PWR Today, 2-17-02

Well, we moved up to #9 again. I'm a little surprised how much movement there still is at this point in the season. In any case, here is how we stack up against the teams around us:

we are just behind:

    Colorado College vs Cornell
          RPI      0.5989  1           0.5700  0
          L16    11- 3- 2  0         12- 4- 0  0
          TUC     8- 6- 0  0          7- 5- 0  1
          H2H              0                   0
          COP     3- 0- 0  0          2- 0- 0  0
          ============================================
          PTS              1                   1

    Cornell vs Michigan State
          RPI      0.5700  0          0.5985  1
          L16    12- 4- 0  1         10- 3- 3  0
          TUC     7- 5- 0  0          9- 4- 5  1
          H2H                 0                      0
          COP     0- 2- 0  0          4- 0- 2  1
          ============================================
          PTS                  1                     3

    Cornell vs Maine
          RPI      0.5700  0           0.5799  1
          L16    12- 4- 0  1          9- 4- 3   0
          TUC     7- 5- 0  1         10- 7- 5  0
          H2H                 0                       0
          COP     3- 5- 0  0          3- 2- 0   1
          ============================================
          PTS                 2                       2

Cornell vs. CC: Although they are currently ranked 6th in PWR wins, this is actually the easiest comparison for us to pick up. We obviously won't catch their RPI, and common opponents won't change between now and the end of the season. Therefore, it will come down to TUC and L16. They will definitely sweep their remaining weekends vs. MTU and UAA, but will probably lose at best in the semis of the WCHA playoffs, so their final L16 would be 13-2-1. Our losses to NMU and Maine drop out of L16 the last weekend of the season, and the first loss to Dartmouth drops out the first weekend of the playoffs. Say we drop another RS game and then sweep the playoffs, our L16 would be 14-2. That would just barely give us L16. CC's only remaining TUC will be their semi-final opponent, who should beat them, so they'll finish 8-7 vs. TUCs. We should do better than that (much more on our TUC possibilities below), so we would keep this category. Therefore, if we make a good run and/or CC falters, we should be able to flip the comparison with CC based on L16. Nice.


Cornell vs. Mighigan State: The best we can hope for here is to flip the TUC comparison, which isn't that likely. Therefore, the best we can home for is a tie. We will probably not catch their RPI based on the competition we will face (all will be pretty close to .500 (and below in the QFs) from here on out). Therefore, MSU will keep this comparison outright or due to the RPI tiebreaker

Cornell vs. Maine: this is what will happen with the Michigan State comparison if we caught their TUC record - we lose based on RPI tiebreaker. Again, we will not catch their RPI nor common opponents record, so this comparison is definitely lost for the season.

So we have a reasonable chance of picking up one more PWR win. Not too bad.

We are just ahead of:

    Cornell vs Michigan
          RPI      0.5700  0          0.5738  1
          L16    12- 4- 0  1          9- 4- 3  0
          TUC     7- 5- 0  1          7- 6- 3  0
          H2H                 0                      0
          COP     1- 2- 0  1          0- 2- 1  0
          ============================================
          PTS                 3                       1

    Alaska-Fairbanks vs Cornell
          RPI      0.5635  0          0.5700  1
          L16    11- 3- 2  0         12- 4- 0  0
          TUC     8- 7- 1  0          7- 5- 0  1
          H2H                 0                      0
          COP     2- 2- 0  1          0- 2- 0  0
          ============================================
          PTS                  1                     2


    Cornell vs Northern Michigan
          RPI      0.5700  1           0.5602  0
          L16    12- 4- 0  1         11- 4- 1  0
          TUC     7- 4- 0  1          8- 6- 1  0
          H2H                 0                      1
          COP     0- 1- 0  0          0- 2- 0  0
          ============================================
          PTS                  3                      1

Cornell vs. Michigan: As with MSU and Maine, we will not catch their RPI. Therefore, this comparison could turn quite easily - if the TUC or L16 flips to Michigan, they win the comparsion based on the RPI tiebreaker. Michigan's remaining games are OSU and WMU, quite winnable for Michigan. Say they make a run and lose the CCHA title game to MSU (not too unlikely) their L16 would improve to 12-3-1. If we keep the same L16 scenario for us (dropping another RS game and winning at Placid), we're 14-2, so we should keep this. Michigan has 5 or 6 TUCs remaining: OSUx2, WMUx2, CCHA semi-final, and CCHA final. Tough to call either way, but I bet we will keep the TUC category, and therefore the comparison.

