[squiggs96]

Not sure why the quote button isn't working for me, so the formatting may be messy.

Bingo says:
Quote:
Las Vegas was a cup finalist. SJ and Tampa preseason favourites. Penguins have two recent cups. Carolina in everyone’s list for pushing the play. Calgary clearly playing well.

Thats 6 of 10

Is 6 of 10 statistically significant enough to put any link between those stats and if a team is good or not? I, personally, don't think it is. 6 of 10 is 1 better than 50/50, so it may as well be a coin flip. That got me thinking. I ordered all 31 NHL teams alphabetically by location, and flipped 31 virtual coins. I assigned heads to a team being good and tails to a team being bad. Heads came up 18 times. I flipped 18 more virtual coins for those teams and 10 were heads and 8 were tails. The list of 10 top teams was then:


1. Arizona
2. Boston
3. Calgary
4. Nashville
5. New York Islanders
6. Ottawa
7. Philadelphia
8. San Jose
9. St. Louis
10. Vegas

I also ran it for nickname with the same process. Coincidentally 18 heads were flipped in the first round. This must mean my experiment is scientifically correct, as you just can't get 18 twice in a row. However, 11 heads were on the next round.My top 11 teams are:

1. Avalanche
2. Blues
3. Bruins
4. Ducks
5. Flyers
6. Jets
7. Kings
8. Lightning
9. Maple Leafs
10. Sharks
11. Wild

To preserve the integrity of my data, screen shots for all coin flips and data charts are available upon request.

By doing it by location we get Boston, Calgary, Nashville, San Jose, and Vegas. That's 5 out of 10, so only one worse than your method.

By doing it by nickname we get the Avalanche, Bruins, Ducks, Jets, Lightning, Maple Leafs, and Sharks. That's 7 out of 11. That's even better than 6 out of 10.

Bingo says:
Quote:
I’d argue that goal differential generally meet standings as a result so it wouldn’t tell you anything new
The general aim of hockey is to score more goals than the other team. The more times you do it, in theory, the more times you win. If you are trying to predict who is playing well, who will win, or who is a better team, then goal differential is a great statistic to look at. I'm not sure it tells you anything new, but it sure tells you who the best teams are. In the last 10 years an average of 1 team per year makes the playoffs with a negative goal differential. I've stated it before, but I'll repeat it again here: You don't get a positive goal differential from winning hockey games, you win hockey games by having a positive goal differential.



You said that goal differential won't tell you anything new. What did the stats that produced the list in post #72 tell us? It told us that by having some underlying numbers it can predict about half of the top 10 teams. In my opinion, that didn't tell us anything new that we can rely on. It's as if the stats don't correlate at all with telling us who the successful teams are.