In this week’s edition of the power rankings, I have completely revamped the model that churns out the numbers. If you’re curious, this is a post explaining the new method, but the gist is that I’m now using score-adjustments rather than score-close. That plus a re-weighting of 5-on-5 and special teams data.
Here are last week’s rankings with the old method, and with the adjustments there’s obviously some big changes this week as some teams went up thanks to the adjustment, and some teams went down.
The biggest news is there’s a new number one this week as Pittsburgh’s long reign at the top is over thanks to scaling back the weight on special teams.
The chart below is how this week’s rankings are shaping up.
10 Thoughts on Rankings
1. First thing’s first let’s talk about who benefits from the adjustments: Los Angeles, Nashville, New York Islanders. All three are horrid on special teams and this helps them out. In Los Angeles’ case they are elite in when leading or trailing, but very mediocre when the game is tied, so using score-adjustment gives them a bump.
2. Now the team’s that drop: Toronto, Washington, Vancouver. All three are average teams who are in the top ten using the old method thanks to very good special teams. Toronto is especially curious because they’re the opposite of LA; ok when it’s tied, but terrible when leading or trailing.
3. Chicago is the best team in the league and that is pretty much non-debatable. They’re the early favourites to win the Cup and it’ll be hard to knock em off the way they’re playing right now.
4. Last week’s ranking on the chart applies to what it was last week using this method, not the actual rankings. As you can see, there’s a lot more stability as very few teams move up or down in a week’s time. That leaves me with very little to actually discuss, so while I improved the model, I may have cannibalized this column. Whoops.
5. Anaheim has the biggest drop this week. That despite having the best record in the NHL. Look at their roster right now and tell me that it’s at all deserving, because I don’t know how they manage it.
6. Two Saturday’s ago I was watching the Flames lose to San Jose and made my friend a $25 bet they wouldn’t make the playoffs. That went as predicted very quickly. Thanks for the free money, Jon.
7. In terms of the upper echelon of the league, there seems to be three distinct tiers. Cup Contenders: Chicago, Pittsburgh, Tampa Bay. Cup Hopefuls: Nashville, St. Louis, Detroit. And the Dark Horse teams: New York Islanders, Minnesota (if they ever get goaltending) and the Los Angeles Kings.
8. On the other end of the spectrum, no one is as bad as Buffalo, don’t let their recent winning streak fool you. They’re still far and away the worst team in the league.
9. Columbus is finally healthy (ish) and I imagine that they start climbing out of the basement in these rankings very soon. They played a very strong game last night against Detroit.
10. Toronto just keeps winning despite being completely average. We’ve seen this story before, here’s hoping it has a better ending.
For my projections, I adjust POP to get a slightly more accurate read on how many points to expect. To do that, I looked at how spread apart the teams are this season, and compared that to how spread apart teams were in the past meaning the ratings are less volatile, which gives off slightly more stable projections.
Here’s this weeks:
Here’s a visualization of how the projections changed daily during the week. Credit to @IneffectiveMath on Twitter for the idea.
Click it to enlarge, because clearly this blog format wasn’t made for pictures.