, and <2 make it to the sweet 16. I�m betting the chalk this year! Exceptions are San Francisco and Alabama.

mwalz9

mwalz9

Threads: 52
Posts: 754

Joined: Feb 7, 2012

March 16th, 2022 at 1:31:14 PM permalink

I like Chattanooga as a first round upset as a 13 seed. I'll be betting money line on that one.

I like Georgia St +23.5 tomorrow early.

I placed a prop for Jared Bynum to score 20+ and Providence to win tomorrow in that game.

I also have a +700 future on Arizona to win it all.

That's all I looked at so far. I do plan to be at the local brick and mortar sportsbook all day tomorrow and bet there or on FanDuel comparing lines and placing my wager where the line is most beneficial to me if there is a difference.

Monkmister1

Threads: 0
Posts: 9

Joined: Mar 13, 2022

March 16th, 2022 at 3:56:12 PM permalink

Quote: SOOPOO
Quote: Monkmister1
Just saying the #1 seeds are weaker does not make it so.
link to original post

I think if you ask any CBB expert, they would tell you the difference between the top four teams and the next eight is much smaller than most years.

I�m guessing <5 seeds 11 and higher win in the round of 64, and <2 make it to the sweet 16. I�m betting the chalk this year! Exceptions are San Francisco and Alabama.
link to original post

Monkmister1

Threads: 0
Posts: 9

Joined: Mar 13, 2022

March 16th, 2022 at 3:57:44 PM permalink

Being new here I will respectfully say your definition of expert and mine differ greatly.

Wizard
Administrator

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 16th, 2022 at 7:43:07 PM permalink

At this time of year I usually spend way too much time analyzing March Madness. Case in point is I never actually bet anything. I just find it a statistical challenge.

The bet at the Rampart on a #5 seed has been bothering me since my last post. A #5 seed has never won. Thus, in my last post, I gave such an event a probability of 0. However, that is surely not correct.

To remedy that, I wrote a program to simulate the entire tournament, billions of times. I won't get into how I modeled it, unless somebody asks.

Here is how often each seed one in my simulation and probability.

Seed	Winner	Probability
1	2,354,770,448	0.418368
2	1,140,618,554	0.202652
3	786,974,307	0.139820
4	437,521,459	0.077734
5	407,525,253	0.072404
6	231,009,077	0.041043
7	133,987,286	0.023805
8	71,534,125	0.012709
9	36,423,246	0.006471
10	17,403,682	0.003092
11	7,293,396	0.001296
12	2,836,470	0.000504
13	448,739	0.000080
14	109,106	0.000019
15	11,516	0.000002
16	632	0.000000
Total	5,628,467,296	1.000000

Putting those probabilities in the Rampart props on the seed to win, gives the following updated table.

Seed	Pays	Probability	EV
1	-150	0.418368	-0.302720
2	200	0.202652	-0.392045
3	375	0.139820	-0.335853
4	900	0.077734	-0.222663
5	1200	0.072404	-0.058744
6 to 16	1000	0.089022	-0.020758

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

gordonm888
Administrator

Threads: 60
Posts: 5043

Joined: Feb 18, 2015

March 17th, 2022 at 6:58:21 AM permalink

2022 is the first year after the rule change that all players are allowed to transfer schools once without any restrictions. This, in combination with the transfer portal has changed everything. At the end of last season, student athletes took advantage of this to transfer in huge numbers, i.e., many good players transferring with the objective of getting playing time because they were stuck on the bench behind other good players

So a team like Texas Tech (a 3 seed) has 8 transfers (players who transferred from other schools last spring) including either 4 or all 5 starters! The Kentucky team has a large number of transfers from other schools - I think 6 of their top 8 players.

The net result in 2022 has been COMPRESSION - a trend towards equalizing the quality of teams. Many, many commentators have said that there is no dominant team and the difference in 'power' between the one seeds and the five seeds is less than it has ever been. Many analysts have said that there are about 15 teams that have legitimate chances of winning the entire tournament.

So, this year is historic for identifiable reasons, and analyses based on how different seeds have performances in the past would have a built-in defect.

