rbg
Postdoc
Posts: 3052
Reg: 10-20-14
|
02-07-19 01:49 PM - Post#275354
Look at the last 2 weeks of playoff odds from the Yale Undergraduate Sports Analytics Group
1/27/19
1 Yale
Auto Bid 62.3%, Playoff Probability 99.4%, 1 Seed 81.9%, 2 Seed 12.3%, 3 Seed 3.8%, 4 Seed 1.4%
2 Brown
Auto Bid 14.1%, Playoff Probability 77.0%, 1 Seed 3.8%, 2 Seed 30.7%, 3 Seed 25.2%, 4 Seed 17.3%
3 Harvard
Auto Bid 8.9%, Playoff Probability 68.0%, 1 Seed 5.7%, 2 Seed 21.1%, 3 Seed 21.9%, 4 Seed 19.3%
4 Princeton
Auto Bid 3.5%, Playoff Probability 65.1%, 1 Seed 4.4%, 2 Seed 17.6%, 3 Seed 20.4%, 4 Seed 22.7%
5 Penn
Auto Bid 10.4%, Playoff Probability 62.1%, 1 Seed 4.1%, 2 Seed 15.3%, 3 Seed 20.5%, 4 Seed 22.2%
6 Dartmouth
Auto Bid 0.7%, Playoff Probability 21.0%, 1 Seed 0.4%, 2 Seed 2.7%, 3 Seed 6.2%, 4 Seed 11.7%
7 Columbia
Auto Bid 0.1%, Playoff Probability 4.9%, 1 Seed 0.0%, 2 Seed 0.4%, 3 Seed 1.2%, 4 Seed 3.3%
8 Cornell
Auto Bid 0.1%, Playoff Probability 3.4%, 1 Seed 0.1%, 2 Seed 0.3%, 3 Seed 0.8%, 4 Seed 2.2%
2/3/19
1 Yale
Auto Bid 57.0%, Playoff Probability 99.4%, 1 Seed 61.0%, 2 Seed 26.8%, 3 Seed 9.3%, 4 Seed 2.3%
2 Harvard
Auto Bid 23.5%, Playoff Probability 96.3%, 1 Seed 27.1%, 2 Seed 40.6%, 3 Seed 20.3%, 4 Seed 8.3%
3 Princeton
Auto Bid 6.3%, Playoff Probability 91.1%, 1 Seed 11.5%, 2 Seed 21.7%, 3 Seed 35.5%, 4 Seed 22.4%
4 Brown
Auto Bid 7.9%, Playoff Probability 63.8%, 1 Seed 0.6%, 2 Seed 6.9%, 3 Seed 23.8%, 4 Seed 32.5%
5 Penn
Auto Bid 5.2%, Playoff Probability 38.7%, 1 Seed 0.9%, 2 Seed 3.5%, 3 Seed 8.9%, 4 Seed 25.4%
6 Dartmouth
Auto Bid 0.2%, Playoff Probability 7.5%, 1 Seed 0.0%, 2 Seed 0.1%, 3 Seed 1.5%, 4 Seed 5.9%
7 Cornell
Auto Bid 0.1%, Playoff Probability 3.4%, 1 Seed 0.0%, 2 Seed 0.1%, 3 Seed 0.7%, 4 Seed 2.6%
8 Columbia
Auto Bid 0.1%, Playoff Probability 0.4%, 1 Seed 0.0%, 2 Seed 0.0%, 3 Seed 0.1%, 4 Seed 0.3%
Edited by rbg on 02-07-19 02:00 PM. Reason for edit: No reason given.
|
rbg
Postdoc
Posts: 3052
Reg: 10-20-14
|
02-12-19 09:59 AM - Post#276024
In response to rbg
2/10/19
1 Yale (Last Week #1)
Auto Bid 65.1%, Playoff Probability 99.9%, 1 Seed 87.8%, 2 Seed 9.6%, 3 Seed 1.8%, 4 Seed 0.7%
2 Princeton (Last Week #3)
Auto Bid 7.0%, Playoff Probability 85.1%, 1 Seed 6.2%, 2 Seed 36.1%, 3 Seed 26.5%, 4 Seed 16.3%
3 Harvard (Last Week #2)
Auto Bid 11.8%, Playoff Probability 79.2%, 1 Seed 5.3%, 2 Seed 31.4%, 3 Seed 24.5%, 4 Seed 18.0%
4 Penn (Last Week #5)
Auto Bid 10.2%, Playoff Probability 60.9%, 1 Seed 0.7%, 2 Seed 11.1%, 3 Seed 22.2%, 4 Seed 26.9%
5 Brown (Last Week #4)
Auto Bid 4.8%, Playoff Probability 45.0%, 1 Seed 0.1%, 2 Seed 8.4%, 3 Seed 15.4%, 4 Seed 21.1%
6 Cornell (Last Week #7)
Auto Bid 1.0%, Playoff Probability 26.7%, 1 Seed 0.4%, 2 Seed 3.3%, 3 Seed 8.3%, 4 Seed 14.7%
7 Dartmouth (Last Week #6)
Auto Bid 0.1%, Playoff Probability 3.6%, 1 Seed 0.0%, 2 Seed 0.2%, 3 Seed 1.1%, 4 Seed 2.3%
8 Columbia (Last Week #8)
Auto Bid 0.1%, Playoff Probability 0.1%, 1 Seed 0.0%, 2 Seed 0.0%, 3 Seed 0.0%, 4 Seed 0.1%
The group notes that Cornell's low numbers is a reflection of its having the hardest remaining strength of schedule. From their numbers, the hardest to easiest remaining strength of schedules are Columbia/Cornell, Dartmouth/Harvard, Penn/Princeton and Brown/Yale.
