FanLax Forum Poll

[laxfan1313]

Based upon this week's poll, Cornell has 2 top ten wins and two others in the top 20.

[FanLax Computer]

Notre Dame was ranked so low early in the season because they hadn't beaten anyone. A "good loss" is still a loss. Wins show what teams a team is better than. Not "good losses". Now ND has some wins and they're in a better place. The moral of the story is, If you want to be ranked higher, you need to beat higher ranked teams.

St. Joe's is... 29.

Rank, Team, Win %, Win Rank, Loss Rank

1 Maryland 100% 1 1
2 Georgetown 90% 4 2
3 Princeton 80% 2 5
4 Cornell 91% 6 4
5 Virginia 80% 7 6
6 Rutgers 83% 13 3
7 Notre Dame 56% 7 7
8 Yale 78% 3 16
9 North Carolina 64% 13 8
10 Jacksonville 83% 5 14
11 Harvard 78% 17 9
12 Penn 50% 7 16
13 Ohio State 73% 10 14
14 Denver 55% 12 18
15 Army 82% 20 11
16 Duke 64% 13 20
17 Boston University 73% 24 10
18 Brown 60% 16 25
19 Richmond 60% 24 12
20 Lehigh 70% 22 20
21 Utah 67% 19 24
22 Johns Hopkins 42% 10 36
23 Loyola 50% 18 29
24 UMass 60% 26 23
25 High Point 55% 34 12
26 Penn State 27% 21 32
27 Villanova 60% 29 22
28 Towson 45% 27 29
29 Saint Joseph's 82% 28 27

[Farfromgeneva]

Based upon this week's poll, Cornell has 2 top ten wins and two others in the top 20.

Got shots with 3 others who may all be too 20 at seasons end as well (not sure about Brown but fairly certain on Army and Princeton is a lock)

Where Harvard finishes seems still up in the air but with RPI math benefit in remaining schedule should easily remain inside top 20 unless some OOC stuff drags them down meaningfully. Beyond BU.

[Farfromgeneva]

Notre Dame was ranked so low early in the season because they hadn't beaten anyone. A "good loss" is still a loss. Wins show what teams a team is better than. Not "good losses". Now ND has some wins and they're in a better place. The moral of the story is, If you want to be ranked higher, you need to beat higher ranked teams.

St. Joe's is... 29.

Rank, Team, Win %, Win Rank, Loss Rank

1 Maryland 100% 1 1
2 Georgetown 90% 4 2
3 Princeton 80% 2 5
4 Cornell 91% 6 4
5 Virginia 80% 7 6
6 Rutgers 83% 13 3
7 Notre Dame 56% 7 7
8 Yale 78% 3 16
9 North Carolina 64% 13 8
10 Jacksonville 83% 5 14
11 Harvard 78% 17 9
12 Penn 50% 7 16
13 Ohio State 73% 10 14
14 Denver 55% 12 18
15 Army 82% 20 11
16 Duke 64% 13 20
17 Boston University 73% 24 10
18 Brown 60% 16 25
19 Richmond 60% 24 12
20 Lehigh 70% 22 20
21 Utah 67% 19 24
22 Johns Hopkins 42% 10 36
23 Loyola 50% 18 29
24 UMass 60% 26 23
25 High Point 55% 34 12
26 Penn State 27% 21 32
27 Villanova 60% 29 22
28 Towson 45% 27 29
29 Saint Joseph's 82% 28 27

Do you score these or just ordinal rank? I’m curious what your delta is between 29 and 19.

Would you say it’s true (or no) with your system that with 20-30% of the reg season left that there’s probably more opportunity to gain, or pick up, relative to teams ranked above you in general? Or there’s more upside to your in the books existing opponents to improve your rank if you are a little lower ranked in those areas (w/l). Subject to actually look at remaining schedulers relative to completed portion.

So I could see upside in SJUs w rank and loss rank relative to someone like BU (win rank) or Richmond (w/l rank).