So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.

gordonm888
Administrator

Threads: 60
Posts: 5043

Joined: Feb 18, 2015

March 17th, 2022 at 6:58:23 AM permalink

2022 is the first year after the rule change that all players are allowed to transfer schools once without any restrictions. This, in combination with the transfer portal has changed everything. At the end of last season, student athletes took advantage of this to transfer in huge numbers, i.e., many good players transferring with the objective of getting playing time because they were stuck on the bench behind other good players

So a team like Texas Tech (a 3 seed) has 8 transfers (players who transferred from other schools last spring) including either 4 or all 5 starters! The Kentucky team has a large number of transfers from other schools - I think 6 of their top 8 players.

The net result in 2022 has been COMPRESSION - a trend towards equalizing the quality of teams. Many, many commentators have said that there is no dominant team and the difference in 'power' between the one seeds and the five seeds is less than it has ever been. Many analysts have said that there are about 15 teams that have legitimate chances of winning the entire tournament.

So, this year is historic for identifiable reasons, and analyses based on how different seeds have performed in the past would have a built-in defect.

So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.

charliepatrick

charliepatrick

Threads: 39
Posts: 2946

Joined: Jun 17, 2011

March 17th, 2022 at 3:22:07 PM permalink

Quote: Wizard
...Total #1 seeds to make Final Four

Over 1.5 +120
Under 1.5 -110...link to original post

Perhaps I've misread it and if I've read this right Over, is 6/5 and Under is 10/11; isn't this overbook. Just bet $115.24 on Under and $100 on Over (or similar) to make money.

I'm just wondering whether if they meant -120 and +110 and it was 54.15% chance that makes sense; it would mean a HE of 0.87% and 3.715% (to the bookie on the fav and outsider).

Wizard
Administrator

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 17th, 2022 at 7:36:24 PM permalink

Quote: charliepatrick
Perhaps I've misread it and if I've read this right Over, is 6/5 and Under is 10/11; isn't this overbook. Just bet $115.24 on Under and $100 on Over (or similar) to make money.

I'm just wondering whether if they meant -120 and +110 and it was 54.15% chance that makes sense; it would mean a HE of 0.87% and 3.715% (to the bookie on the fav and outsider).
link to original post

You're right. I made a typo in posting the lines. Now that the tournament has started, I can't go back and see the original lines. However, they usually had 30-cent lines on their prop bets, so it probably was something like -140 and +110.

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

Wizard
Administrator

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 18th, 2022 at 7:57:50 AM permalink

I just created a new page on March Madness. I welcome all comments.

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

SOOPOO

Threads: 122
Posts: 10992

Joined: Aug 8, 2010

March 18th, 2022 at 8:47:43 AM permalink

Quote: Wizard
I just created a new page on March Madness. I welcome all comments.
link to original post

I don't mean to be petty, but you start with a major error. In NO WAY is the tournament made up of the top 68 college basketball teams. Since each conference is guaranteed at least one participant, some of the lower level conferences have NO teams that would we considered a top 68 team, maybe there are 12 (guess) such conferences or so. So there are 12 teams NOT amongst the top 68 that make the tournament, and 12 teams that ARE amongst the best 68 that do NOT make the tournament.

Edit... You also left out a word in point 13.

ThatDonGuy

Threads: 117
Posts: 6267

Joined: Jun 22, 2011

March 18th, 2022 at 10:07:33 AM permalink

Quote: Wizard
I just created a new page on March Madness. I welcome all comments.
link to original post

1. As already has been pointed out, it is not the "best 68 teams," but the winners of the 32 Division I conference championships plus the best 36 remaining teams.

2. It is not necessarily 17 in each region. The tournament committee can put two, or even three, of the "First Four" games in the same region. It cannot do all four as two of them must be for one of the four #16-seed spots.

3. The teams are ranked before the First Four games. The winners of the two First Four games that do not involve the four weakest teams in the tournament are pre-assigned a seed in one of the regions. In 2022, one of the games was for an 11 seed, and the other was for a 12.

4. The Final Four matchups are not the same, in terms of regions, each year. The tournament committee ranks the four #1 seeds at the start of the tournament. The winner of the region with the #1 overall seed will play the winner of the region with the #4 overall seed, and the winner of the region with the #2 overall seed will play the winner of the region with the #3 overall seed.

