No... the point is that a 55% face off man who is generally ineffective at the other vital parts of the game shouldn't be out on the ice in crucial situations.
- jtommyt

I guess most if not all NHL coaches disagree...which is why guys that have the best face off % are typically out there at the end of the game if there's a faceoff

Faceoffs:

The main reason faceoffs don't matter, and what I suspect the reason for their non-correlation to winning is, is because it's impossible to be very good or bad at faceoffs.

If someone was so terrible at faceoffs, they wouldn't likely take any. So what happens is that at the professional level, no one wins or loses at faceoffs by a very big discrepancy.

Toews lead the league with a 58.6% faceoff percentage. Ribeiro was last with 37.5%

If they took 100 faceoffs vs each other, it'd be 60-40 Toews, but all Ribeiro has to do is win an extra 1 faceoff out of 10 and they are even.

So that's why faceoffs don't really matter: at the professional level the best and worst guy are barely any different. And therefore, everything else being equal, saying a guy's faceoff skills as a reason to keep or play him becomes ridiculous.

Bozak, who I believe is the reason we're having this conversation, is an OK player, but even if he was the best faceoff player in the NHL, which he isn't, that wouldn't be a reason you should factor in your decision to play him.

Yes, a faceoff win gives you possession, but possession changes like every 2 seconds or something similarly small, so it barely matters.

Yes, sometimes a goal is scored off a faceoff, but this is exceptionally rare. There are more than 50 faceoffs per game, and you see a goal scored off the faceoff a couple times per season if you only watch one teams' games.

The math has been done, and it's estimated you need a faceoff +/- of +250 to equal a goal.

In the end, if your team never one a faceoff, you would likely not win many games, but since it's impossible to get an advantage big enough to really matter (outside of random anomalous games which I am sure occasionally occur) it is safe to say that within the confines of what can reasonably be expected to occur in an NHL game, that faceoffs don't really matter.
- James_Tanner

Why don't they just give the home team the puck first and rotate each stoppage then ?
Seriously I don't see how anyone who watches hockey can not see how faceoff wins are important

I guess most if not all NHL coaches disagree...which is why guys that have the best face off % are typically out there at the end of the game if there's a faceoff
- Redmile247

That's why new Era stats work is so important... some long held beliefs really don't bear out in the face of evidence.

Toews lead the league with a 58.6% faceoff percentage. Ribeiro was last with 37.5%

If they took 100 faceoffs vs each other, it'd be 60-40 Toews, but all Ribeiro has to do is win an extra 1 faceoff out of 10 and they are even.

So that's why faceoffs don't really matter: at the professional level the best and worst guy are barely any different. And therefore, everything else being equal, saying a guy's faceoff skills as a reason to keep or play him becomes ridiculous.


- James_Tanner

For someone who presents themselves as an analytics guy that's some pretty awful math. Statistics don't work that way. You can't take a statistic that's based on face offs against everyone and transfer it directly to a one on one comparison. If Toews was always against chumps in the face off circle, his winning percent would go up.

If Toews and Ribero took 100 face offs against each other, Toews probably wins 90 of them.

Why don't they just give the home team the puck first and rotate each stoppage then ?
Seriously I don't see how anyone who watches hockey can not see how faceoff wins are important
- Redmile247

I feel like people are having two separate arguments.

Individual faceoffs are important, example being last minute of the game, pp draws etc.

The *statistic* of faceoff % is useless, there is little variance over a period of time. Therefore judging a player, or players off of faceoff % is in fact useless

I feel like people are having two separate arguments.

Individual faceoffs are important, example being last minute of the game, pp draws etc.

The *statistic* of faceoff % is useless, there is little variance over a period of time. Therefore judging a player, or players off of faceoff % is in fact useless
- sbroads24

It's not useless ...just maybe not as useful as some might think I get that ...but there's a reason why centers with terrible face off % don't just ignore it cuz it's useless ....they work hard at it in practice ...vets who are good at draws will show rookies little tricks etc ...I'm willing to buy into what NHL players and coaches believe over a few internet guys who think they know what they are talking about

That's why new Era stats work is so important... some long held beliefs really don't bear out in the face of evidence.
- jtommyt

New era stats still are completely flawed and only tell a small amount ...you can't use stays to quantify why things happen in hockey ...there are too many variables that make comparing invalid

It's not useless ...just maybe not as useful as some might think I get that ...but there's a reason why centers with terrible face off % don't just ignore it cuz it's useless ....they work hard at it in practice ...vets who are good at draws will show rookies little tricks etc ...I'm willing to buy into what NHL players and coaches believe over a few internet guys who think they know what they are talking about
- Redmile247

Buying what exactly? There is 0 proof that face offs as a statistic have any bearing on team success long term. The best 3 teams in the NHL in face off% all missed the post season. How can that possibly be if it were an important facet of the game ?

Any singular event can be labeled as important. An event repeated hundreds of times where the results are similar to a coin flip from team to team become somewhat meaningless

Buying what exactly? There is 0 proof that face offs as a statistic have any bearing on team success long term. The best 3 teams in the NHL in face off% all missed the post season. How can that possibly be if it were an important facet of the game ?

Any singular event can be labeled as important. An event repeated hundreds of times where the results are similar to a coin flip from team to team become somewhat meaningless
- sbroads24

you can make that arguement with corsi , any advanced stats if you try hard enough ...I just try and use common sense which tells me I'd rather have the puck than be chasing it ...what percentage of goals are scored when you don't have possession ? Seriously how can anyone argue against having centers who win more draws than they lose

you can make that arguement with corsi , any advanced stats if you try hard enough ...I just try and use common sense which tells me I'd rather have the puck than be chasing it ...what percentage of goals are scored when you don't have possession ? Seriously how can anyone argue against having centers who win more draws than they lose
- Redmile247

Who in the hell is arguing this? The argument is that you don't employ a player who is inferior in other aspects of the game, simply because they are good at faceoffs

Who in the hell is arguing this? The argument is that you don't employ a player who is inferior in other aspects of the game, simply because they are good at faceoffs
- sbroads24

It happens more than you think ...players that excel at one aspect are employed all the time

It happens more than you think ...players that excel at one aspect are employed all the time
- Redmile247

No it happens as much as I think it does. And they are usually bad players. I see Paul Gaustad and his FO prowess is still team-less at the moment

New era stats still are completely flawed and only tell a small amount ...you can't use stays to quantify why things happen in hockey ...there are too many variables that make comparing invalid
- Redmile247

But what they do tell is based on evidence. Which is still miles better than old school thoughts which are based on nothing.

No it happens as much as I think it does. And they are usually bad players. I see Paul Gaustad and his FO prowess is still team-less at the moment
- sbroads24

Well if you say so since you apparently know all

But what they do tell is based on evidence. Which is still miles better than old school thoughts which are based on nothing.
- jtommyt

Meh when players who are not that great are apparently better than a Shea weber or whoever else becomes the pro analytics whipping boy I start to put less and less stock in the evidence ...good tools but still non fancy stat intangibles can play just as much of a role as playing players who are simply better corsi wise

Well if you say so since you apparently know all
- Redmile247

I don't know all, I do know that I wouldn't bank on being a good team because I win more faceoffs than other teams

I don't know all, I do know that I wouldn't bank on being a good team because I win more faceoffs than other teams
- sbroads24

You might not be ...but the odds would be more in your favour of you did

Faceoffs:

The main reason faceoffs don't matter, and what I suspect the reason for their non-correlation to winning is, is because it's impossible to be very good or bad at faceoffs.

If someone was so terrible at faceoffs, they wouldn't likely take any. So what happens is that at the professional level, no one wins or loses at faceoffs by a very big discrepancy.

Toews lead the league with a 58.6% faceoff percentage. Ribeiro was last with 37.5%

If they took 100 faceoffs vs each other, it'd be 60-40 Toews, but all Ribeiro has to do is win an extra 1 faceoff out of 10 and they are even.

So that's why faceoffs don't really matter: at the professional level the best and worst guy are barely any different. And therefore, everything else being equal, saying a guy's faceoff skills as a reason to keep or play him becomes ridiculous.

Bozak, who I believe is the reason we're having this conversation, is an OK player, but even if he was the best faceoff player in the NHL, which he isn't, that wouldn't be a reason you should factor in your decision to play him.

Yes, a faceoff win gives you possession, but possession changes like every 2 seconds or something similarly small, so it barely matters.

Yes, sometimes a goal is scored off a faceoff, but this is exceptionally rare. There are more than 50 faceoffs per game, and you see a goal scored off the faceoff a couple times per season if you only watch one teams' games.

The math has been done, and it's estimated you need a faceoff +/- of +250 to equal a goal.

In the end, if your team never one a faceoff, you would likely not win many games, but since it's impossible to get an advantage big enough to really matter (outside of random anomalous games which I am sure occasionally occur) it is safe to say that within the confines of what can reasonably be expected to occur in an NHL game, that faceoffs don't really matter.
- James_Tanner

Correction: + 50 ish for a goal + 250 for a win. Or something like that. Ill try and find the original article and post it.