Cornell vs. UAF: Awfully close on L16 and TUC. If we choke vs upcoming TUCs, we could lose this comparison. UAF has 2 left vs UNO (TUC) and 2 vs UAF (not TUC), while we have TUC games vs. Clarkson (maybe - at .500 now, with games left vs Cornell/Colgate, and UVM/Dartmouth. Clarkson definitely the toughest remaining ECAC schedule, but they should win a playoff game or two, so they could maybe stay just at .500), RPI, and Union. Throw in the possibility of UAH becoming a TUC: they currently need 7 points in their last 2 games vs. Findlay (should win) and 2 vs. Bemidji (probable split). UAH should win some playoff game though, so I'd say there's a reasonable chance they become a TUC. The only problem is that they're 2-7 in their last 9, so maybe they're not playing that well. On the other hand, their 7 losses were to St. Cloud, UNO, Mankato, and Air Force, so I hope they'll pull it out. In any case, this comparison could flip, but shouldn't if Cornell takes care of business.

Cornell vs. NMU: Wow is this one close. See comments on UAF - if we take care of business and maybe get some help from UAH, we'll keep this one. Definitely have to beat all TUCs from here on out! Dropping the SLU game or one QF game wouldn't hurt us nearly as badly as losing to CLK, RPI, or Union.

Anyway, there are the details. Here's the summary: We can only pick up one more PWR comparison, and could drop three. All three that we could lose hinge on the same factors: TUC and L16. Therefore, all of these comparisons could go the same way - we win them all or we lose them all. In any case, our PWR finish looks to be 8th at the best and 12th at the worst. If we beat Clarkson, RPI, and Union, our L16 and TUC should be good enough to guarantee an NCAA slot regardless of ECAC tourney result. If we drop one or more to those three, we pretty much need to win at Placid to punch our ticket. The irony could be that our PWR would be good enough to get in if we win at Placig, and not good enough to get in if we lose at Placid, rendering the auto-bid entirely meaningless for us!

Sorry for the typos - wouldn't let me edit because the message was too long (URI too large?)

One more thought: if we have to win at Placid to get in, don't you think the rest of the ECAC should let us win, just so we put a team the NCAAs who has a resonable chance at a respectable showing?

Excellent summary, Robb -- thanks very much.

Hopefully the team is fully aware that the final weekend could be critical to NCAA hopes even if Cornell clinches the regular season title this weekend.

I join Greg in thanking you, Robb. Very well done, and much appreciated.

F'n Dartmouth!

So here's the question - what about catching the likes of SCSU, Denver, and Minnesota? I haven't looked this week, but last week they were our best chances for additional comparisons, assuming that Colgate, Brown, and UAH finish as TUCs. Gotta love the flawed PWR, where taking comparisons that you have no business taking can boost your ranking into the top 5.

Nice summary Robb. I doubt Cornell will catch those three teams in the TUC comparison anyway, since they have several games left against likely TUCs. Of course, the WCHA powers could always lose most of those games. . . that'd be nice.

We already win the comparison with Minnesota:

Cornell Minnesota
18-6-1 .5700 RPI .6121 22-7-4
7-5 .5833 TUC .5000 4-4-2
12-4 .7500 L16 .6562 10-5-1
2-0 1.0000 COp 1.0000 2-0
0-0 H2H 0-0

But Minnesota wins comparisons with BU, MSU, CC, and Maine, who win comparisons with us.