5. The tournament committee is allowed to move a team up or down one seeding level (in "extraordinary circumstances," two levels) if necessary to maintain the "seeding principles":
Teams from the same conference cannot be placed in the bracket so that they would play in the first round, or in the second round if they have played each other twice so far in the season (including the conference championship tournament), or in the third round if they have played each other three or more times in the season.
Teams that have played each other this season "should not" play each other in the First Four or the first round.

billryan

billryan

Threads: 240
Posts: 16282

Joined: Nov 2, 2009

March 18th, 2022 at 10:22:33 AM permalink

That seems incredibly over-complicated. I can see trying to avoid 1st round matchups between teams that have played each other two or three times previously, but having to avoid second and third-round matchups seems nightmarish. I assumed they simply went by RPI.

The difference between fiction and reality is that fiction is supposed to make sense.

ThatDonGuy

Threads: 117
Posts: 6267

Joined: Jun 22, 2011

March 18th, 2022 at 10:58:26 AM permalink

Quote: billryan
That seems incredibly over-complicated. I can see trying to avoid 1st round matchups between teams that have played each other two or three times previously, but having to avoid second and third-round matchups seems nightmarish. I assumed they simply went by RPI.
link to original post

They don't use RPI any more. Starting in 2019, the NCAA switched to something developed in house called NET.

NET uses two metrics: "team value," which is based on who you played, where (away / home / neutral), and who won, although when you played does not matter (a game played early in the season counts the same as one played the day before Selection Sunday), and "net efficiency", which is points scored per possession minus points given up per opponent's possession - well, sort of; "possessions" is not a statistic, so it is estimated as shots taken minus offensive rebounds plus turnovers plus 0.475 x free throws taken. There used to be two other metrics - margin of victory, capped at 10 points per non-overtime game and 1 point per overtime game, and win percentage, adjusted to count games won by the visiting team as 1.4 and won by the home team as 0.6 - but they were dropped after the 2018 season. Just how team value is calculated appears to be as closely guarded a secret as a PAR sheet.

billryan

billryan

Threads: 240
Posts: 16282

Joined: Nov 2, 2009

March 18th, 2022 at 11:51:09 AM permalink

I guess putting the best team first and the worst team last wouldn't justify the seeding department's budget.

The difference between fiction and reality is that fiction is supposed to make sense.

gordonm888
Administrator

Threads: 60
Posts: 5043

Joined: Feb 18, 2015

March 18th, 2022 at 8:38:05 PM permalink

Quote: ThatDonGuy
Quote: billryan
That seems incredibly over-complicated. I can see trying to avoid 1st round matchups between teams that have played each other two or three times previously, but having to avoid second and third-round matchups seems nightmarish. I assumed they simply went by RPI.
link to original post

They don't use RPI any more. Starting in 2019, the NCAA switched to something developed in house called NET.

NET uses two metrics: "team value," which is based on who you played, where (away / home / neutral), and who won, although when you played does not matter (a game played early in the season counts the same as one played the day before Selection Sunday), and "net efficiency", which is points scored per possession minus points given up per opponent's possession - well, sort of; "possessions" is not a statistic, so it is estimated as shots taken minus offensive rebounds plus turnovers plus 0.475 x free throws taken. There used to be two other metrics - margin of victory, capped at 10 points per non-overtime game and 1 point per overtime game, and win percentage, adjusted to count games won by the visiting team as 1.4 and won by the home team as 0.6 - but they were dropped after the 2018 season. Just how team value is calculated appears to be as closely guarded a secret as a PAR sheet.
link to original post

To be clear, the NCAA Selection Committee doesn't pick the 32-at-large teams by their NET rating - not even close. Rhode Island made the tournament and they had a NET ranking of 77; whereas Oklahoma had a NET ranking of 34 and was not selected for an at-large bid. The selection process is often described by outsiders as "flawed."

Indeed, the NCAA does not claim to seek to pick the "best" teams for the 32 at-large tournament slots. Instead, they seek to pick the 32 teams who have "achieved the most", based on the entirety of their season, i.e., if a team loses its first 12 games and then improves dramatically during the season and has gone undefeated for the last 20 games that is not weighed any differently than a team that won their first 20 games and then fell apart and lost their last 12 games. That's an extreme example, but I think it provides the gist of the selection rule.

So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.

ThatDonGuy

Threads: 117
Posts: 6267

Joined: Jun 22, 2011

March 19th, 2022 at 6:29:23 AM permalink

Quote: gordonm888
To be clear, the NCAA Selection Committee doesn't pick the 32-at-large teams by their NET rating - not even close. Rhode Island made the tournament and they had a NET ranking of 77; whereas Oklahoma had a NET ranking of 34 and was not selected for an at-large bid. The selection process is often described by outsiders as "flawed."