For someone who presents themselves as an analytics guy that's some pretty awful math. Statistics don't work that way. You can't take a statistic that's based on face offs against everyone and transfer it directly to a one on one comparison. If Toews was always against chumps in the face off circle, his winning percent would go up.

If Toews and Ribero took 100 face offs against each other, Toews probably wins 90 of them.
- Hunkulese

Thats a fair point but your overlooking what i was trying to say on a technicality. Its not meant to be a real world example just to show you how close the best and worst guy are.

This faceoff debate really illistrates a problem with out society: feelings.

Feelings are not facts.

The only answer to the facts presented here: I did not know that. It's counter-intuitive, but it makes sense. I will now change my mind.

If someone tells you factual information and you choose to ignore it, that is ridiculous.

That's it.

Faceoff percentage doesn't matteer. Facesoffs don't correlate to winning, aren't a fact in player evaluations and rarely effect the outcome of a game. You shouldn't have a "faceoff specialist" on your team. You should never mention a player's faceoff skill as a reason to play him or keep him on the team.


And this isn't anyone's opinion. This is fact.

You might not be ...but the odds would be more in your favour of you did
- Redmile247

Not really though. Arizona led the league in FO%, by your logic, wouldn't they have possession more than any other team as well?

This faceoff debate really illistrates a problem with out society: feelings.

Feelings are not facts.

The only answer to the facts presented here: I did not know that. It's counter-intuitive, but it makes sense. I will now change my mind.

If someone tells you factual information and you choose to ignore it, that is ridiculous.

That's it.

Faceoff percentage doesn't matteer. Facesoffs don't correlate to winning, aren't a fact in player evaluations and rarely effect the outcome of a game. You shouldn't have a "faceoff specialist" on your team. You should never mention a player's faceoff skill as a reason to play him or keep him on the team.


And this isn't anyone's opinion. This is fact.
- James_Tanner

Wow. Here is the part I don't understand, and it's really important to the whole discussion. That second last paragraph you wrote is not fact. It is your opinion.

Faceoffs are a facet of the game. To be victorious in the game, you must prove to be better than your opponent in enough facets of the game to tilt the outcome in your favour. No one facet is the end all and be all in any game. But to speak as if it does not factor into the outcome on any level is nonsense.
If your logic was sound, centers would just stand at the faceoff dot with their stick across their knees. Lost draw, no possession, chasing the play, getting tired in my own zone? who gives a f#$k? 'Cause draws aren't important.
For someone who spouts about being so open-minded, you really like to submit your opinions as facts and dismiss any others' opinions as misguided or ill-conceived. A little medicine for you James: being intelligent doesn't make other people stupid.

Wow. Here is the part I don't understand, and it's really important to the whole discussion. That second last paragraph you wrote is not fact. It is your opinion.

Faceoffs are a facet of the game. To be victorious in the game, you must prove to be better than your opponent in enough facets of the game to tilt the outcome in your favour. No one facet is the end all and be all in any game. But to speak as if it does not factor into the outcome on any level is nonsense.
If your logic was sound, centers would just stand at the faceoff dot with their stick across their knees. Lost draw, no possession, chasing the play, getting tired in my own zone? who gives a f#$k? 'Cause draws aren't important.
For someone who spouts about being so open-minded, you really like to submit your opinions as facts and dismiss any others' opinions as misguided or ill-conceived. A little medicine for you James: being intelligent doesn't make other people stupid.
- Tonybere

Yes, but some facets are 100x more important than others.

Not one single person has presented a shred of data or evidence to dispute what James or others have said.

Faceoff % does not matter over the course of a season. That is a stone cold fact.

Yes, but some facets are 100x more important than others.

Not one single person has presented a shred of data or evidence to dispute what James or others have said.

Faceoff % does not matter over the course of a season. That is a stone cold fact.
- sbroads24

NO! I don't get how this isn't obvious, but that is not a fact. It is your opinion.
A fact is, losing ONE draw does not matter over the course of a season. The idea that if your team didn't win a single draw all season it wouldn't matter is what you people are saying with your ridiculous "fact."

Yes, but some facets are 100x more important than others.

Not one single person has presented a shred of data or evidence to dispute what James or others have said.

Faceoff % does not matter over the course of a season. That is a stone cold fact.
- sbroads24

AND, James and the others have not presented on shred of EVIDENCE to support what they have said.