|
rbg
Postdoc
Posts: 3052
Reg: 10-20-14
|
02-13-19 10:10 AM - Post#276134
In response to rbg
Luke Benz of the YUSAG posted an article at the Yale Daily News discussing the most recent probabilities.
https://yaledailynews.com/blog/2019/02/13/by-th e-n...
Looking at all the information, his present rankings of the teams are:
1 - Yale (trending up)
2 - Harvard (trending up)
3 - Cornell (trending up)
4 - Penn (trending down)
5 - Princeton (trending down)
6 - Brown (trending down)
7 - Dartmouth (trending neutral)
8 - Columbia (trending down)
|
TigerFan
PhD Student
Posts: 1888
Reg: 11-21-04
|
02-13-19 10:14 PM - Post#276298
In response to rbg
Tigers beat your #3 and 4 on their home courts but whatever...
|
palestra38
Professor
Posts: 32812
Reg: 11-21-04
|
02-14-19 08:36 AM - Post#276337
In response to TigerFan
I think that is based on all games. Penn has better wins. But your point is well-taken and Penn is in trouble in the League.
|
penn nation
Professor
Posts: 21193
Reg: 12-02-04
|
02-14-19 08:11 PM - Post#276410
In response to palestra38
I think that is based on all games. Penn has better wins. But your point is well-taken and Penn is in trouble in the League.
Yes, in some trouble. The rematch with Cornell is a must win for Penn, IMHO, in order to make the 4 team tourney. Losing to Harvard this weekend would already put us on extremely shaky ground.
|
TigerFan
PhD Student
Posts: 1888
Reg: 11-21-04
|
02-14-19 08:14 PM - Post#276412
In response to penn nation
Very interesting weekend. Harvard could sweep or be swept. Same with Cornell and Brown. Quite a race!
|
rbg
Postdoc
Posts: 3052
Reg: 10-20-14
|
02-16-19 06:55 PM - Post#276745
In response to TigerFan
2/16/19
1 Yale (2/10/19 #1)
Auto Bid 63.6%, Playoff Probability 99.7%, 1 Seed 87.2%, 2 Seed 10.5%, 3 Seed 1.6%, 4 Seed 0.4%
2 Harvard (2/10/19 #3)
Auto Bid 17.9%, Playoff Probability 91.4%, 1 Seed 10.0%, 2 Seed 50.6%, 3 Seed 19.7%, 4 Seed 11.1%
3 Princeton (2/10/19 #2)
Auto Bid 4.4%, Playoff Probability 74.0%, 1 Seed 1.6%, 2 Seed 17.2%, 3 Seed 34.9%, 4 Seed 20.3%
4 Penn (2/10/19 #4)
Auto Bid 9.2%, Playoff Probability 60.1%, 1 Seed 0.8%, 2 Seed 12.1%, 3 Seed 19.3%, 4 Seed 27.9%
5 Cornell (2/10/19 #6)
Auto Bid 1.7%, Playoff Probability 46.0%, 1 Seed 0.9%, 2 Seed 7.7%, 3 Seed 14.4%, 4 Seed 23.0%
6 Brown (2/10/19 #5)
Auto Bid 3.2%, Playoff Probability 26.3%, 1 Seed 0.0%, 2 Seed 1.6%, 3 Seed 9.5%, 4 Seed 15.2%
7 Dartmouth (2/10/19 #7)
Auto Bid 0.1%, Playoff Probability 3.0%, 1 Seed 0.0%, 2 Seed 0.1%, 3 Seed 0.7%, 4 Seed 2.2%
8 Columbia (2/10/19 #8)
Auto Bid 0.1%, Playoff Probability 0.1%, 1 Seed 0.0%, 2 Seed 0.0%, 3 Seed 0.0%, 4 Seed 0.1%
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
Yale UG Sports Analytics Group Playoff Odds - Men
02-17-19 01:38 PM - Post#276989
In response to rbg
Dusted off the simulator this morning and the number one thing that popped out to me was how likely ties are to be a factor in the Ivy tourney - and not just a simple tie for fourth and fifth.