Just curious. I wouldn’t be upset with most of those teams in the 18-20 range in a poll through 29 and think SJU is just as deserving but also could see some of their OOC picking it up and helping them on both their w rank and l rank relative to the strength of their remaining schedule.

(Could see SBU being neutral but already not bad on pure record basis. Upside to w strength on PSU, Towson, not sure on Prov, little upside on hobart, not sure on Monmouth, losses to Del and Drexel may look better come season end but there’s some counterbalance w towson unless those three run a collective train on UMass, Fairfield & Hofstra from here on out. Remaining opponents are NEC not great but right now Hobart is terrible so some look better today on paper 6-5 LIU, 5-5 Merrimack and whatever Wagner is plus’s a presumed L to UPenn but has a fighting chance there)

[Gobigred]

Notre Dame was ranked so low early in the season because they hadn't beaten anyone. A "good loss" is still a loss. Wins show what teams a team is better than. Not "good losses".

What HGK doesn't seem to understand is that the 50th to 60th best teams in the country could lose to those same three schools. But they wouldn't be ranked very high because they don't have the cachet Notre Dame has. As Fanlax Computer says above, you have to then look at whom a team has beaten to slot them in the ranking. Losing to the top three teams in the country proves nothing. Nearly every team in Division I could accomplish that.

[Farfromgeneva]

Notre Dame was ranked so low early in the season because they hadn't beaten anyone. A "good loss" is still a loss. Wins show what teams a team is better than. Not "good losses". Now ND has some wins and they're in a better place. The moral of the story is, If you want to be ranked higher, you need to beat higher ranked teams.

St. Joe's is... 29.

Rank, Team, Win %, Win Rank, Loss Rank

1 Maryland 100% 1 1
2 Georgetown 90% 4 2
3 Princeton 80% 2 5
4 Cornell 91% 6 4
5 Virginia 80% 7 6
6 Rutgers 83% 13 3
7 Notre Dame 56% 7 7
8 Yale 78% 3 16
9 North Carolina 64% 13 8
10 Jacksonville 83% 5 14
11 Harvard 78% 17 9
12 Penn 50% 7 16
13 Ohio State 73% 10 14
14 Denver 55% 12 18
15 Army 82% 20 11
16 Duke 64% 13 20
17 Boston University 73% 24 10
18 Brown 60% 16 25
19 Richmond 60% 24 12
20 Lehigh 70% 22 20
21 Utah 67% 19 24
22 Johns Hopkins 42% 10 36
23 Loyola 50% 18 29
24 UMass 60% 26 23
25 High Point 55% 34 12
26 Penn State 27% 21 32
27 Villanova 60% 29 22
28 Towson 45% 27 29
29 Saint Joseph's 82% 28 27

Struggle w Yale below ND-guess strictly driven by that (early but still) loss to PSU mainly.

[FanLax Computer]

Do you score these or just ordinal rank? I’m curious what your delta is between 29 and 19.

Ordinal Rank. It's a Ranking based 100% on ordinal Rank. That is to say, whether a team is good or bad is not based on an objective formula but 100% based on subjective relativity.

Would you say it’s true (or no) with your system that with 20-30% of the reg season left that there’s probably more opportunity to gain, or pick up, relative to teams ranked above you in general? Or there’s more upside to your in the books existing opponents to improve your rank if you are a little lower ranked in those areas (w/l). Subject to actually look at remaining schedulers relative to completed portion.

20-30% left in the season is, in general, irrelevant. *Who* the team is playing is extremely relevant. The critical path for a team is improving their Best Three Wins and maintaining their Three Worst Losses.

SJUs
(Could see SBU being neutral but already not bad on pure record basis. Upside to w strength on PSU, Towson, not sure on Prov, little upside on hobart, not sure on Monmouth, losses to Del and Drexel may look better come season end but there’s some counterbalance w towson unless those three run a collective train on UMass, Fairfield & Hofstra from here on out. Remaining opponents are NEC not great but right now Hobart is terrible so some look better today on paper 6-5 LIU, 5-5 Merrimack and whatever Wagner is plus’s a presumed L to UPenn but has a fighting chance there)

SJU's best wins are against 26, 28, and 42. They need to replace that #42 and, if possible, improve their Best Wins for the 26 and 28. And, looking at their schedule, they basically have one opportunity to do this: UPenn. Not including post-regular season games and... Maybe the teams they already beat improve their rankings. This'd help SJU as well.

