Eddie Schorr
Member
So, Ohio State enters the weekend at 9, wins the series against a higher ranked team and leaves the weekend as 10. Someone want to explain that to me. That's a pretty flawed formula.
USCHO Fan Forum
This is a sample guest message. Register a free account today to become a member! Once signed in, you'll be able to participate on this site by adding your own topics and posts, as well as connect with other members through your own private inbox!
-
The USCHO Fan Forum has migrated to a new plaform, xenForo. Most of the function of the forum should work in familiar ways. Please note that you can switch between light and dark modes by clicking on the gear icon in the upper right of the main menu bar. We are hoping that this new platform will prove to be faster and more reliable. Please feel free to explore its features.
Latest pairwise rankings
- Thread starter Speed Kills
- Start date
Well, I do see what you're saying here. I don't think anyone will argue it's a perfect system. But the answer is probably as simple as Western Michigan won both their games this weekend even if it was only against Miami. I would agree though. If we were adding a human element, then yes, Ohio State shouldn't drop behind WMU based on the results this weekend.
Okay....one can read the PWR Prob Matrix on CHN, and it looks like everyone down to Michigan Tech at 11 is in. I like Alaska's situation, because I think that, for example, Mankato only stays ahead of them by winning the CCHA tourney. Personally, I don't see any upsets, so I would guess that Minnesota knocks out MSU on Saturday night, and MTU and MSUM end up playing in a week for the CCHA. 4 B10s, 2 CCHAs, 3 NCHCs, Alaska, BU, AHA champion, Quinn, Harvard. That's 14. And, leaves, seemingly, 2 spots for some combo of Cornell, Merrimack, or NEastern.
Anyone have predictions?
Anyone have predictions?
Not a prediction but it would be cool to see some of these postseason meetings have added value. So as an example if Merrimack plays NU in HE semis it would be cool if that turned out to be a play in game to NCAA tournament. The Hockey East title game should be fun though if somehow BU isn't in it and it's a win and you're in for both teams or even if BU is in it and the other team has to win to get in.
Eddie Schorr
Member
And I can almost see the Western aspect despite the pathetic level of competition they played. The real puzzler is how Penn State only drops one to 8 and stays the same distance ahead of Ohio State that they were before the series. That's an awful lot of weight given to one OT win on the road versus losing the other two in regulation.
Lost_Husky
Late to the Party
Short high level answer, Pairwise isn't solely based on your wins. Yes, how your team does is a significant input, but how the teams you have wins over performed has an influence on your RPI/PWR as well.
PW spots 8-11 are incredibly tight. That margin is miniscule (.0008 RPI points) between OSU (10) from PSU(8)/WMU (9) who are tied in RPI. MTU (11th) is right there if OSU slips only .0017 points behind. Those are small margins and a lot of movement can happen.
If the schools OSU beat (say Notre Dame) had a bad weekend and falls in the rankings, that hurts your totals as well. If a team like Western sweeps and the teams Western beat had a great weekend, this gives Western a extra boost.
Going 2-1 against a high PWR team and other game results could have hurt you more than someone like Tech and WMU sweeping low PWR teams, especially if the quality teams they previously beat had a great weekend.
That is convoluted formula with a lot of variables. To give you a better idea of what you can expect from OSU, go over and check CHN's Probability Matrix. It won't clear this issue up, but it will give you an idea of variability and probability for where everyone can end up.
Last edited:
StateofHockey78
New member
I'd rather use math, than a committee's subjectivity though. The formula is well understood by all the teams and it all comes down to winning games. You can't lose to teams like LIU and expect for it not to come back to bite you. That's exactly what is happening right now with OSU's position based on losing a key comparison with UConn due to them having both the H2H and Common Opp points (LIU). You win the LIU game and I guarantee your RPI and comparisons would have put you ahead of both those teams. All that being said, OSU is in the driver's seat to improve their position. A win against Michigan will vault you over PSU and possibly Western. Western is playing CC who is low pairwise and you'd get bonus pts for beating UM on the road. Big opportunity to jump to 7-8 IMO.
bruinsfan11
Member
I feel like Ohio St is gonna end up in Minnesota’s region in most scenarios between the math and Penn St hosting in Allentown
mookie1995
there's a good buck in that racket.
It is only complete at the end of the season. That is what it was created to do.
it is nice to look at before the end, but again it only comes into play after the regular season is completed.
StateofHockey78
New member
if they end up 8,9, or 10 I would agree.
