So this sent me down a random, looking at stats wormhole.
Last season, for playoffs and regular season combined, the Stars won 56 percent of faceoffs in victories. That's 58 percent in defensive zone, 56 in the neutral zone, and 55 percent in the attacking zone.
In losses, the Stars won 53 percent of faceoffs. They were 55 percent in the defensive zone, 52 percent in neutral, and 52 percent in attacking zone.
This is where we get anecdotal and how, and why coaches would use this data to motivate a team. Faceoffs are a reflection of effort, when you give a little bit more effort, 3 percent more in the faceoff circle, we win games. When we get less effort we lose.
In the playoffs only, the Stars in victories, they won 57 percent of the draws overall. 58 percent in the defensive zone, 53 percent in neutral, and 59 percent in the attacking zone.
In playoff losses, they were 53 percent total in the circle. 53 percent in the defensive zone, 48 percent in neutral, and 56 percent in the attacking zone.
Again anecdotal, but from a coaching message standpoint, 4 percent extra effort in the circle is the difference between winning and losing in the playoffs. That's a message you can package and deliver when asking for a bit more from tired lineup in the locker room.
To finish off my rambling, faceoffs matter for Dallas not because they win or lose games necessarily, but faceoff percentage is a pretty good reflection of whether Dallas played it's desired game in a a respective evening. It's not all about effort, but it was one of the few stats that we can anecdotally connect to it that isn't impacted by outside noise.
Great article about the nuance of faceoffs… two points, obviously possession is preferable, so does a possession advantage derived from faceoff wins show as a determining factor in wins?
Secondly, and this idea throws the ability to analyze faceoffs out the window… I once heard it said that faceoff win % isn’t important, what’s important is winning KEY faceoffs… ie Power Play offensive zone, PK in defensive zone… and I think some guys hold back their faceoff “trick” until those key moments… like holding back your super curve ball for strikeouts!
Really good point. I think the real key (statistically at least) is figuring out that inflection point for puck possessions: where do faceoff wins rank against shots off the cycle versus shots off the rush, or rush entries versus dump ins, etc. If 80 percent of hockey is played along the exterior of the ice, then how much do faceoff wins contribute to that 20 percent that's played along the interior, etc?
Id like add another baseball reference, I think faceoffs are akin to a catcher framing a pitch in baseball.
Its overall effect on the game is basically negligible, but when done well enough can create an advantage that can tilt the scales of human error in a game.
Or, to bring it closer to home.
It's a quick jab.
Your not winning the fight with it, but you can do some damage and set up spacing for a KO sequence.
Oct 4, 2023·edited Oct 4, 2023Liked by David Castillo
This is the most well-rounded analysis on faceoffs I've seen; very nice primer for the rabbit hole.
If a team had a faceoff win percentage of like 75%, I could see it making more of a significant difference in overall GF and GA. But the fact that the best faceoff teams in the league have a win rate of only 55-60% makes it hard to believe that it has significant impact over the course of a season. Given 60 faceoffs or so per game, that's only 6 more wins per game than the opposition (33-27) if you're at 55% for the game. So if you dominate the dots for an entire game at like an 75% clip (45-15), maybe it tilts the win-o-meter a bit for that night, but it's likely an overvalued stat.
As noted, hockey has too many other significant factors that are better determinants of success and is simultaneously too random for such minor differences to provide much in terms of predictability. Still, while it washes out in the long run, you didn't overlook the fact that a mere single faceoff (or zone entry, bad read, shot, etc) has the ability to change the outcome of a game - well played.
I went to high school with a dude named James Stevens, so I always get a little nostalgic when I see your name pop up. Haha. And appreciate it. I knew I wanted to cover where the research used to be and where it's headed without getting into the weeds, so I'm glad it's connecting with fans/subscribers.
I never finished writing it up, but even at 3-on-3 where everyone agrees possession is so important, winning a faceoff is not predictive of scoring the next goal, or even who takes the next shot. Bowness's insistence on using no-offense faceoff guys at the start of overtime still drives me batty.
David, it seems to me that if you have positive faceoff wins in both the offensive and defensive zones, then you should increase your ability to score and to keep the opposing team from scoring. Then theoretically, should this not improve your chances to have a positive scoring differential and positive puck possession?
Jerry! Thanks for joining the discussion. (Really surprised, honestly: I thought this would just be a throwaway post for readers)
Using the tornado vs. lightning analogy, I think faceoffs are similar: while a faceoff win (the tornado) seems to have broader implications on a team's performance, the real question is about the threat level (lightning)? A faceoff win is not simply a win versus a loss but a sequence: winning the puck (which doesn't always guarantee possession, although for the sake of argument we'll assume that we're talking about a CLEAN faceoff win), ensuring its retrieval (not guaranteed), maintaining possession (not guaranteed), and leveraging that possession into a unit of either offense or defense (also not guaranteed).
I think that's the real crux of the debate. Rather than ask what a faceoff is worth, it's better to ask what a faceoff sequence is worth, and within that, what kind of sequence converts faceoff wins into tangible offensive and defensive impacts. If that makes sense.
