View Single Post
Old 10-07-2021, 01:38 PM   #1
mrkajz44
First Line Centre
 
mrkajz44's Avatar
 
Join Date: Oct 2010
Location: Deep South
Exp:
Default Five Thirty Eight - Now with NHL predictions! - Update: 20 games

I don't wade into the Fire on Ice forum too much, so apologies if this has already been posted. I really enjoy following Five Thirty Eight for their sports predictions for MLB and NFL, but they never had one for NHL until this year:

https://projects.fivethirtyeight.com...l-predictions/

For those following the offseason, the projection for the Flames is pretty much expected. Their ELO rating is 1502, which is basically as "average" as you can get. They are forecast to get 91 points and have a 54% chance of making the playoffs. Hard to argue with much of that in my opinion.

These will be updated as the season goes on. It's basically a more powerful (and likely more accurate) version of SportsClubStats, which I haven't used in a few years. I enjoying watching it for other sports, so I figured others might like it for hockey. They also have a section about which games are "good" to watch based on the quality of the teams and the playoff importance to the result. Lots of middle of the road matchups to start off of course as playoff implications are far down the road.

In their introduction to the predictions (https://fivethirtyeight.com/features...t-this-season/), I found this interesting:

Quote:
The NHL gives out points for losing. Does that matter?

It does in our season forecast, which simulates the rest of the schedule 50,000 times and tracks where every team ends up in the standings. In losses that go into overtime — both in real life and our simulated future contests — a team will get the Bettman Point. However, one of the interesting findings in the research we conducted for NHL Elo ratings was that how a team won or lost — in regulation, overtime or the dreaded shootout — doesn’t seem to matter when measuring a team’s quality. When it comes to predicting a team’s future games, all one-goal contests are created equal, even if they happen in a glorified skills competition and the loser winds up receiving different numbers of points.
It seems the loser point really does skew the standings quite a bit, where team strength is mostly unaffected by if they lose close in regulation vs OT/shootout.

Don't worry Squiggs - this will never be brought up in the snake thread.

*For those curious, there is a detailed breakdown of how the predictions work:

https://fivethirtyeight.com/methodol...dictions-work/

It gets pretty detailed, but the high level approach is they use ELO ratings to simulate the season 50,000 times. For those not familiar with ELO, the average is 1500 and you go up or down through the season. If you win your ELO goes up, and losing makes your ELO go down. The further the pre-game spread in ELO, the larger adjustment that is made after the game (ie. if you beat a team that has a much higher rating than you, you'll go up much more than if you had beat a team with a similar or lower rating to you).
__________________
Much like a sports ticker, you may feel obligated to read this

Last edited by mrkajz44; 11-24-2021 at 11:27 AM.
mrkajz44 is offline   Reply With Quote
The Following 7 Users Say Thank You to mrkajz44 For This Useful Post: