Highlander
Active member
Ya when I was a kid, I didn't care if Tim Horton was making $25.00 a game or Dave Keon was making $75 a game, I just didn't care.caveman said:another example of the bloody salary cap... the discussion about paycheques....
Dermott Re-Signs
- Thread starter louisstamos
- Start date
Guilt Trip
Active member
Dermott signed his qualifying offer sheet so it's good for the Leafs, this season. I'm just guessing but I think Dubas offered him something just over 1 mill per for 2 years. Either way, the Leafs don't have to protect him in next year's expansion draft and Dermott has a year to prove himself.Frank E said:
Also, from what I can gather, Dermott has ARB rights after next season as this will be his 4th year of pro hockey and he signed his ELC in the 19-20 year old range.
Guilt Trip
Active member
Malholtra should help him out in the dot.Deebo said:
Kin
New member
Joe S. said:
I guess but the thing is that pre-cap every team did have a cap to one extent or another. Whether a result of basic economics or an owner's largesse every team still had a budget and guys getting paid a lot did affect what could be spent elsewhere. One of the biggest lies the Pro-Owner folks had during the lockout was that before the lockout big market teams were just spending like crazy and making the game uncompetitive and it's just never been true.
dappleganger
Active member
Deebo said:
Let me just say this, I'll be very interested to see where Kerfoot lines up this season, at centre or on the wing.
Heroic Shrimp
New member
https://twitter.com/ChartingHockey/status/1320111161057120256
herman
Well-known member
Zanzibar Buck-Buck McFate said:
It's pretty hard to answer with 3 stats for all skaters so it's going to be long
different roles and playstyles profiles lend themselves to different emphases. It's kind of why Wins Above Replacement/Goals Above Replacement models are created to try to be profile agnostic.The basic 'advanced' stats are just a zoom out of the standard shots/goals/points/+- stats.
Consider the sales funnel of a sales organization: at the mouth, the widest portion, you have your passive advertisement, marketing, social engagement efforts; once you get to direct sales you have sales calls, then quotations, then sales that are actually agreed to and delivered on.
In hockey 'advanced stats' you have players' on-ice decisions (and puck bounces) within two competing schemes (and how well executed they are at any given instance) happening; the stats are just results logged of those micro events. At the mouth is having the puck on your team's stick: currently untracked on NHL.com without the chip trackers. Within the funnel proper, you have
- shot attempts (Corsi): CF/CA
- shot attempts that don't get blocked (Fenwick): FF/FA
- shot attempts that make it on net (shots): SF/SA
- shot attempts that make it into the net (goals): GF/GA
More advanced is simply taking those stats and basically doing a +/- for those events, and then calculating it over a rate of time (per 60 min), either isolating to the individual or logging when that individual was on the ice (and with whom). Even more advanced is taking an aggregation of shot metrics crossed with shot locations and developing an Expected Goals model, which applies an average success rate for a shot from any position on the ice given a league average shooter against a league average goaltender. This is the xGF/xGA you might see.
The unfortunate thing is we currently do not have public data on events that lead up to shots (where the passes come from and go to, where the shots target on the net).
Cross-referencing actual goal results against 'effort' results (upstream events) sort of gives a clearer picture of what's happening. Standard shooting percentage is a simplistic version of this (goals/shots on goal for). Auston Matthews, for example, has fairly above average shot attempts share, somewhat average xG, but a GF% that outperforms his xG. This means he controls play decently, takes shots from mid-range, but has either really good shooting talent or very good setups (he has both!). Zach Hyman has great CF%, exceptional xGF, but not the greatest number of goals (historically speaking). As you know from watching him, he frees up pucks for his teammates to get chances, goes hard to the net for rebounds, but his hands are somewhat granitey.
So to finally summarize, the stats that seem most pertinent to me at the moment are the ones that players have direct control over, be it because of their talent, adherence to a good system, or some other micro-decision that yields results (Matthews' toe-drag release). It'll be a bit of a different focus per player type if we also mix in microstats that are currently hand-tracked (zone exits/entries).
Very basic forward stats
- Shot attempts share (CF%)
- Expected goals share (xGF%)
- Goals share (GF%)
Very basic defenseman stats
- Shot attempts against (CA60): do they give up a lot of shot attempts
- Expected goals against (xGA60): do they help prevent dangerous chances
- Zone exits/assists: do they help move the puck out of the zone
ZBBM
Active member
herman said:Zanzibar Buck-Buck McFate said:
It's pretty hard to answer with 3 stats for all skaters so it's going to be long
The basic 'advanced' stats are just a zoom out of the standard shots/goals/points/+- stats.