Great work! Bones had the same faceoff guys as DeBoer with different results. It would be interesting to see stats on how their teams took advantage of the faceoff wins.
I'm just guessing but maybe there's more advantage to a faceoff win for a good puck possession team. If you can convert faceoff wins to successful zone exits it should matter?
Your comment made me realize I cut something out that I should have left in (refresh my post!) so huge thanks Sami for the comment. It doesn't necessarily address your point, but it's something I mean to point out. But yes: definitely worth a deep dive at some point.
So we aren't going to miss Luke "faceoff dude" Glendening this season.
Faceoffs I think take on extra importance because they are a stat that is simple to track. I'd say that hits probably also fall into this category. What does a hit mean, what does it do. Just like the last second PK faceoff has importance, so can a hit. Likewise the opening faceoff as the game is just as trivial as many, if not most, hits. Turnovers and takeaways I can see as being more important, but still in the +/- camp of needing context to really be meaningful.
Exactamundo. It's the reason I so angry over seeing Dellandrea get scratched for Glendening. "Great, so Glendening is effective for one second. What about the other 12 minutes a game?"
Dellandrea is such an interesting case. One of those guys that wouldn't clear waivers due to multiple teams, but still this team doesn't know what to do with him. Oddly this year he's fighting for a spot on the 4th line, but with talented players and not just bit part guys. Though last night's game showcased that even with centers banged up, Dallas can modify the lineup pretty easily. I don't love it, but it at least makes PDB have to try out new things (or recycle Seguin to the top line which we all saw worked)
I always enjoy your work because it seems you nerd out a bit (same as I do)... Not to nit pick...but I believe the word you are looking for is homicide, not “homocide”.
Holy shit dude. Major thanks for catching that early.
And thanks thanks. I think that's part of what drew me into analytics to begin with. There's a rabbit hole for everything, and despite the reputation that hockey nerds have (know-it-alls), it's a grimy process really. Plenty of question marks, and open-ended questions in addition to the hard data.
I contemplated even mentioning the spelling error because I didn’t want to be a dick, but in this current world climate, I figured it would be worse for me not to mention it. I’ve never been satisfied with the generalized observations, claims or proclamations devoid of rationale or supportive evidence/analytics and the researched information almost always stimulates expansive or deeper questions worth researching. I appreciate it
So this sent me down a random, looking at stats wormhole.
Last season, for playoffs and regular season combined, the Stars won 56 percent of faceoffs in victories. That's 58 percent in defensive zone, 56 in the neutral zone, and 55 percent in the attacking zone.
In losses, the Stars won 53 percent of faceoffs. They were 55 percent in the defensive zone, 52 percent in neutral, and 52 percent in attacking zone.
This is where we get anecdotal and how, and why coaches would use this data to motivate a team. Faceoffs are a reflection of effort, when you give a little bit more effort, 3 percent more in the faceoff circle, we win games. When we get less effort we lose.
In the playoffs only, the Stars in victories, they won 57 percent of the draws overall. 58 percent in the defensive zone, 53 percent in neutral, and 59 percent in the attacking zone.
In playoff losses, they were 53 percent total in the circle. 53 percent in the defensive zone, 48 percent in neutral, and 56 percent in the attacking zone.
Again anecdotal, but from a coaching message standpoint, 4 percent extra effort in the circle is the difference between winning and losing in the playoffs. That's a message you can package and deliver when asking for a bit more from tired lineup in the locker room.
To finish off my rambling, faceoffs matter for Dallas not because they win or lose games necessarily, but faceoff percentage is a pretty good reflection of whether Dallas played it's desired game in a a respective evening. It's not all about effort, but it was one of the few stats that we can anecdotally connect to it that isn't impacted by outside noise.
Hope this all makes sense.
Really great point about how stats can be superficial without being meaningless.
Great article about the nuance of faceoffs… two points, obviously possession is preferable, so does a possession advantage derived from faceoff wins show as a determining factor in wins?
Secondly, and this idea throws the ability to analyze faceoffs out the window… I once heard it said that faceoff win % isn’t important, what’s important is winning KEY faceoffs… ie Power Play offensive zone, PK in defensive zone… and I think some guys hold back their faceoff “trick” until those key moments… like holding back your super curve ball for strikeouts!
Really good point. I think the real key (statistically at least) is figuring out that inflection point for puck possessions: where do faceoff wins rank against shots off the cycle versus shots off the rush, or rush entries versus dump ins, etc. If 80 percent of hockey is played along the exterior of the ice, then how much do faceoff wins contribute to that 20 percent that's played along the interior, etc?
It's funny to see the stats from tonight's preseason game, where the Stars decimated the Blues on the dots and lost 4-0. Interesting indeed.
Great write up David.👍🏻
I loves me some faceoff talk!
Id like add another baseball reference, I think faceoffs are akin to a catcher framing a pitch in baseball.
Its overall effect on the game is basically negligible, but when done well enough can create an advantage that can tilt the scales of human error in a game.
Or, to bring it closer to home.
It's a quick jab.
Your not winning the fight with it, but you can do some damage and set up spacing for a KO sequence.
Damn that's a great analogy.
Jorge Posada is the only reason I ever considered myself a Yankees fan.
