[Discussion] Stats for 2021/2022 Canucks vs league comparison. Bonus! Did i figure out what a faceoff is worth?

[Bad_BOI_pete]

I Couldn't sleep so I tried to come up with a rudimentary way to calculate what a faceoff win is worth. When i finished my calcs I compared the canucks vs the league in 2021/2022.

 

My math for the faceoff factor seems to check out. Based on the Known of `10 shots = 0.9 goal. I calculated the best faceoff teams create 5 more shots on net each game. (5*82)*.9=36.9 goals. The total goals i Calculated from my Faceoff factor is 36. Thats only a 2.5 percent error, very close.

 

Whats a faceoff win worth? ES is worth about about 1.5% of a goal and on special teams its worth about 4% of a goal. Almost 3X more valuable.

 

Anyways on to the discussion, The canucks should have performed a lot better considering there faceoff numbers. This just proves that the Canucks have a long way to go.

 

 

	all situations shots	PP	PK
Faceoff factor  *	net + 5/GP	1.50%	2.00%
NHL average total	31.6	20.50%	79.40%
canucks Total	32	23.50%	74.90%
faceoff rank	6th	9th	17th
Canucks actual rank	15th	9th	31st
* The difference between the top 50% and worst 50% faceoff teams based on league averages.
	even strength Goals	PP Goals For	SH goals Against
Faceoff Factor  *	28	5	3
NHL Average Goals	Net 0	48.8	48.8
Canucks Goals	Net Plus 16	Total 58	Total 58
Faceoff rank	6th	9th	17th
Canucks actual rank	15th	8th	27th
* The difference between the top 50% and worst 50% faceoff teams based on league averages.

Edited October 22, 2022 by Petey_BOI

[204CanucksFan]

I'm very interested by the concept of your metric but I'm very confused how you came up with "Based on the Known of `10 shots = 0.9 goal", how is that "known"?

 

As well as what you are basing the best faceoff teams getting 5 more shots per game on.

[Googlie]

Interesting analysis.  Conventional wisdom would suggest that O-zone faceoff wins lead to more shots, and that D-zone wins help prevent shots.  And what role would be played by faceoff wins after an icing, where the objective isn't a shot on goal, but a dump in to facilitate a line change?

 

I have always thought that faceoff wins data was an underappreciated stat.  Your analysis certainly suggests that the Canucks should be better, being that they are (were) one of the better faceoff teams

[Bad_BOI_pete]

20 minutes ago, Googlie said:

Interesting analysis.  Conventional wisdom would suggest that O-zone faceoff wins lead to more shots, and that D-zone wins help prevent shots.  And what role would be played by faceoff wins after an icing, where the objective isn't a shot on goal, but a dump in to facilitate a line change?

 

I have always thought that faceoff wins data was an underappreciated stat.  Your analysis certainly suggests that the Canucks should be better, being that they are (were) one of the better faceoff teams

I could probably figure out most of what your suggesting, but to get league averages for 32 teams for each different faceoff point would be super time consuming. The problem with all the specialty faceoffs is that no one breaks them down by league average. If they did, it would be super easy.

 

 

[Bad_BOI_pete]

1 hour ago, 204CanucksFan said:

I'm very interested by the concept of your metric but I'm very confused how you came up with "Based on the Known of `10 shots = 0.9 goal", how is that "known"?

 

As well as what you are basing the best faceoff teams getting 5 more shots per game on.

The average save percentage in the league is about 0.91. 1 shot * 10= 10 *.91= .9

 

like i said its rudimentary calcs. i took the top 16 and bottom 16 teams for faceoff win% in ES , PP and PK and for each respective side came up with a average for shots, PP% PK% ES Goals, PP Goals and shorthanded goals. then i found the diffrence between the top and bottom for each Stat. That is how I came up with the faceoff factor.

 

Each Stat is calced separately and to make sure the calculations were close I compared The Expected net shots Vs the expected Net goals and thats how I know the calculations are close within 2.5%

 

of course there is room for error, but with over 4000 faceoffs that is a significant amount of data.

[-AJ-]

Interesting. I remember years ago, I tried to statistically prove that teams with higher face-off percentages won more games, but found that the result was statistically insignificant. Even though it did result in higher wins, the difference wasn't big enough to be statistically significant.

 

Nice to know that you have a good sample size of face-off data though, so sample size shouldn't be much of an issue.

[Bad_BOI_pete]

30 minutes ago, -AJ- said:

Interesting. I remember years ago, I tried to statistically prove that teams with higher face-off percentages won more games, but found that the result was statistically insignificant. Even though it did result in higher wins, the difference wasn't big enough to be statistically significant.

 

Nice to know that you have a good sample size of face-off data though, so sample size shouldn't be much of an issue.

The heart of the teams are all very close in faceoff win % the 8th place team last year and the 24th only had 3% difference. But the difference between 1st and last was 9% which is very significant. I think at some point in the game Key faceoffs can be a tactic used to dominate possesion time. Remember when Boone Jenner took all the faceoffs in OT? or the Vegas series a few years back? 

 

The more polarizing the advantage, the more tactics can arise. You will see that even teams like colorado with a low faceoff percentage overall have a very decent offensive zone win percentage and a terrible Nuetral zone win% and the polar opposite is a team like the islanders where they win a whole bunch in the neutral zone and are just average in the offensive and defensive zone's.

 

Another team that uses faceoffs tactically was florida even though they were ranked 22 overall in faceoff win% but 4th in Powerplay faceoffs.

 

Is it a coincidence that Carolina Dominated the NHL in both SH faceoffs and PK?