Incidentally, if we catch CC but all the other comparisons remain the same, the PWR will look like:

4. Minnesota 24 beats BU, CC, MS, Me
Boston U 24 beats CC, MS, Cr, Me
6. Col Coll 22 beats MS, Me
Mich St 22 beats Cr, Me
Cornell 22 beats Mn, CC
9. Maine 21 beats Cr

Minnesota would beat out BU on the individual comparison, and CC/MSU/Cr would come down to RPI, in which Cornell trails the other two. (Plus, if they ignore the comparison with Minnesota because they're clear of the bubble or already have a bye, we drop back into the tie with Maine.) Note that for selection purposes, we'd benefit from MSU winning the CCHA tournament, since then CC wouldn't get credit for the comparison with them in the selection phase. (I think.)

I was recently accepted to cornell, and I have just begun to follow NCAA hockey. I was under the impression that both the ECAC winner and the ECAC tournament winner both got spots in the NCAA tournament, but it seems from reading this thread that I am wrong. Can somebody please email or post an explanation of how a team qualifies for the NCAA tournament for me? Thank you very much!

Eric -

A few years ago, before the MAAC autobid, there was an automatic berth to the NCAA's for the RS *and* the Tourney champions in the ECAC. If a team should win both the regular season and the tournament titles, that team would receive a first round bye in the NCAA tournament.

Now, there are 5 autobids for 12 spots. So instead of giving 2 to each conference, the NCAA only gives one automatic bid. The ECAC (in fact, I think every conference does this) awards the automatic bid to the winner of the conference tournament. This makes the regular season title more or less meaningless, apart from getting the best seed in the tourney.

For a team to make the NCAAs out of the ECAC, that team must either:

Win the ECAC tournament.
Win enough PairWise comparisons to receive an at-large bid.

There's a pretty good explanation of the pairwise on the uscho.com site someplace. There's also a good explanation at [slack.net] although I'm not sure where it is. The slack.net/hockey site also has a great layout for viewing the pairwise rankings and the individual comparisons, which will give you a good idea of who is currently in, out, or on the bubble for an NCAA at-large bid. JTW, can you provide the direct links?

Thanks a lot...that clears many things up...

Flipping the comparison with Maine isn't really that hard. A difference of 0.01 in RPIT isn't insurmountable even at this point in the season if Maine loses a few. The TUC criteria is also very much up for grabs since Maine still has two games at BU, which, if they lose both, would give us the TUC lead (7-5-0 beats 10-9-5). Not to mention both sets of playoffs.
Without looking again I think the St. Cloud comparison is possible to flip.

I try to point people at [slack.net] in hopes that they will do a better job of remembering the shorter address (and mentioning it on the air ). But here are a few direct links as well:

Explanation of the NCAA Selection Procedure (which I just updated to reflect the whole reduced-travel thing):
[slack.net]

"You Are The Committee" (yes, it's back): [slack.net]

Incidentally, the self-service hypothetical interface at[slack.net] can be used to make such depressing discoveries as: if Cornell had won the Everblades tourney, we'd be in the top four of the PWR and talking about a bye.

just wanted to say welcome to eric. its nice to see someone checking in on what i believe to be one of the greatest aspects of cornell before even attending freshman orientation. lynah is a great place. you'll love it!

Watching Cornell hockey (men's and women's, at least for me) ranks closely behind Ultimate and road-tripping to see concerts as one of the defining experiences I've had in college (and grad school). Oh, right, there were those pesky classes and shaping my academic future and becoming a well-rounded person and all of that, too. )

Enjoy college, Eric!

Excuse me if I'm being thick, since I don't really understand entirely how PWR is used in the tourny selection ( and I know it's not EXACTLY the criteria they use)...but, if we have a PWR of 12 at worst, and there are 12 spots in the tourny....wouldn't we get a slot? The 5 autobids are excluded, leaving 7 at large bids. In what situations would teams ranked #13 and below be able to leapfrog us and steal a bid?

jordan the problem this year is this...
if there are no upsets in the tournaments for each of the conferences then you will get 5 bids. Now all of the conference leaders are in the top 12 in the computers EXCEPT mercyhurst, the maac leader. This means that if there are zero upsets, only numbers 1-11 plus mercyhurst will into the NCAAs.
BUT
for each tourney upset that does not result in a top 11 team dropping, then the top 11 drops one.
so if NEstern and wisconsin win their tourneys, then you have to be in the top 9 to get into the tourney. (12= mercyhurst, 11=wisconsin, 10-NEstern).
hope i explained it well (and accurately)
welcome to cornell and cornell hockey, eric!

