Could someone please come on and tell how the rankings will be done at the end of Oct. Will it be from a computer system. I can’t see how it can be done from myhockeyrankings.com since that had been rumored. Most teams don’t report all there games there. I hope how the rankings will be done will be posted here/
From M2 Commissioner Jay Peacock email on Oct 13th, 2016:
Rankings-
Because we can’t add all the teams across the country who are non-ACHA opponents and it’s not fair to teams that play tougher opponents that take a loss to improve play, we are going to be purging the games played against non-ACHA opponents.
I have been in conversation with MYhockeyrankings a number of times over the last week. The director has told me a number of you have been writing him because your win loss is not up to date. Please note that the first official ranking will not be until November 2nd. He is getting up to speed as we speak and he will begin preliminary rankings soon. Attached is some information/descrition from them that may help you better understand the process.
Attachment content is below…
MYHockey ‘s ratings are computed mathematically, with no subjective weighting or human determined values. MYHockey rates teams based upon how well they play against other teams and how good those teams are. These two factors are “AGD” or “average goal differential” and “SCHED” or “strength of schedule”. AGD is currently calculated by accumulating the goal differential of each game, to a maximum of 7, and divides it by the number of games played. Using a max game goal differential of 7 does “penalize” teams that blow out opponents, but this is common in rating systems as it takes away the incentive for teams to run up scores and allows them to have a bad game by capping the mathematical damage. The strength of schedule is computed by averaging the rating of each game opponent. AGD and SCHED are added together to compute a team’s rating.
There is no preferential treatment for teams from certain leagues. All teams are assumed to be equal and having a rating of zero before the system mathematically calculates your SCHED and AGD based entirely upon game performance. A team’s SCHED is not only determined by the teams it plays, but by the teams its opponents play and its opponents’ opponents play, etc. All teams in a given age level are ranked in one statistical pool. The math does not know if the teams are traditionally strong or not. It simply computes based upon who a team has played (SCHED) and how well they have done (AGD).
In general, as teams start getting 20+ games under their belts, the system tends to be very accurate. There are, however, some occasional exceptions.
If a group of teams (i.e. a league) doesn’t play others outside its group, their placement relative to all other teams cannot be accurately determined, but relative to one another it is accurate. Groups of teams with limited outside play can be inaccurately skewed by the scores of only a few games, which might not be representative of the group’s overall ability. Accuracy significantly increases with more games and more interplay and it makes sense for there to be a minimum number of games before teams are ranked.
Assuming teams have all played enough games against both league and non-league teams, inaccuracies may still exist. Some teams play differently (a lot differently) depending upon which goalie is in net for a specific game. In this case, the ratings average out the play and may not completely reflect either scenario. Missing players, home vs. away, double rostered players are a few other scenarios that could influence the accuracy of the ratings. It’s not an exact science. The ratings should be used to determine what teams might be good non-league opponents and which bracket of a tournament may be more appropriate. They are not meant to be the AP Poll for youth hockey.
There are a couple weaknesses to the current system. The first deficiency is with teams that emphasize development in some games. They are “penalized”. That same team may then focus on winning big tournaments and succeed, but may never achieve the rating that they “deserve”. Secondly, teams that “come on” in the second half of the season are “penalized” by early poor performances. All games throughout the season count equally and that may not reflect a teams performance near the end of the season.
1 Like
thanks for your quick response i know this is the procedure now but in my opinion myhockeyrankings .com nor anyone else will convince me that all teams ARE EQUAL. In the Northeast there is no way teams in the American North Division and most of the American South teams are even close to the Sechl teams. In the Southeast how can teams in the ACC and CSCHC be considered equal to teams in the MACH and TSCHL. Most of the teams in the Southeast have enough games for being ranked there shouldn’t be a team from the ACC I’m the top 10 I don’t see how this will work Since this is new will have to see how it works and giver it a chance I have been following d2 hockey for a little over 3 years but am not a youngster just like going to watch the kids play
Myhockeyrankings is hilarious. Somebody is gonna get screwed BIG time with this set up. You need a mix of both. Computer and the “old school committee”. That will give you the best overall rankings come tournament time.
With myhockeyrankings it’s all about winning games period!!! Throw SOC out the window. If it stays this way next season, watch the middle of the road teams load up on lousy D3 squads. A wins a win and that’s what’s going to push teams into the regional picture if you stay with this “new and improved” setup.