Consider the sales funnel of a sales organization: at the mouth, the widest portion, you have your passive advertisement, marketing, social engagement efforts; once you get to direct sales you have sales calls, then quotations, then sales that are actually agreed to and delivered on.
In hockey 'advanced stats' you have players' on-ice decisions (and puck bounces) within two competing schemes (and how well executed they are at any given instance) happening; the stats are just results logged of those micro events. At the mouth is having the puck on your team's stick: currently untracked on NHL.com without the chip trackers. Within the funnel proper, you have
- shot attempts (Corsi): CF/CA
- shot attempts that don't get blocked (Fenwick): FF/FA
- shot attempts that make it on net (shots): SF/SA
- shot attempts that make it into the net (goals): GF/GA
More advanced is simply taking those stats and basically doing a +/- for those events, and then calculating it over a rate of time (per 60 min), either isolating to the individual or logging when that individual was on the ice (and with whom). Even more advanced is taking an aggregation of shot metrics crossed with shot locations and developing an Expected Goals model, which applies an average success rate for a shot from any position on the ice given a league average shooter against a league average goaltender. This is the xGF/xGA you might see.
The unfortunate thing is we currently do not have public data on events that lead up to shots (where the passes come from and go to, where the shots target on the net).
Cross-referencing actual goal results against 'effort' results (upstream events) sort of gives a clearer picture of what's happening. Standard shooting percentage is a simplistic version of this (goals/shots on goal for). Auston Matthews, for example, has fairly above average shot attempts share, somewhat average xG, but a GF% that outperforms his xG. This means he controls play decently, takes shots from mid-range, but has either really good shooting talent or very good setups (he has both!). Zach Hyman has great CF%, exceptional xGF, but not the greatest number of goals (historically speaking). As you know from watching him, he frees up pucks for his teammates to get chances, goes hard to the net for rebounds, but his hands are somewhat granitey.
So to finally summarize, the stats that seem most pertinent to me at the moment are the ones that players have direct control over, be it because of their talent, adherence to a good system, or some other micro-decision that yields results (Matthews' toe-drag release). It'll be a bit of a different focus per player type if we also mix in microstats that are currently hand-tracked (zone exits/entries).
Very basic forward stats
- Shot attempts share (CF%)
- Expected goals share (xGF%)
- Goals share (GF%)
Very basic defenseman stats
- Shot attempts against (CA60): do they give up a lot of shot attempts
- Expected goals against (xGA60): do they help prevent dangerous chances
- Zone exits/assists: do they help move the puck out of the zone
Thanks again ? this is super-helpful. I literally took a screenshot to keep as reference. And so, by the powers vested in me,* I hereby confer upon herman a Doctorate in Hockimetry, with all the rights and appurtenances thereof. 8) 8) :-* :-*
I actually sort of followed you.
So here's my 30,000-foot reaction: all these stats measure micro-events, you say. I would disagree on one count: GF/GA. That's not a micro-event; it's not even a macro-event. It is THE macro-event ? the only one that counts, ultimately. The whole point of this exercise, the reason some of us have spent the better part of our mortal finite lives on this site, is because we are emotionally invested in seeing a particular group of athletes get their names etched onto a trophy. (Why we are invested is the subject for another dissertation.)
What I don't see, or at least if it's there I don't understand it, is a statistical correlation between all these other advanced stats and GF/GA. I am skeptical of putting too much weight on a statistic (or bundle of statistics) that can't be shown to have a clear effect on the only statistic that counts.
More than that, really useful metrics need to be predictive. Can advance stats, singly or (more likely) in combination predict who is likely to win the most games over the course of the season, in short. If they can't (yet), they need further development ? in my untutored opinion. But maybe these formulations exist, and I just don't know about them ... quite possible.
* I self-vested
Just thinking out loud here. The predictions that sites like moneypuck have take some advanced stats into account I believe, but because of parity generally you don't see a team with a slam dunk predictive win and because of the random nature of hockey the better team doesn't always win (hot goalie etc.). It doesn't make the stats bad, it just means hockey is a very random sport which is very hard to predict and account for in some way and the league has lots of parity.Zanzibar Buck-Buck McFate said:herman said:Zanzibar Buck-Buck McFate said:
It's pretty hard to answer with 3 stats for all skaters so it's going to be long
The basic 'advanced' stats are just a zoom out of the standard shots/goals/points/+- stats.
