Tracking the Bracket Matrix throughout the Regular Season | Page 2 | Syracusefan.com

Tracking the Bracket Matrix throughout the Regular Season

20Plus said:
Good work I appreciate it I don't know what's wrong with some of these people
Expand...

That was a while back and the person apologized. So all good... I can understand why people are upset with bubble talk again... this is Syracuse after all.

It's February I will probably ramp up my game tracking, but for now it will all be in this thread.
 
Last night the 16 teams around the bubble line went 2-5. So at this point it could mean the bubble is weakening or make it easier for teams to quickly shoot up. Our losses are depressing, and it makes it hard to believe we are still in this. But as of now we are right in the heart of it... again.

Now that is Feb 1, I will track the 8 teams in or around the bubble on both sides per the matrix:

upload_2018-2-1_17-58-32.png


SMU is the second last team in right now. They have some quality victories OOC in Arizona, USC, and Boise St, and they won at Wichita St. The new quadrant system may help a team like SMU because they have had losses at Uconn and Tulane, which may have been "bad" in the past as they are outside the top 100, but may be OK now as they may be within the top "135" which is the new guideline for a bad road loss.

upload_2018-2-1_18-3-54.png



Western Kentucky is the first team out right now. They have 3 quadrant one wins based on projected RPI (neutral Purdue, neutral SMU, at Old Dominion). Old Dominion is RPI #74, and this the new type of road win the committee will now apparently value.

And while the quadrant system raises the RPI for a bad loss on the road, it lowers the floor for a bad one at home. Based on projected RPI, Western Kentucky has 4 quadrant three or quadrant 4 losses. (Home Missouri St, Home #78 Belmont, at Ohio, at Wisconisn). The losses at home vs Missouri St or Belmont would not have been bad in the past but they are now considered as such.
The win over Purdue can only carry you so far.

They are also #46 in KP, which I think is will be a consideration for how they assess the mid majors going forward.
 
(Updated for before Virginia game)

For the first time in conference play, we are now out of the consensus bracket.

We are the first team out... we are still on 44% of the brackets, down from 84% prior to the Georgia Tech game. Things are quite fluid.

Since last game Syracuse and SMU have fell out, and Missouri and Washington have moved in. For Washington it is the first time all year they are in the tourney "as of now" per consensus. Washington moved up a bit after a nice win over Arizona St, and Missouri had a quadrant one win at Alabama.

It is quite tight around the bubble line:

(out of 79)
Viriginia Tech 48
Washington 44
Marquette 39
--------
Syracuse 35
Western Kentucky 35
Boise State 34
SMU 29

A today would shoot us comfortably back in the short term... and is of course essential we get a quality scalp for long term.



upload_2018-2-3_13-32-51.png


upload_2018-2-3_13-32-0.png
 
After the Virginia game / before the Louisville game.

Well, I will continue to do this through the regular season just because I enjoy the bubble whether Syracuse is in or not.

We are still right on the bubble "as of now". A quadrant one win tonight would put us back on the right side of the line in the eyes of many. So we don't have a hard task "record wise" to get in, but I, like most, find that even a modest record to the end of the season is hard to fathom at this point. I expect to slide off the bubble over the next two games, but who knows.

The lost to Virginia did not crush us, but it was a blown opportunity, and we need to somehow take advantage of a few quad one games down the stretch.

We have fell from 44% in, to 26% in after the Virginia game. However, we are still the second team out.

Since last week Marquette has moved out, and Boise St has moved into the tournament.

upload_2018-2-5_16-12-46.png


upload_2018-2-5_16-13-41.png
 
Last edited:
Let's talk #1 seeds. 3 teams seem to be in really good shape right now to get a #1 seed, but there is still time left. But Villanova and Virginia look fr real, and Purdue keeps in winning.

The race for the last #1 seed is wide open and quite shaky. Xavier who keep squeaking by in home games against mediocore teams does not seem like a #1. But who is it then? Kansas, Duke, or Auburn? Duke has had two bad losses. Kansas has a nice set of wins, but will pile up losses as well. Auburn has a great record in a very good SEC, but they still have not built up enough quality wins. I think Kansas being in the big 12 has the inside track on it if they perform.


The 3 seed line doens't look that inspiring either other than Michigan St because the team has talent.