Indeed, the NCAA does not claim to seek to pick the "best" teams for the 32 at-large tournament slots. Instead, they seek to pick the 32 teams who have "achieved the most", based on the entirety of their season, i.e., if a team loses its first 12 games and then improves dramatically during the season and has gone undefeated for the last 20 games that is not weighed any differently than a team that won their first 20 games and then fell apart and lost their last 12 games. That's an extreme example, but I think it provides the gist of the selection rule.
link to original post

First of all, and pardon me for being pedantic (as usual), but there are 36 at-large teams.

Second, the "Principles and Procedures for Establishing the Bracket" says, "The committee selects the 36 best teams not otherwise automatic qualifiers for their conference to fill the at-large berths," so, in effect the NCAA does claim to seek to pick the "best" teams for the at-large slots. What it doesn't do is, it doesn't tell the committee members how to determine the best teams. (In the men's ice hockey tournament, on the other hand, the at-large selection process is entirely objective.) I was under the impression that the committee was supposed to choose the 36 teams that they thought were the best "at the moment," without necessarily taking how well they played earlier in the season into account. For example, a team that was weak early in the season because a key player was injured/suspended but played better late is given more consideration than one that was strong early on but then got weaker as the season progressed.

gordonm888
Administrator

Threads: 60
Posts: 5043

Joined: Feb 18, 2015

March 19th, 2022 at 8:37:25 AM permalink

Quote: ThatDonGuy
Quote: gordonm888
To be clear, the NCAA Selection Committee doesn't pick the 32-at-large teams by their NET rating - not even close. Rhode Island made the tournament and they had a NET ranking of 77; whereas Oklahoma had a NET ranking of 34 and was not selected for an at-large bid. The selection process is often described by outsiders as "flawed."

Indeed, the NCAA does not claim to seek to pick the "best" teams for the 32 at-large tournament slots. Instead, they seek to pick the 32 teams who have "achieved the most", based on the entirety of their season, i.e., if a team loses its first 12 games and then improves dramatically during the season and has gone undefeated for the last 20 games that is not weighed any differently than a team that won their first 20 games and then fell apart and lost their last 12 games. That's an extreme example, but I think it provides the gist of the selection rule.
link to original post

First of all, and pardon me for being pedantic (as usual), but there are 36 at-large teams.

Second, the "Principles and Procedures for Establishing the Bracket" says, "The committee selects the 36 best teams not otherwise automatic qualifiers for their conference to fill the at-large berths," so, in effect the NCAA does claim to seek to pick the "best" teams for the at-large slots. What it doesn't do is, it doesn't tell the committee members how to determine the best teams. (In the men's ice hockey tournament, on the other hand, the at-large selection process is entirely objective.) I was under the impression that the committee was supposed to choose the 36 teams that they thought were the best "at the moment," without necessarily taking how well they played earlier in the season into account. For example, a team that was weak early in the season because a key player was injured/suspended but played better late is given more consideration than one that was strong early on but then got weaker as the season progressed.
link to original post

Thanks for your correction viz 36 teams.

I based my post after watching many interviews of the Selection Committee Chairman and incoming Committee Chairman who explicitly said that achievements in the early season were weighed as much as achievement recently. Many of the interviewers accused the chairman of ignoring
- what happens in the conference tournament, and
- ignoring what happens in the last two weeks before the bracket is announced (because the final bracket was identical to the bracket that the selection committee announced two weeks earlier except for the automatic qualifiers -who had to be included because they had won conference tournaments).

The Selection Committee Chairman repeatedly stated that the criteria was a team's accomplishments over the totality of their season. I cannot explain why the statement of principles say something different.

I am aware of the exceptions due to injury of a significant player or due to covid absences* - but when a team simply improves during a season, e.g. freshmen point guards stop turning the ball over and a dominating interior player emerges, - NCAA representatives seem to insist that that is not taken into account. I heard "totality of the season" until my ears bled.

* The statement that they considered covid absences appears to many observers to just be posterior protection. Committee representatives never cited examples of how this affected decisions, and many of us can cite counter-examples where it doesn't seem to have been considered.

