Better at faceoffs

[ricosuave]

The Flames, and in particular, our young players have been way better at faceoffs then we have ever seen in the past.

I just stumbled across this article and it is exactly what many of us have been preaching about for years. Why don't more teams do this? Why don't the Flames bring Otto back for faceoff coaching, and maybe even other power play or penalty kill specialists (Yelle?) to help the coaching staff out?

http://www.sportsnet.ca/hockey/nhl/y...guru/?shawct=1

Discuss!

[driveway]

Because, in the long run, faceoffs don't actually matter.

[Geeoff]

Haven't we been getting killed at faceoffs in the playoffs?

[ignite09]

Originally Posted by driveway

Because, in the long run, faceoffs don't actually matter.

Would like to see you expand on that. Do you have anything to back that up with?

[AcGold]

Originally Posted by Geeoff

Haven't we been getting killed at faceoffs in the playoffs?

Yeah really badly, especially against Anaheim. Monahan is in real bad shape and that's the worst part of it.

[Caged Great]

Originally Posted by driveway

Because, in the long run, faceoffs don't actually matter.

It makes things a lot harder when you're losing 10% more of the draws and that can bite you in situations where you're pushing for a goal or preventing one. It's small, but it does add up over the course of a season, and the playoffs can be changed entirely by a single bit of good/bad luck and having the possession off the draw decreases the chances for bad luck and increases them for a good bounce.

[Street Pharmacist]

Originally Posted by ignite09

Would like to see you expand on that. Do you have anything to back that up with?

If you look at the best teams at face offs vs the worst teams, there's very little correlation with winning. Face offs are helpful in certain situations, but on the whole, they're not as important as they get made to be

[thefoss1957]

Originally Posted by driveway

Because, in the long run, faceoffs don't actually matter.

Just about the only goals the 'Hawks allowed to the Wild were directly off lost d-zone draws...Faceoffs drive possession, and possession drives scoring...to say that faceoffs don't matter seems fallacious.

The 'Hawks have Yannick Perrault coaching Centers in their system, and he has done wonders for Marcus Kruger, to the benefit of the 'Hawk PK.

[Street Pharmacist]

Originally Posted by thefoss1957

Just about the only goals the 'Hawks allowed to the Wild were directly off lost d-zone draws...Faceoffs drive possession, and possession drives scoring...to say that faceoffs don't matter seems fallacious.

The 'Hawks have Yannick Perrault coaching Centers in their system, and he has done wonders for Marcus Kruger, to the benefit of the 'Hawk PK.

3 of the top 5 face off teams in the league are either eliminated or about to be. Overall face off % means little. Strategically, a great face off guy is useful, but not essential.

Is very complex. Like you said, face offs drive possession, possession drives scoring, and scoring drives winning. However, there's enough chance involved in each of those links to really minimize the impact.

The problem is magnitude. If the best teams win 80% and the worst won 20%, you'd have a point. However, the best face off teams are maybe 10% better than the worst. If a goal is scored off a defensive zone loss once every 20-30 lost d zone face offs (that's a huge stretch, it's likely more rare), and your only going to lose 10% less than your opponent, I wouldn't call that much of an advantage over a seven game series. Maybe 1 goal?

[Mister Yamoto]

Originally Posted by ricosuave

Why don't the Flames bring Otto back for faceoff coaching, and maybe even other power play or penalty kill specialists (Yelle?) to help the coaching staff out?

Tim Hunter - Fight Coach

[driveway]

Originally Posted by ignite09

Would like to see you expand on that. Do you have anything to back that up with?

http://statsportsconsulting.com/main...lysis12-12.pdf

As part of this analysis we analyzed 211,372 faceoffs from the 2008-9, 2009-10, and 2010-11 regular seasons.
...
a player must win about 76 more faceoffs than they lose in order to obtain a goal differential for his team.
...
A team that moves from winning 50% of their faceoffs to winning 60% of them gains just over 12 goals per season which is equivalent to two additional wins.

No team really approaches 60% of faceoff wins over the course of a season, the best in the last few seasons was Boston at 56.4% in 12-13. While the best players do add to their team's chances, and overall being good at winning faceoffs or losing faceoffs can impact a team positively or negatively, the effort needed to improve isn't really worth the gain you get out of it.

In particular, for the seasons that we studied, San Jose as a team gained approximately 6.1 goals per season as a result of their prowess winning faceoffs ... this means that San Jose has earned approximately one additional win per season via winning faceoffs.

Yes, one win can be important, but San Jose was at the extreme high-end of the scale for the study (Edmonton, as they are no good, was at the opposite end). There are more impactful ways to spend time, energy, and money than on improving faceoff winning percentage.

There was exactly 0 difference between the Flames faceoff wins and Canucks faceoff wins last series. Right now the Ducks are up on us by 28 faceoff wins, or about .3 of a goal. It's worth noting that almost all of that discrepancy came in game 3, where they dominated us 40-19 in faceoffs, and we won.

Faceoffs are weird, they feel important. They look important. People talk about them like they are important and every now and then a pivotal goal is scored off a faceoff, or a faceoff win results in a vital clearing of the puck, or seals a win in the last seconds.

Players should practice faceoffs. It's good to have players on your team who are good at winning faceoffs. They are absolutely not worth stressing over, losing sleep over, and should not be a priority for training, as their impact is minimal.