As a side note I use bet365 which may not be available to Americans, but you can still get following odds on these current #2 seeds:
Auburn : 150-1
Clemson : 350-1 (the injury hurt them, but they are still playing well)

Unfortunately the max return is $2,500. If any of those two made the final four it is a great hedging opportunity.


upload_2018-2-5_16-16-24.png
 
TUESDAY KEY BUBBLE GAMES (Last 8 In, First 8 Out)

Note - I base the last 8 in based on # of brackets the are in, not average seed.

upload_2018-2-5_22-26-6.png


Arkansas is the 7th last team in per the matrix. A 4-6 team in the SEC being in, speaks to the bubble being as it normally has been in recent years (underwhelming), and the SEC being much stronger than usual. They have a neutral win over Oklahoma as well that really helps them.


upload_2018-2-5_22-50-31.png


Providence is the 5th last team in. Providence has some good home wins in Confrerence (Xavier, Creighton, Butler) and they beat Washington on a neutreal floor. But they lost to UMass (projected sub 200 RPI). Go Hoyas :(

upload_2018-2-5_22-57-51.png


Missouri is the 8th last team in. No quad one wins OOC.. they have 3 in conference. Lost vs Illinois, Nothing overwhelming here. Ole Miss is not good, but no road game is easy for a bubble team.










 
upload_2018-2-5_23-1-18.png


Boise St is the last team in the tourney right now. They have one of those hard to compare resumes. The thing in their favour is a KP Rating of 47, which puts them on the same page as other bubble team. They have 1 quadrant one win at Oregon. All very meh. Sort of like Illinois St last year.
 
This seems to be as good a place as any - jncuse's excellent bubble-tracking threads.

After yesterday's games, here is where we stand, quadrant-wise. Full disclosure - I used RPI as the basis for determining which quadrant a win or loss falls in. I am not 100% sure what the committee will use. They have not said officially, other than "metric-based measurements".

Q1 wins = 1
Q2 wins = 5
Q3 wins = 5
Q4 wins = 5

Note - Buffalo is just outside being a Q1 win (a 2 spot move up would make then a Q1, and not a Q2, win).

For completeness' sake:

Q1 losses = 4
Q2 losses = 2
Q3 losses = 2
Q4 losses = 0
 
For some perspective compared to other teams. We were second last out before last night.

upload_2018-2-6_19-59-51.png


Not much to separate these teams if we are only looking at quadrant wins.
 
Providence just gets by Georgetown, after trailing several parts in the game.
Missouri just gets by Ole Miss.
Arkansas got by South Carolina.

Boise St may lose to New Mexico so at least that would be a positive.

EDIT - And New Mexico blows it ... and at the end of the game quite a cheapshot forearm.
 
Last edited:
Well no gains were made tonight, except that none of the wins were that impressive. Maybe it will be different tomorrow.

upload_2018-2-7_0-38-46.png


Marquette is the 4th last team out on the matrix, There schedule has been really hard (1o quad one games), and they have nothing close to a bad loss. But still 2-8 is not going to get it done. their quality wins are at providence and home vs Seton hall.

upload_2018-2-7_0-40-9.png
 
upload_2018-2-7_0-43-27.png



The Bonnies hang around as the 5th last team out. They are 5-5 against the top 2 quadrants, with 2 quality road wins (Buffalo and Syracuse). They can't afford more mistakes as they already have a real bad loss, and only have one top 100 RPI opponent the rest of the way (#6 Rhode Island at home)

upload_2018-2-7_0-45-56.png
 
The Bonnies are such a tough conundrum for us. They're a fellow bubble buddy so normally we'd want themt o lose, but being in the A10 there are a lot of chances for "bad losses". The further away they get from the bubble (in the negative direction) the worse of a loss it becomes for us. And the better they play, their metrics improve, but they could also take an at-large spot from us.
 
Here's a question I have for jncuse, I know you looks at the rpi forecast a lot, so maybe you have some insight here. I'm not good at pasting pictures, so apologies, but here is our rpi forecast
Final Record Expected RPI Probability
23-8 14.7 0.17%
22-9 19.8 2.55%
21-10 26.8 12.59%
20-11 35.5 27.45%
19-12 44.7 31.81%
18-13 53.8 19.11%
17-14 63.6 5.62%
16-15 76.5 0.69%

It seems like we have a relatively wide range out comes; ie, 20-11 we are like a 36 RPI, one extra win moves you up to 27, one fewer win is 45. Pretty big swing. Is this normal? (I feel like no, but you definitely have a lot more experience with it than me)

I was also looking at some of the teams we have already beaten, (warning, lot of text incoming!!!!)