[RURICK]

Can someone please explain to me how UNC is ranked in the top 10 and OSU is ranked 13th after OSU beat ND 14-11 and beat UNC 20-8, that's right, 20-8, lol. I will wait.

[HGK]

FanLax Computer wrote: ↑
Notre Dame was ranked so low early in the season because they hadn't beaten anyone. A "good loss" is still a loss. Wins show what teams a team is better than. Not "good losses". Now ND has some wins and they're in a better place. The moral of the story is, If you want to be ranked higher, you need to beat higher ranked

That’s my exact point if you read my post. So ND played a great schedule early and lost to all top 5 teams. Nobody wanted them ranked. Michigan played a terrible schedule but was undefeated and everyone ranked them high. Does anyone actually look at the quality of the contests? I guess i just don’t get the methodology. But if I was playing a team in the QF for example, i’d much rather play the team with a great record and weak wins than the team with the worse record and good losses i can tell you that.

[masondixonlax]

FanLax Computer wrote: ↑
Notre Dame was ranked so low early in the season because they hadn't beaten anyone. A "good loss" is still a loss. Wins show what teams a team is better than. Not "good losses". Now ND has some wins and they're in a better place. The moral of the story is, If you want to be ranked higher, you need to beat higher ranked

That’s my exact point if you read my post. So ND played a great schedule early and lost to all top 5 teams. Nobody wanted them ranked. Michigan played a terrible schedule but was undefeated and everyone ranked them high. Does anyone actually look at the quality of the contests? I guess i just don’t get the methodology. But if I was playing a team in the QF for example, i’d much rather play the team with a great record and weak wins than the team with the worse record and good losses i can tell you that.

Exactly the reason why ND or Penn would be a very dangerous team to make it into the tournament. Even Duke when they click and everyone is on the same page.

[RURICK]

I am still waiting for an answer to my specific question in my last post. Thanks to anyone responding. Computer rankings are BS.

[FanLax Computer]

...Computer rankings are BS.

[FanLax Computer]

Can someone please explain to me how UNC is ranked in the top 10 and OSU is ranked 13th after OSU beat ND 14-11 and beat UNC 20-8, that's right, 20-8, lol. I will wait.

My rankings are based on each team's resume, not head-to-head. Not individual games but each team's entire body of work. And, with that said, UNC's resume is better than OSU's.
UNC beat Denver, Richmond, and High Point. OSU beat ND, Harvard, and JHU... and UNC. Yes, OSU has better wins (but only barely).
The issue for OSU is... UNC lost to Duke, OSU, and UVA. OSU lost to Denver, Rutgers, and Cornell. UNC's Losses are less painful than OSU's. Consequently, UNC has a better ranking.

[RURICK]

FanLax Computer, That's pretty funny.

[HGK]

Can someone please explain to me how UNC is ranked in the top 10 and OSU is ranked 13th after OSU beat ND 14-11 and beat UNC 20-8, that's right, 20-8, lol. I will wait.

My rankings are based on each team's resume, not head-to-head. Not individual games but each team's entire body of work. And, with that said, UNC's resume is better than OSU's.
UNC beat Denver, Richmond, and High Point. OSU beat ND, Harvard, and JHU... and UNC. Yes, OSU has better wins (but only barely).
The issue for OSU is... UNC lost to Duke, OSU, and UVA. OSU lost to Denver, Rutgers, and Cornell. UNC's Losses are less painful than OSU's. Consequently, UNC has a better ranking.

So now as I have been saying since February - good losses do matter. I give up.

[rolldodge]

Can someone please explain to me how UNC is ranked in the top 10 and OSU is ranked 13th after OSU beat ND 14-11 and beat UNC 20-8, that's right, 20-8, lol. I will wait.

