Is QRF Accurate?

[MNHockeyFan]

GRG beats Edina 3-1 in December. Edina beats Hopkins 5-1 a couple of weeks ago. Hopkins "squeaks out" as you say, a win over GRG 2-1 in OT. Around and around we go...

The only two teams that have approached true dominance this year are Blake in Class A and Eden Prairie in AA. However compared to GRG, Hopkins, Edina, Stillwater and Roseville, EP has played a much weaker schedule. But they've won almost all of the big games they have played so far, the lone blemishes being a 3-4 loss to Roseville and a 1-1 tie against Warroad.

Blake plays in a weaker conference, but has many tough non-conference wins to their credit, including Edina, Minnetonka, Irondale, Mayo, Eagan, and Warroad. Their matchup against Breck this week should be a good one. In their last game Blake and Breck tied 1-1, and Breck will be fired up to break their long winless streak against what has become their No. 1 nemisis.

[ice29]

Not sure if this is the answer, but my guess is how QRF assesses game results based not only on wins and losses, but scoring differential. I believe a couple of blowout wins or losses or an overall trend of close games, can affect the ranking to a significant degree under this model, as can scoring differential against common opponents. And as we all know, score differential can be affected by a number of factors such as a few empty net goals, injuries, or playing of backup players that made a game appear closer than it really was. I'm not going to analyze Silver Bay and Hibbing's game results to verify this (be my guest), but this might be the X-factor you're overlooking in your question.

I wouldn't get too lathered up about these statistical rankings. A formula could be tweaked to skew the results in a number of ways. QRF is simply one formula that the author has determined to be their best approximation based on the entire season to date. It's objective and unbiased, which is what's good, but if you don't agree with underlying criteria such as the goal differential factor or how the model weights games between the classes, it's probably going to yield results you don't agree with or some that seem "off".

An aside-- one major factor that is overlooked in a model such as this is how a team is performing now versus earlier in the year. A more sophisticated model might weigh recent results higher than those in November as the season moves ahead-- this is the thinking behind a "power ranking". QRF is more like the Bowl Championship Series computer ranking that weighs season-long results evenly.

Here's a riddle: If A beats B, and B beats C, but C beats A, who is the best? Answer: whoever won on February 23rd

I'll accept this as an answer, thank you!

Good Luck to All Sectional Participates!

[State Champ 97]

How does Elk River lose to Centennial and move up in the QRF?

[chickendance]

Ok, even better...Moorhead wins 3 stays at 20, Bemidji wins one loses 2 and moves up to 21 (+2), Irondale loses 1 and drops to 24 (-2)?

It's just a formula and the equation will always yield a number.