1) Buffalo currently has an RPI of 32. Expected RPI is 33, but using their numbers, they have a 50-50 shot (53% actually) of finishing with an RPI better than 30, which would make them a quad one win.
2) Toledo is currently 73, projected to finish right around there. 75 is the important number here, if they remain 75 or better (about 59%) that's a quad 2 win.
3) UConn is currently at 95, but projected for 104. For neutral site wins, 100 is the dividing line between quad 2 and 3. They have about a 55% chance of staying in the top 100.
4) Georgetown currently 149, projected around 150. They'd need to move into the top 135 to make it a quad 2 win. About a 25% chance, I hesitate to even include it.

We also have some shots coming up, Clemson, UNC, Miami, and Duke all look to be quad 1 teams. NC State would currently be a quad 2 win for us (63 RPI) but is projected to drop to 76, right outside the top 75.

Of the wins we've already banked, I think the Buffalo one is pretty big, would be nice to flip it to tier 1 (though per the Warren Nolan site, they are 27 and not 33, so it's already one)
 
Based on last night' games, here is where we stand, quadrant-wise. Buffalo moved into a Q1 win.

Q1 wins = 2
Q2 wins = 4
Q3 wins = 5
Q4 wins = 5

Q1 losses = 4
Q2 losses = 3
Q3 losses = 1
Q4 losses = 0
 
Knicks411 said:
Here's a question I have for jncuse, I know you looks at the rpi forecast a lot, so maybe you have some insight here. I'm not good at pasting pictures, so apologies, but here is our rpi forecast
Final Record Expected RPI Probability
23-8 14.7 0.17%
22-9 19.8 2.55%
21-10 26.8 12.59%
20-11 35.5 27.45%
19-12 44.7 31.81%
18-13 53.8 19.11%
17-14 63.6 5.62%
16-15 76.5 0.69%

It seems like we have a relatively wide range out comes; ie, 20-11 we are like a 36 RPI, one extra win moves you up to 27, one fewer win is 45. Pretty big swing. Is this normal? (I feel like no, but you definitely have a lot more experience with it than me)

I was also looking at some of the teams we have already beaten, (warning, lot of text incoming!!!!)

1) Buffalo currently has an RPI of 32. Expected RPI is 33, but using their numbers, they have a 50-50 shot (53% actually) of finishing with an RPI better than 30, which would make them a quad one win.
2) Toledo is currently 73, projected to finish right around there. 75 is the important number here, if they remain 75 or better (about 59%) that's a quad 2 win.
3) UConn is currently at 95, but projected for 104. For neutral site wins, 100 is the dividing line between quad 2 and 3. They have about a 55% chance of staying in the top 100.
4) Georgetown currently 149, projected around 150. They'd need to move into the top 135 to make it a quad 2 win. About a 25% chance, I hesitate to even include it.

We also have some shots coming up, Clemson, UNC, Miami, and Duke all look to be quad 1 teams. NC State would currently be a quad 2 win for us (63 RPI) but is projected to drop to 76, right outside the top 75.

Of the wins we've already banked, I think the Buffalo one is pretty big, would be nice to flip it to tier 1 (though per the Warren Nolan site, they are 27 and not 33, so it's already one)
Expand...

Some great analysis on those teams around the quadrant borders. As to your first question, that RPI variability from record to record is within my expectations. It's been around mid 30's with 20 wins, and mid 40's with 19 wins since the end of December.
 
The matrix has been updated. Just note that of the 102 brackets that are out, only 40 were done after our game on Monday, so the overall totals are lagging. We are only in 16% of the older brackets, but we are up to 45% on the new brackets.

Out of 40 New
Virginia Tech 24
Boise St 22
----------------
Syracuse 18
UCLA 10

So we are still tracking as the last team out, but it is almost a coinflip.
 
You must log in or register to reply here.

Forum statistics

Threads
168,200
Messages
4,755,539
Members
5,944
Latest member
cusethunder

Share this page

Online statistics

Members online
176
Guests online
1,564
Total visitors
1,740
Totals may include hidden visitors.


Top Bottom