In 41% of simulations there's a tie for the last spot, but in only 24% of simulations is the tie exclusively between fourth and fifth place. That leaves 17% of sims where three (or more) teams are tied in a situation where at least one will be left out. The most common is a tie for fourth, fifth and sixth (10%), while a tie for third, fourth and fifth is next most at 6%. There is a 1% chance of a crazy tie between third, fourth, fifth and sixth.
That obviously brings into play the oft-maligned Ivy tiebreaker rules, which despite being widely discussed, haven't actually been used to break a fourth-place tie yet (but were used to break the first-place tie for seeding and NIT purposes last year).
Right now the odds have Yale as having a 97% chance to be a top two seed (I have them at 77% to be the one-seed). And Harvard is at ~80% to be a top two seed. At the same time, Columbia and Dartmouth have essentially zero chance of making the tourney. So, we can focus on what the breakers mean for that middle group of four vying for two spots.
If there's a sweep between two teams, things are pretty simple. From there, we need to analyze top wins.
So, right now, Princeton is in the best shape, having sealed the tie-breaker with Penn (and thus being effectively up three games) and having an early leg up in the tiebreaker with Cornell with the road win. It would need to beat Brown at home to have a shot at that tie-breaker. At the same time, the Tigers have zero wins against Harvard and Yale to help with the second tie-breaker (best win).
Cornell is in the next-best shape with wins over Penn and Brown AND a win over Harvard. That means if the Crimson stays in the top two, the Bears can, at best, tie the Big Red for the first two tie-breakers (beating Cornell AND Harvard to match Cornell's win) and Penn would need to beat Yale or Harvard as well as taking the return from the Big Red. Cornell did have the loss to Princeton at home and is in danger of losing that tiebreak.
Penn is in okay shape. The tiebreaker with Princeton is obviously lost, so it is effectively 3 games behind the top three. And its best shot at the tiebreaker with Cornell would require not only beating the Big Red but taking down Yale at home during the final weekend of the year. That being said, winning a game that would tilt the tiebreak back in its favor is probably necessary to getting into a tie for fourth anyway.
Brown is in the worst shape, needing to win the return by Cornell AND matching its best win to have a shot at that tie-breaker and needing to win at Penn to have a shot at that tie-breaker. It does have the early leg up over Princeton, but if Princeton holds serve at home next weekend, the odds of catching the Tigers even with a sweep that would secure the tie-breaker would diminish greatly.
The race should be pretty wild. Princeton is leaking oil as opponent three-point shooting has regressed to the mean making the Tigers' defense look pretty human, exposing an anemic offense. With the tiebreaker situation, though, if Princeton wins Friday, it'll be hard to catch. If not, Saturday becomes a must win with the closing schedule the Tigers have.
Penn and Brown have to sweep and should be favored in all of their games (Harvard-Brown might be close to a pick).
Meanwhile, Cornell, which won't be less than a 6 or so point dog in any game over the next two weekends could easily find itself at 5-7... BUT it likely only needs one upset over those four games (especially if it's at Penn) to be in very good shape.
Oh yeah... and the tourney odds I got:
Yale - 99.9%
Harvard - 99%
Princeton - 87%
Cornell - 40%
Penn - 35%
Brown - 34%
Dartmouth - 3%
Columbia - 0%
Edited by mrjames on 02-17-19 01:40 PM. Reason for edit: No reason given.
|
PennFan10
Postdoc
Posts: 3585
Reg: 02-15-15
|
02-17-19 06:33 PM - Post#277013
In response to mrjames
So many close games in the league. Every night the games are a battle. Dartmouth went to OT @Penn and lost by 1 @Princeton. What a brutal 0-2 for them. Harvard and Penn could have easily been 0-2 this weekend instead of 2-0/1-1.