Professional sports gamblers seek to exploit weaknesses in betting lines especially when those weaknesses arise from inefficiencies in the real-world seeding process. So, it helps to look beyond the bureaucratic "Vision statements" of an organization and listen to what their representatives claim that they are doing when challenged by the media.

So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.

avianrandy

avianrandy

Threads: 8
Posts: 1551

Joined: Mar 7, 2010

March 19th, 2022 at 5:43:34 PM permalink

Ok show of hands. Who had the st Peter's peacocks knocking off Kentucky wildcats?

Wizard
Administrator

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 20th, 2022 at 6:51:55 AM permalink

Thank you for the corrections on the selection process. I decided to just leave out words like "best" and "top."

The average seed in the second round is exactly 6. The average in previous years was 5.83.

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

Wizard
Administrator

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 21st, 2022 at 11:02:30 AM permalink

Quote: Wizard
The average seed in the second round is exactly 6. The average in previous years was 5.83.
link to original post

The average seed in the third round is 5.31. Not including 2022, the previous average in round 3 is 4.55. So the low-seeded teams are exceeding expectations a little.

The news is making a big deal about a 15-seed making it this far, but this happened last year as well as Oral Roberts made it to the Sweet 16. Historically, in the 36 seasons played, 15 seeds have made the Sweet 16 three times.

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

Ace2

Ace2

Threads: 32
Posts: 2672

Joined: Oct 2, 2017

March 21st, 2022 at 11:06:22 AM permalink

Quote: avianrandy
Ok show of hands. Who had the st Peter's peacocks knocking off Kentucky wildcats?
link to original post

No one except maybe some St P alum

It�s all about making that GTA

SOOPOO

Threads: 122
Posts: 10992

Joined: Aug 8, 2010

March 21st, 2022 at 12:20:49 PM permalink

Quote: Wizard

The average seed in the third round is 5.31. Not including 2022, the previous average in round 3 is 4.55. So the low-seeded teams are exceeding expectations a little.

I'll repeat myself. Lower seeded teams may be meeting expectations, NOT necessarily exceeding them. Since for all the reasons I and others have mentioned many times there is more parity than in the years you derive your 4.55 number, it would have been a surprise to NOT see the number higher than that 4.55. Maybe not as high as 5.31, but the landscape of NCAA BBall has changed.

By the way, if you had been fading my picks, you could now be on your private plane to your luxury villa in the Grand Cayman islands....
If I was a tout I could see my commercial.... VERIFIED.... PROVEN..... INDISPUTABLE.... 22% win rate versus the spread...

gordonm888
Administrator

Threads: 60
Posts: 5043

Joined: Feb 18, 2015

March 21st, 2022 at 1:13:07 PM permalink

Quote: SOOPOO
Quote: Wizard

The average seed in the third round is 5.31. Not including 2022, the previous average in round 3 is 4.55. So the low-seeded teams are exceeding expectations a little.

I'll repeat myself. Lower seeded teams may be meeting expectations, NOT necessarily exceeding them. Since for all the reasons I and others have mentioned many times there is more parity than in the years you derive your 4.55 number, it would have been a surprise to NOT see the number higher than that 4.55. Maybe not as high as 5.31, but the landscape of NCAA BBall has changed.
-snip-
link to original post

I strongly agree with Soopoo. Given that players had greatly increased mobility awarded to them last year and transferred in large numbers, perhaps the slant of your article might be the quantification of the effect of increased player mobility on the competitiveness of the lower seeds (using the historical analysis that you have done as a basis for comparison.)

So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.

gordonm888
Administrator

Threads: 60
Posts: 5043

Joined: Feb 18, 2015

March 21st, 2022 at 1:36:12 PM permalink

By the way, you might establish the range of possible outcomes by round by taking a case where the lower(higher?) seeds (better teams) always win and another case where the results of all games are 50/50 coin flips because the seeds are meaningless.

If the seeds were meaningless then the average rank of winners would be 8.5 for all rounds of the tournament.

If the seeds were 100% determinative then the average rank of winners would be:

Rnd of 32: 4.5
Sweet sixteen: 2.5
Elite 8: 1.5
Final 4 and beyond: 1

So, the shift in the average rank of winners in the Sweet Sixteen from 4.55 to 5.31 should be considered in the context that the full range of possibilities between determinism and randomness is only 2.5 - 8.5. It's an interesting theoretical problem to try to come up with the significance of such a shift in the predictive significance of the seedings, but I think your comment about "exceeding expectations a little" may not be proportional to what has occurred.