[EldrickOnIce]

That's the thing with most stats though - in a best of 7000, faceoffs would likely mean little.
But situationally, a faceoff win (or loss) can be the difference between winning and losing a single game, and therefore winning or losing a short series.

This seems important to me.

[Itse]

5 of the 10 best PP teams didn't even make the playoffs. Does not mean powerplay doesn't matter? Of course not. It just means that powerplays ALONE will not get you anywhere. But you still want to have a good powerplay.

Then there's the idea that faceoff percentages are all pretty close to each other anyway and gaining an advantage in them is really hard. This is a terrible argument. The NHL is an ultra-competitive environment, lots of stats are going to be close to each other, and gaining an advantage in anything is going to be really hard. Just because your team is not very likely to be in significantly better shape than the opposing team doesn't mean that you don't want to be in the best shape you can be.

Anybody who is dismissive of the idea that faceoffs might get you a couple of extra wins over a season is out to lunch. 4 points is a huge deal in today's NHL. Even 1 point can be a huge deal. You want that point. Everything matters. Teams even want to have good equipment guys, and they're not even on the ice.

One thing I've been really disappointed in hockey statistics is the seeming lack of searching for statistically significant combinations. For an easy example, it's somewhat obvious that winning faceoffs by itself will not do much. But how about being a good faceoff team AND a good special teams team, especially on the powerplay? Does that give you notably more goals? How many?

Or, does it matter how the team likes to score otherwise? Are faceoffs more important for teams that like to cycle in comparison to teams that like to score off the rush?

Then there's the point Eldrick mentioned. In a short series things are less about statistics and more about probabilities. If every offensive zone faceoff win gives you a really, really small chance to score, you should try really hard to win them. Sure you might get just one goal out of it, but that goal could be a GWG that ultimately is the difference between winning and losing the series.

You can also compare them to puck battles in the corner. Two evenly matched teams are probably going to split those about 50/50. The odds of any one battle making a difference is really, really small. But you still want to try and win them all, because you never know which ones will matter and what will the series come down to in the end.

[Poe969]

Originally Posted by ignite09

Would like to see you expand on that. Do you have anything to back that up with?

The Flames are pretty terrible at face offs and they're in the second round. Winning face offs would help because as pointed out, getting the puck off the draw when your in the offensive zone and trying to score or winning a big defensive draw in your zone when the other team is trying to score just gives you more position time and a better chance to do what you're trying to. I think the Flames hustle and compete level is what's won them games.

[Enoch Root]

It's a microscosm vs macrocosm issue.

Can a faceoff be important? Absolutely. If it is in one end in a close game, winning or losing it could very well impact the game.

But the thing is, there are thousands of them over a season. As driveway pointed out above, even though FO wins can result in goals, it averages out over a season and it is very difficult to be dominant enough for it to have a net impact.

If it takes 60% to be worth 2 extra wins, then the very best team in the league each year is only generating 1 extra win. One. And the vast majority of teams would all be net zero.

Yet every night, FOs are discussed as a key component of the game.

Someone brought up PPs. Are they important? Of course. But over the course of the season, the total goal differential from team to team is marginal. A strong PP might be worth a handful of goals at best, and maybe a win or two more over the season than an average team.

This is precisely why the Corsi-cans look at 5 on 5, close stats.

[Erick Estrada]

I don't know. The Ducks have had a lot of zone time in the Flames end because of faceoff wins. That zone time may not result in a goal against but it's also time the Flames can't score so I take any stats with a grain of salt seeing you can't score when the puck is in your own end and when you take faceoffs in the offensive zone and keep losing them you also aren't scoring when you are losing puck possession. Poor faceoff percentage is one of the reasons the Flames are not a great advanced stat possession team. Ideally you would like to improve this as it does make a difference.

[bubbsy]

here's another example of where stats seem really messed up for hockey.

Possession is a MASSIVE concept to articulate a team's control of a game (having the puck should not only mean you have a higher chance of creating scoring opportunities, but it also means reducing the chances created by the opposition ). the concept of "shot attempts" being the indicator for a team's "possession" seems incorrect to me.

i think the ability to quantify the puck's state in all 3 zones may provide more context and value, where state could be team A possession, team B possession, contested possession (board battles, turned over pucks, etc).

if this could be tracked, we could have stats like:
avg time/game Calgary possesses the puck in (off/def/neutral) zone.
avg time/game Calgary opposition possesses the puck in (off/def/neutral) zone.
avg time/game Calgary is contesting for puck possession (off/def/neutral) zone

(my idea is likely stupid, my point is that shots being used in isolation to quantify possession seems incomplete).

[Street Pharmacist]

Originally Posted by Erick Estrada

I don't know. The Ducks have had a lot of zone time in the Flames end because of faceoff wins. That zone time may not result in a goal against but it's also time the Flames can't score so I take any stats with a grain of salt seeing you can't score when the puck is in your own end and when you take faceoffs in the offensive zone and keep losing them you also aren't scoring when you are losing puck possession. Poor faceoff percentage is one of the reasons the Flames are not a great advanced stat possession team. Ideally you would like to improve this as it does make a difference.

If it makes a difference, why don't better face off teams win more games?

[DuffMan]

I'd rather see a better at power plays change.

[Blixa]

Originally Posted by Street Pharmacist

If it makes a difference, why don't better face off teams win more games?

Maybe they had poor goaltending or lot of injuries in defence? It´s about a summary of lot little things and faceoffs are one off them.