I think the exception to this is if either OSU wins the B10, or if Mich beats OSU, and MTU wins the CCHA, in which case the Huskies will end up ahead of OSU, and will go to Fargo.
purpleinnebraska
Well-known member
Yeah, but "math" is still somewhat subjective. Someone has to decide what counts, and how much it counts. What was the old component, last 10 games or something like that? I thought that was a good idea, for teams that were missing players earlier in the season but now were at full strength. But that was taken out because people felt it favored (or didn't favor, depending on who was complaining) the smaller conferences. Anyway, not to relitigate the battles over components of the formula (55%? 65%?), but calling it math doesn't eliminate subjectivity.
StateofHockey78
New member
I think that is a logical viewpoint. We can certainly debate about the formula, but IMO the pairwise works pretty well (understanding nothing will please everyone). Pretty sure I heard they were moving to the NPI next year so we'll all have to relearn the math/formula yet again!
Although I agree with the OSU posters (I think OSU is a much better hockey team right now than PSU), It wasn't surprising to me to see PSU maintain their place going 2-1 on the weekend just based on the math.
Could absolutely see this happening for Merrimack provided they get through to the semi-finals (remember, this is unfamiliar territory for my Warriors!). I "think" MC might have enough points to get in even if they lose to BU in the final since Merrimack swept BU in the regular season (well, took 5 of 6 points).
Oh, and by the way, I'd LOVE to see a QU vs MC match in Bridgeport.
You are correct, in my view, about math being subjective. Statistically, the CHN Power Rankings (what used to be called KRACH) are the best, because they can retro-predict every team's record. But I don't like them because I think that the math is too heavily weighted toward the non-conference schedule, and it is very possible for a particular league to have a strong OOC record based on playing the middle teams in the other conferences. So, even that isn't perfect.
The L16 criterion was removed because of the possibility that some teams might have an easier end of their schedule.
Ultimately, one can find reasons to complain about every system involved.
The PWR is reasonable, and beats the eye test.
Also, hockey, at least in individual games, can be a game of bounces. There is something inherently unpredictable about it. Hot goalies, broken sticks, penalty calls. All combine to make one-and-done not really 'fair', but we will never get anything else.
Bronco
Onondaga United IV.6
LOL. Says the guy that didn't even schedule a NCHC team this season. If you want better comparisons then play at least one series against the other big conferences instead of loading up on wins against Atlantic Hockey celler dwellers.
Bronco
Onondaga United IV.6
Looking at the NCHC tourney...
Omaha. If they can get past NoDak, they have a winning record against WMU this season. I don't see them winning over Denver in the championship, but if they can get there, perhaps that is enough to leap them into the top 14.
UMD has had SCSU's number this season. I wouldn't be shocked to see the Bulldogs in the semifinals against Denver. Don't think they will, or anyone, will touch Denver, but they could make it interesting.
moose97
New member
No upsets? When almost every conference has gone to single elimination? If you say so...
This is a good point. I see Ohio State went a combined 5-1 against Mercyhurst, Bentley and Long Island and added two wins against subpar Bowling Green. Their toughest non conference test was a series against UConn and they went 0-1-1. So 7-2-1 in non conference play and not a single impressive win in the lot.
Penn State's non conference is even more pathetic. They happened to luck into the fact that Alaska was good this year but I can promise you Guy Gadowsky didn't schedule them with that in mind:
Canisius
Mercyhurst
St. Thomas
Alaska
RIT
OK fine, RIT was good this year. But sorry, that's embarrassing scheduling.
But hey the counter is always that it's just math. But if you take this scheduling philosophy and take care of business in the non conference, just scratch and claw your way in conference play to around a .500 record or so. Seems like a successful formula. If the goal is simply to make the field anyway.
StateofHockey78
New member
I'm not a OSU/PSU supporter and I don't push the B1G on people either, but the schedules don't surprise me. Are they weak non-conf schedules? Absolutely, but that is not uncommon. All the power conferences schedule games at a fairly equal rate with the weaker schools (NCHC played 10 games against AHA, B1G 11, HEA 20. NCHC played 21 games against the CCHA, B1G 15, HEA 2. NCHC 10 against ECAC, B1G 3, HEA 48).
If I was in their shoes and I saw what MN & Michigan (and ND/Wisco to a lesser extent) were bringing in every year talent wise, I would not want to load up my non-conference schedule on top of an already challenging conference either (unless I felt what I had in the current class of players was a serious contender). If your team is mid-level or worse then yes the goal is grind out a .500 conference record and hope you can take care of business in non-conference (College Basketball is a prime example of this strategy working time after time for mid-level teams). I think Rohlik/OSU does a better job landing a few top talent players, but neither of these schools is competing with the top schools nationally in terms of recruits IMO. PSU and OSU for example are not going to be stealing many recruits from DU, NoDak, UMD, St. Cloud, BU, BC, MN , Mich, etc, etc.