Just trying to help you think through an interesting phenomenon.

So many better men, a few of them friends, are dead. And a thousand thousand slimy things live on, and so do I.

unJon

unJon

Threads: 14
Posts: 4592

Joined: Jul 1, 2018

March 21st, 2022 at 2:22:04 PM permalink

The seed data impact on probability of victory should also show heteroscedasticity. It will be right at the break point between the last �at large� bid and the rest of the conference winners that would not have gotten an �at large� bid.

The race is not always to the swift, nor the battle to the strong; but that is the way to bet.

Wizard
Administrator

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 21st, 2022 at 2:38:38 PM permalink

Quote: SOOPOO
I'll repeat myself. Lower seeded teams may be meeting expectations, NOT necessarily exceeding them. Since for all the reasons I and others have mentioned many times there is more parity than in the years you derive your 4.55 number, it would have been a surprise to NOT see the number higher than that 4.55. Maybe not as high as 5.31, but the landscape of NCAA BBall has changed.
link to original post

I don't dispute any of that. Do you dispute my figures?

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

Monkmister1

Threads: 0
Posts: 9

Joined: Mar 13, 2022

March 21st, 2022 at 2:39:41 PM permalink

Quote: SOOPOO
Quote: Monkmister1
Just saying the #1 seeds are weaker does not make it so.
link to original post

I think if you ask any CBB expert, they would tell you the difference between the top four teams and the next eight is much smaller than most years.

I�m guessing <5 seeds 11 and higher win in the round of 64, and <2 make it to the sweet 16. I�m betting the chalk this year! Exceptions are San Francisco and Alabama.
link to original post

Monkmister1

Threads: 0
Posts: 9

Joined: Mar 13, 2022

March 21st, 2022 at 2:41:30 PM permalink

And yet 3 #1 seeds ate in the sweet 16. I wonder how the CBB experts explain that?

avianrandy

avianrandy

Threads: 8
Posts: 1551

Joined: Mar 7, 2010

March 21st, 2022 at 2:45:09 PM permalink

Quote: Ace2
Quote: avianrandy
Ok show of hands. Who had the st Peter's peacocks knocking off Kentucky wildcats?
link to original post
No one except maybe some St P alum
link to original post

I believe that would be the #15 seed the news is talking about. As a resident of a state that borders Kentucky sometimes I hear about it from Kentucky fans from whatever sport.not really a basketball fan but I would like to see Duke go all the way

SOOPOO

Threads: 122
Posts: 10992

Joined: Aug 8, 2010

March 21st, 2022 at 2:54:53 PM permalink

Quote: Wizard
Quote: SOOPOO
I'll repeat myself. Lower seeded teams may be meeting expectations, NOT necessarily exceeding them. Since for all the reasons I and others have mentioned many times there is more parity than in the years you derive your 4.55 number, it would have been a surprise to NOT see the number higher than that 4.55. Maybe not as high as 5.31, but the landscape of NCAA BBall has changed.
link to original post

I don't dispute any of that. Do you dispute my figures?
link to original post

I do not dispute any of your figures. My point is that using pre transfer portal and 'NIL payment' years to try and figure out what will happen this year is fraught with error.

SOOPOO

Threads: 122
Posts: 10992

Joined: Aug 8, 2010

March 21st, 2022 at 3:01:31 PM permalink

Quote: Monkmister1
And yet 3 #1 seeds ate in the sweet 16. I wonder how the CBB experts explain that?
link to original post

Silly question! Regardless of how much better or worse the number one seeds are compared to previous years, those were the 4 best teams out of the 68 selected. One is gone. Overall top pick Gonzaga trailed by 10 at halftime. Arizona needed OT to win its second round match. Kansas struggled against Creighton. These 3 number one seeds that DID advance did not play a team that would have been considered a top 25 team before the tourney began (not fact checked!, but I think...)

Wizard
Administrator

Threads: 1493
Posts: 26485

Joined: Oct 14, 2009

March 21st, 2022 at 6:41:03 PM permalink

Quote: SOOPOO
I do not dispute any of your figures. My point is that using pre transfer portal and 'NIL payment' years to try and figure out what will happen this year is fraught with error.
link to original post

I was just throwing out a truthful statistic.