I feel bad for the programs that actually carry a heavy work load this season and will get beat out by the programs who don’t deserve it. Makes regionals a joke overall when the best aren’t even in the tournament. Which is t that the point. Having the 10-12 “BEST” programs go???
I’m actually supportive of the ACHA trying out MyHockeyRankings.com. As noted in the explanation above, the system produces a more accurate output as more games are factored into the algorithm. At first the rankings may seem a little wonky, but they should improve by December.
I like that the algorithm takes strength of schedule and average goal differential into account, and that the system not only calculates based on opponent-to-opponent, but also the other teams that each opponent plays. That way it’s actually beneficial to play teams that a) are at a similar level or tougher; and b) play other challenging opponents. Seems to me there’s a disincentive to play a cupcake schedule and blow out weak teams if one wants to rank well. I also like that there’s a built in disincentive to run gains up past a differential of 7 points.
I know that D1 uses computer rankings once the season gets underway. Initially, there’s not enough data to produce accurate computer rankings, so coaches vote to start the process. But as more data is accumulated, a computer ranking system takes over and generate pretty good rankings.
In past years it seems there were problems getting rankings out in a timely manner using a committee-based ranking system. It would be great if rankings could start off with a coaches/committee vote like in D1 until enough data is available to produce better computer rankings. Who knows, maybe they’ll try that system next season.
If you have followed the MyHockeyRankings in past seasons, they were generally pretty accurate compared to the ranking committee. The math does take into great consideration strength of schedule when computing rankings.
Here are the best examples that should ease your concerns:
West:
Cal State Northridge - is ranked at #15 despite holding a 6-0-1 record. This includes all teams in the West - even those that have not played in minimum 5 games.
Southeast:
Princeton - ranked at #12 currently despite a 9-0-0 record. This also includes all SE teams without 5 games recorded.
If you select a team and then the tab ‘Rating Math’ it shows which game results help or hurt the overall ‘Rating’. Cal State’s win over Loyola Marymount is the best example dropping them 3.88 overall despite a 1 goal win.
1 Like
Right - I forgot about the Rating Math tab. Explains a lot when a team wins big against a weak opponent and their ranking drops or stays steady or notches up just a skosh.
I think you should re read the part from my hockey rankings that caps the games at 7 goal differential. That actually hurts you a lot if you schedule those cup cake teams.
And I really enjoy the my hockey rankings addition that takes into account only playing league teams over a mixed schedule to verify the legitimacy of a potentially ranked team come Dec.
So to bad for the super leagues with 10-14 teams. Not a lot of non conference games to add into the math. Darn
1 Like
MHR says a loss is s loss, But ACHA has an OTL category. Will MHR adjust its point system to treat OTL as a lesser hit?
MyHockeyRankings has had their algorithms and format for years, I would not expect them to change for the ACHA. So yes, losing a game 4-3 in regulation versus losing a game 4-3 in OT will be weighted the same.
I think the addition of MHR is fantastic for the ACHA. I feel that over the past few seasons, the ACHA has made some mistakes when it has come to rankings. Introducing a standardized system takes away the user error, as well as team/conference bias (which undeniably does exist). I am confident that come late February, you will be seeing the most deserving teams at regionals. If you look back over the past few years (in the past rankings tab), you’ll notice quite a few .500 teams that were ranked fairly high, and many .750+ teams ranked low. A strong schedule really does matter.
1 Like
An OTL or SOL should really count as a tie.
I agree that MHR a good addition to the ACHA ranking system. From my experience it’s been a relevant, but fun system until now. It is now both irelevant and meaningful. In that regard I would like MHR to continue to develop and refine. Make it even better with time and further minimize the “borders”. In the end we know it will be the border between #2 and 1-2 spots just below as well as #10 and 1-3 spots below. Having said that I realize those arguments will always exist.
This is a positive move as it also improves transparency.
For many years, there have been complaints about certain leagues getting preferential treatment when it came time for rankings.
There will probably still be complaints come auto-bid and regional tournament time, which happens in every college sport.
Regardless of format if you schedule good teams and win games, the ranking will take care of itself.
Is there a timeline as to when scores get posted to the ACHA?
Some scores take a long time to show up, therefore I assume it affects the rankings for that week.
Is it up to the home team to submitt?
Anybody can add scores to MHR. You can sign up to add or add as a guest. The site gets updated on Wednesdays.