The league is tough.
|
rbg
Postdoc
Posts: 3052
Reg: 10-20-14
|
02-18-19 02:24 PM - Post#277067
In response to PennFan10
2/18/19
1 Yale (2/16/19 #1)
Auto Bid 64.4%, Playoff Probability > 99.95%, 1 Seed 80.9%, 2 Seed 16.9%, 3 Seed 1.8%, 4 Seed 0.4%
2 Harvard (2/16/19 #2)
Auto Bid 21.3%, Playoff Probability 98.7%, 1 Seed 18.1%, 2 Seed 63.9%, 3 Seed 12.4%, 4 Seed 4.3%
3 Princeton (2/16/19 #3)
Auto Bid 4.7%, Playoff Probability 78.5%, 1 Seed 1.0%, 2 Seed 10.3%, 3 Seed 44.6%, 4 Seed 22.6%
4 Cornell (2/16/19 #5)
Auto Bid 1.4%, Playoff Probability 49.6%, 1 Seed 0.7%, 2 Seed 6.1%, 3 Seed 16.4%, 4 Seed 26.4%
5 Penn (2/16/19 #4)
Auto Bid 4.9%, Playoff Probability 40.7%, 1 Seed 0.0%, 2 Seed 1.5%, 3 Seed 10.9%, 4 Seed 28.3%
6 Brown (2/16/19 #6)
Auto Bid 3.1%, Playoff Probability 29.4%, 1 Seed 0.0%, 2 Seed 1.2%, 3 Seed 12.8%, 4 Seed 15.4%
7 Dartmouth (2/16/19 #7)
Auto Bid 0.1%, Playoff Probability 3.8%, 1 Seed 0.0%, 2 Seed 0.0%, 3 Seed 1.0%, 4 Seed 2.8%
8 Columbia (2/16/19 #8)
Auto Bid 0.1%, Playoff Probability 0.1%, 1 Seed 0.0%, 2 Seed 0.0%, 3 Seed 0.0%, 4 Seed 0.1%
|
HARVARDDADGRAD
Postdoc
Posts: 2691
Loc: New Jersey
Reg: 01-21-14
|
02-21-19 12:30 PM - Post#277382
In response to rbg
So, if Yale runs the table but loses in the Ivy League Tournament, do the 24-5 Eli's (13-1 in conference) get an at large bid to the NCAA's?
|
PennFan10
Postdoc
Posts: 3585
Reg: 02-15-15
|
02-21-19 02:26 PM - Post#277397
In response to HARVARDDADGRAD
they would be a quality candidate but won’t get picked.
|
HARVARDDADGRAD
Postdoc
Posts: 2691
Loc: New Jersey
Reg: 01-21-14
|
02-21-19 02:40 PM - Post#277398
In response to PennFan10
I feared that.
So - is a second Ivy bid ever a realistic option?
|
westcoast
Senior
Posts: 302
Reg: 03-08-16
|
02-21-19 02:46 PM - Post#277399
In response to HARVARDDADGRAD
I don't think their record will be strong enough. Yale is around 70 in most ranking systems, and maybe could move up to 60 if they run the table convincingly. But they will probably not have any Q1 wins, and probably only two Q2 wins (Miami, @Penn).
|
westcoast
Senior
Posts: 302
Reg: 03-08-16
|
02-21-19 02:48 PM - Post#277400
In response to HARVARDDADGRAD
Sure, a second bid is a realistic option. Even this year, Yale had 3 Q1 games (@Memphis, @Vermont, @Duke) and went 0-3. Winning at Duke is probably not realistic, but the other two games were very winnable, and I think Yale would have had a strong case if they had gone 26-3 win two Q1 wins.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
02-21-19 03:22 PM - Post#277408
In response to westcoast
So, objectively, the answer based on resume would be yes. Yale is currently 0.9 wins above bubble. If they ran the table (and let’s throw in there a semis win for good measure), they’d probably pick up about an extra win.
A loss against a 100-125 team in the finals, even at home, would probably only cost them 0.8ish of a win at most. So, they’d likely end up a full win-plus above bubble.
Now, because narratives are easier than math, that’s not how teams are subjectively judged - even though it’s the accurate way to do so. Illinois St, in recent memory, was nearly two wins above bubble and left out. But in that scenario they would become the Ivy first team since 2011 Harvard to finish with a positive WAB and not make the tourney.
|
mrjames
Professor
Posts: 6062
Loc: Montclair, NJ
Reg: 11-21-04
|
02-21-19 04:24 PM - Post#277415
In response to HARVARDDADGRAD
On the “second bid being realistic†front... in both the objective and subjective world, a Penn team that went 12-2 or better in the Ivy and lost at Yale in the tourney, on top of its non-conf resume, would have been a second bid.
|
PennFan10
Postdoc
Posts: 3585
Reg: 02-15-15
|
02-21-19 04:27 PM - Post#277416
In response to mrjames
What he said
|
penn nation
Professor
Posts: 21193
Reg: 12-02-04
|
02-21-19 04:53 PM - Post#277418
In response to mrjames
On the “second bid being realistic†front... in both the objective and subjective world, a Penn team that went 12-2 or better in the Ivy and lost at Yale in the tourney, on top of its non-conf resume, would have been a second bid.
Not with that home Monmouth loss.
|
|