Jordan,

The way I understand it (someone please correct me!) is that the PWR is a model of the real selection process, which occurs behind closed doors. However, the committee has (intentionally) leaked enough information over the years for laymen to construct the PWR statistical model that mimics the behavior of the actual committee. As best as I can recall, the PWR process has correctly predicted the 12 teams selected correctly for several years in a row now, so it's a pretty good model (but possibly NOT the actual process). The much trickier part is the seeding, which is completely separate from the selection. Seeding is not based strictly on PWR; it takes many other factors (travel, attendence, intra-conference matchups, etc) into account.

I apologize for not looking at the SCSUs and UMinns of the world - I had (wrongly) assumed that they would dominate us in the PWR just because they won so many more comparisons than we did. However, as several people have pointed out, the insularity of scheduling allows a couple of bizarre results (read: Brown beating SCSU) to take on far too much significance in the grand scheme of things. Keith - if you say that RPI can move .01 at this point in the season, I certainly can't argue with you. I was thinking of it strictly from our point of view - our RPI is probably NOT going to go up by .01 given that most of the teams we have left to play are around .500, so I was thinking that we couldn't catch Maine. However, if we go up .005 and they drop .005, we are in business...

Thanks for all the thanks - I figured as long as I was poring over the comparisons, I may as well share my thoughts.

The reason why finishing in the top 12 is not enough for Cornell to get an at-large bid is that neither anyone in the MAAC nor any ECAC team besides Cornell is in the top 12 according to the pairwise. So if Cornell doesn't win the ECACs, the MAAC and ECAC champions will come from outside the top 12; even if there are no other upsets in conference tournaments, it will mean a top-ten finish is needed for an at-large bid.

The five selection criteria used to define pairwise comparisons between teams are spelled out in the NCAA D1 Hockey Championships Manual, available on line at
[www.ncaa.org]
So no mystery there. The part that is more complicated is how to use those comparisons to pick out the 7 at-large teams. Sometimes you can find seven teams that win all the comparisons with everyone else (aside from each other and the teams with automatic bids), but since the pairwise comparisons are not transitive (e.g., right now Cornell win the comparison with Minnesota who win the comparison with Michigan State who win the comparison with Cornell), it can get complicated, and the NCAA has never spelled out how they sort it out when it's not obvious.

PWR is a name USCHO gave to the total number of comparisons won by a team, and ranking teams by the number of comparisons won will always give the right answer if things are simple, and usually even if they're somewhat complicated. But I think USCHO oversells PWR to some extent, since there are conceivable situations where PWR does not give the same answer as we would expect from a consideration of the pairwise comparisons themselves.

So the "PWR has always predicted the tournament field" line (which is only true if you take out Quinnipiac for 1999 and 2000) is not so much a case of having constructed a rating that imitates some mysterious process as having chosen a simplification of the process which gives the same answer when things are simple, and getting lucky when they were not.

No...

Thanks..I knew I was missing something obvious.

Simple reasoning about RPI movements. Term #1 in the RPI is win pct, with 35% weight. Our current Win% is .740 (18-6-1). If we win out from here it will be .803 (26-6-1). That's a delta of +0.063, which when weighted by .35 gives an increase of +.022. Now, certainly adding games against 8 more ECAC teams will hurt the SOS portion of the statistic, but it probably won't hurt much. My rationale for saying that is that we'll be averaging more bas schedules into our already bad SOS number, so it shouldn't change much (as opposed to averaging 1.000 for wins into our current .740 win pct.)
The point is, I think, that at this point in the season changes in winning percentage drive the RPI more than the SOS factors, even though SOS is more highly weighted.