My rankings are based on each team's resume, not head-to-head. Not individual games but each team's entire body of work. And, with that said, UNC's resume is better than OSU's.
UNC beat Denver, Richmond, and High Point. OSU beat ND, Harvard, and JHU... and UNC. Yes, OSU has better wins (but only barely).
The issue for OSU is... UNC lost to Duke, OSU, and UVA. OSU lost to Denver, Rutgers, and Cornell. UNC's Losses are less painful than OSU's. Consequently, UNC has a better ranking.

OSU has three top 15 RPI wins. UNC has none (edit:1. Richmond at 15). Average RPI of OSUs loses is 8. UNC is 10 and they have one more loss. It’s not even close.

[44WeWantMore]

Can someone please explain to me how UNC is ranked in the top 10 and OSU is ranked 13th after OSU beat ND 14-11 and beat UNC 20-8, that's right, 20-8, lol. I will wait.

My rankings are based on each team's resume, not head-to-head. Not individual games but each team's entire body of work. And, with that said, UNC's resume is better than OSU's.
UNC beat Denver, Richmond, and High Point. OSU beat ND, Harvard, and JHU... and UNC. Yes, OSU has better wins (but only barely).
The issue for OSU is... UNC lost to Duke, OSU, and UVA. OSU lost to Denver, Rutgers, and Cornell. UNC's Losses are less painful than OSU's. Consequently, UNC has a better ranking.

So now as I have been saying since February - good losses do matter. I give up.

When you lose to a bad team it tells you more about how bad you are than when you lose to a good team. Everybody will lose to number 1, but only a bad team loses to number 60.

Just like when you beat a good team it tells you more about how good you are than when you beat a bad team. Everybody will beat the lowest-ranked team, but only a good team will beat a top team.

The challenge is that you have cases where A beats B, B beats C, and C beats A. Then you need a way of sorting it out. Top three wins and losses is one way. There are other algorithms. Here is another one (PDF): http://college.laxpower2.com/algorithm_rating.pdf Here is another: https://www.collegehockeynews.com/info/?d=krach

[CU77]

Well Dr.F's score-based power rating doesn't much like UNC, has them at #27:
http://college.laxpower2.com/menx/rating01.php

[HGK]

Notre Dame was ranked so low early in the season because they hadn't beaten anyone. A "good loss" is still a loss. Wins show what teams a team is better than. Not "good losses".

What HGK doesn't seem to understand is that the 50th to 60th best teams in the country could lose to those same three schools. But they wouldn't be ranked very high because they don't have the cachet Notre Dame has. As Fanlax Computer says above, you have to then look at whom a team has beaten to slot them in the ranking. Losing to the top three teams in the country proves nothing. Nearly every team in Division I could accomplish that.

Your pomposity is amazing. And I couldn’t disagree with your views on losing to quality teams more.

[Matnum PI]

Notre Dame was ranked so low early in the season because they hadn't beaten anyone. A "good loss" is still a loss. Wins show what teams a team is better than. Not "good losses".

What HGK doesn't seem to understand is that the 50th to 60th best teams in the country could lose to those same three schools. But they wouldn't be ranked very high because they don't have the cachet Notre Dame has. As Fanlax Computer says above, you have to then look at whom a team has beaten to slot them in the ranking. Losing to the top three teams in the country proves nothing. Nearly every team in Division I could accomplish that.

Watched a HS game this afternoon and a team went up big, 10-3 or so. Then the losing team slowly crept back into the game (it was 13-10 at one point) until... The team that was winning all along won 14-10. Now, people will walk away from this game and say, Wow! Losing Team did pretty well only losing 14-10 (and it was 13-10 at one point)! and... I dunno. To me, it looked like Winning Team was winning handily so they took their foot off the gas. Looked to me like if Losing Team was ever to pose a real threat, Winning Team would've turned on the gas and built another 7 goal lead. On the whole, people/teams play to win, not to win by as many goals as possible. From where I'm sitting...