This is the most well-rounded analysis on faceoffs I've seen; very nice primer for the rabbit hole.
If a team had a faceoff win percentage of like 75%, I could see it making more of a significant difference in overall GF and GA. But the fact that the best faceoff teams in the league have a win rate of only 55-60% makes it hard to believe that it has significant impact over the course of a season. Given 60 faceoffs or so per game, that's only 6 more wins per game than the opposition (33-27) if you're at 55% for the game. So if you dominate the dots for an entire game at like an 75% clip (45-15), maybe it tilts the win-o-meter a bit for that night, but it's likely an overvalued stat.
As noted, hockey has too many other significant factors that are better determinants of success and is simultaneously too random for such minor differences to provide much in terms of predictability. Still, while it washes out in the long run, you didn't overlook the fact that a mere single faceoff (or zone entry, bad read, shot, etc) has the ability to change the outcome of a game - well played.
I went to high school with a dude named James Stevens, so I always get a little nostalgic when I see your name pop up. Haha. And appreciate it. I knew I wanted to cover where the research used to be and where it's headed without getting into the weeds, so I'm glad it's connecting with fans/subscribers.
I never finished writing it up, but even at 3-on-3 where everyone agrees possession is so important, winning a faceoff is not predictive of scoring the next goal, or even who takes the next shot. Bowness's insistence on using no-offense faceoff guys at the start of overtime still drives me batty.
Good point, and thanks for stopping by! I'll be honest, this is my first time reading your stuff, but keep up the inspired work.
Thanks! I love reading your stuff, maybe I'll finish that look at 3-on-3 sometime.
David, it seems to me that if you have positive faceoff wins in both the offensive and defensive zones, then you should increase your ability to score and to keep the opposing team from scoring. Then theoretically, should this not improve your chances to have a positive scoring differential and positive puck possession?
Jerry! Thanks for joining the discussion. (Really surprised, honestly: I thought this would just be a throwaway post for readers)
Using the tornado vs. lightning analogy, I think faceoffs are similar: while a faceoff win (the tornado) seems to have broader implications on a team's performance, the real question is about the threat level (lightning)? A faceoff win is not simply a win versus a loss but a sequence: winning the puck (which doesn't always guarantee possession, although for the sake of argument we'll assume that we're talking about a CLEAN faceoff win), ensuring its retrieval (not guaranteed), maintaining possession (not guaranteed), and leveraging that possession into a unit of either offense or defense (also not guaranteed).
I think that's the real crux of the debate. Rather than ask what a faceoff is worth, it's better to ask what a faceoff sequence is worth, and within that, what kind of sequence converts faceoff wins into tangible offensive and defensive impacts. If that makes sense.
Great work! Bones had the same faceoff guys as DeBoer with different results. It would be interesting to see stats on how their teams took advantage of the faceoff wins.
I'm just guessing but maybe there's more advantage to a faceoff win for a good puck possession team. If you can convert faceoff wins to successful zone exits it should matter?
Your comment made me realize I cut something out that I should have left in (refresh my post!) so huge thanks Sami for the comment. It doesn't necessarily address your point, but it's something I mean to point out. But yes: definitely worth a deep dive at some point.
So we aren't going to miss Luke "faceoff dude" Glendening this season.
Faceoffs I think take on extra importance because they are a stat that is simple to track. I'd say that hits probably also fall into this category. What does a hit mean, what does it do. Just like the last second PK faceoff has importance, so can a hit. Likewise the opening faceoff as the game is just as trivial as many, if not most, hits. Turnovers and takeaways I can see as being more important, but still in the +/- camp of needing context to really be meaningful.
Exactamundo. It's the reason I so angry over seeing Dellandrea get scratched for Glendening. "Great, so Glendening is effective for one second. What about the other 12 minutes a game?"
Dellandrea is such an interesting case. One of those guys that wouldn't clear waivers due to multiple teams, but still this team doesn't know what to do with him. Oddly this year he's fighting for a spot on the 4th line, but with talented players and not just bit part guys. Though last night's game showcased that even with centers banged up, Dallas can modify the lineup pretty easily. I don't love it, but it at least makes PDB have to try out new things (or recycle Seguin to the top line which we all saw worked)
Honestly I think they just got tired of him taking penalties, which admittedly, IS a problem.
I always enjoy your work because it seems you nerd out a bit (same as I do)... Not to nit pick...but I believe the word you are looking for is homicide, not “homocide”.
Holy shit dude. Major thanks for catching that early.
And thanks thanks. I think that's part of what drew me into analytics to begin with. There's a rabbit hole for everything, and despite the reputation that hockey nerds have (know-it-alls), it's a grimy process really. Plenty of question marks, and open-ended questions in addition to the hard data.
I contemplated even mentioning the spelling error because I didn’t want to be a dick, but in this current world climate, I figured it would be worse for me not to mention it. I’ve never been satisfied with the generalized observations, claims or proclamations devoid of rationale or supportive evidence/analytics and the researched information almost always stimulates expansive or deeper questions worth researching. I appreciate it
Huge thanks to everyone in the comments section. Absolute banger after banger of insightful observations.