"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)

smoothgrh

smoothgrh

Threads: 87
Posts: 1287

Joined: Oct 26, 2011

March 21st, 2022 at 7:56:49 PM permalink

Wizard, thanks for your newsletter note about the lazy journalistic statistics published about a "perfect bracket" being equivalent to the cumulative total of coin flips.

Obviously not all games are a coin flip. I look forward to your next newsletter with updated stats.

As an aside, those of us in the newsroom back in 1998 watched the first-ever #16 seed upsetting a #1 seed (Allison Feaster and the Harvard women upsetting the #1 Stanford women), and were cheering for this amazing statistical oddity!

Monkmister1

Threads: 0
Posts: 9

Joined: Mar 13, 2022

March 22nd, 2022 at 3:54:00 AM permalink

Would Coulda Shoulda. Luckily games are played on court with no ccb "experts" on the team.

SOOPOO

Threads: 122
Posts: 10992

Joined: Aug 8, 2010

March 22nd, 2022 at 5:27:59 AM permalink

I haven�t watched much college basketball the past couple of years. Having watched a lot this tourney, I was (unpleasantly) surprised at how frequently the refs are going to the TV monitor to see if the foul was flagrant, or who touched the ball last, or if the shooter�s foot was on the line, or if the shot clock had expired, or if the defender was in the restricted area, or�.

For those wanting to make money, my best bet for this round is Purdue -12.5. I can�t see them not scoring at will with their two big men. My predictions like this over the years tend to hit 49% of the time�. but I�m not doing nearly that well this year�

ThatDonGuy

Threads: 117
Posts: 6267

Joined: Jun 22, 2011

March 22nd, 2022 at 7:01:23 AM permalink

Quote: SOOPOO
I haven�t watched much college basketball the past couple of years. Having watched a lot this tourney, I was (unpleasantly) surprised at how frequently the refs are going to the TV monitor to see if the foul was flagrant, or who touched the ball last, or if the shooter�s foot was on the line, or if the shot clock had expired, or if the defender was in the restricted area, or�.
link to original post

There is a rather lengthy list of just what is reviewable. What almost always happens is, something happens during a tournament game that couldn't get reviewed (for example, a missed offensive basket interference call), and it gets added to the list the following year "to prevent it from happening again."

There are only two instances where they "have to" check the replay: when the half/overtime period ends (to see if either the shot was released before the time ran out, or the ball went into the basket before the time ran out and time has to be put back on the clock), and when there is a fight (to see which players are subject to the automatic one-game suspension for fighting). Things like checking to see if the shooter was behind the three-point line are at the discretion of the officials.

Also, there are only two instances where a coach can ask for a review: whether or not a foul was a Flagrant 1 or Flagrant 2, and whether or not a defender was in the No-Charge Arc when a foul underneath the basket was called (and for this one, only in the last two minutes of the second half or an overtime period).

ThatDonGuy

Threads: 117
Posts: 6267

Joined: Jun 22, 2011

March 23rd, 2022 at 5:36:20 PM permalink

Here's some Vegas-related March Madness news: South Point and Rampart (also Virgin River and Casablanca in Mesquite) are offering -105 point spread lines on all of the remaining games in the tournament

mwalz9

mwalz9

Threads: 52
Posts: 754

Joined: Feb 7, 2012

March 23rd, 2022 at 8:19:00 PM permalink

Quote: ThatDonGuy
Here's some Vegas-related March Madness news: South Point and Rampart (also Virgin River and Casablanca in Mesquite) are offering -105 point spread lines on all of the remaining games in the tournament
link to original post

The lines in line with other books or are the making you pay the vig in a half a point?

SOOPOO

Threads: 122
Posts: 10992

Joined: Aug 8, 2010

March 23rd, 2022 at 8:37:48 PM permalink

With all the profit boosts/free bets/ parlay insurance/odds boosts I have a few guaranteed win positions. A bunch of parlays which I�m scheduled to make a little on could be BIG if�.
I have Arkansas +9 on some and Gonzaga -8.5 on others.
And Michigan +5 on some and Villanova -4.5 on others. Wife can get a new pair of shoes If one of those games ends on the exact number. A purse if both do.

ThatDonGuy