Consider the sales funnel of a sales organization: at the mouth, the widest portion, you have your passive advertisement, marketing, social engagement efforts; once you get to direct sales you have sales calls, then quotations, then sales that are actually agreed to and delivered on.
In hockey 'advanced stats' you have players' on-ice decisions (and puck bounces) within two competing schemes (and how well executed they are at any given instance) happening; the stats are just results logged of those micro events. At the mouth is having the puck on your team's stick: currently untracked on NHL.com without the chip trackers. Within the funnel proper, you have
- shot attempts (Corsi): CF/CA
- shot attempts that don't get blocked (Fenwick): FF/FA
- shot attempts that make it on net (shots): SF/SA
- shot attempts that make it into the net (goals): GF/GA
More advanced is simply taking those stats and basically doing a +/- for those events, and then calculating it over a rate of time (per 60 min), either isolating to the individual or logging when that individual was on the ice (and with whom). Even more advanced is taking an aggregation of shot metrics crossed with shot locations and developing an Expected Goals model, which applies an average success rate for a shot from any position on the ice given a league average shooter against a league average goaltender. This is the xGF/xGA you might see.
The unfortunate thing is we currently do not have public data on events that lead up to shots (where the passes come from and go to, where the shots target on the net).
Cross-referencing actual goal results against 'effort' results (upstream events) sort of gives a clearer picture of what's happening. Standard shooting percentage is a simplistic version of this (goals/shots on goal for). Auston Matthews, for example, has fairly above average shot attempts share, somewhat average xG, but a GF% that outperforms his xG. This means he controls play decently, takes shots from mid-range, but has either really good shooting talent or very good setups (he has both!). Zach Hyman has great CF%, exceptional xGF, but not the greatest number of goals (historically speaking). As you know from watching him, he frees up pucks for his teammates to get chances, goes hard to the net for rebounds, but his hands are somewhat granitey.
So to finally summarize, the stats that seem most pertinent to me at the moment are the ones that players have direct control over, be it because of their talent, adherence to a good system, or some other micro-decision that yields results (Matthews' toe-drag release). It'll be a bit of a different focus per player type if we also mix in microstats that are currently hand-tracked (zone exits/entries).
Very basic forward stats
- Shot attempts share (CF%)
- Expected goals share (xGF%)
- Goals share (GF%)
Very basic defenseman stats
- Shot attempts against (CA60): do they give up a lot of shot attempts
- Expected goals against (xGA60): do they help prevent dangerous chances
- Zone exits/assists: do they help move the puck out of the zone
Thanks again ? this is super-helpful. I literally took a screenshot to keep as reference. And so, by the powers vested in me,* I hereby confer upon herman a Doctorate in Hockimetry, with all the rights and appurtenances thereof. 8) 8) :-* :-*
I actually sort of followed you.
So here's my 30,000-foot reaction: all these stats measure micro-events, you say. I would disagree on one count: GF/GA. That's not a micro-event; it's not even a macro-event. It is THE macro-event ? the only one that counts, ultimately. The whole point of this exercise, the reason some of us have spent the better part of our mortal finite lives on this site, is because we are emotionally invested in seeing a particular group of athletes get their names etched onto a trophy. (Why we are invested is the subject for another dissertation.)
What I don't see, or at least if it's there I don't understand it, is a statistical correlation between all these other advanced stats and GF/GA. I am skeptical of putting too much weight on a statistic (or bundle of statistics) that can't be shown to have a clear effect on the only statistic that counts.
More than that, really useful metrics need to be predictive. Can advance stats, singly or (more likely) in combination predict who is likely to win the most games over the course of the season, in short. If they can't (yet), they need further development ? in my untutored opinion. But maybe these formulations exist, and I just don't know about them ... quite possible.
* I self-vested
herman
Well-known member
Zanzibar Buck-Buck McFate said:
GF/GA is extremely co-related to wins
The other stats that eventually lead to GF/GA are more loose for sure, CF being the furthest. At this point, Corsi is just a proxy for who has the puck enough to direct shots at the net. If anything, the number of shot attempts and the ratio against, is more an indicator of the team system/talent base than it is about individual players. How a line is deployed, who they are matched up against, even how they line change, affects Corsi individually. On a team level, score effects also play a part: teams with big goal leads get turtley, and vice versa.What data from 2007 to now says, is that Team CF% is pretty stable and consistent year over year provided the core or coaching system doesn't overhaul. Goal events can be very spiky for stretches (goalie goes nuts for 4 wks, or Mitch Marner/John Tavares/Morgan Rielly shoot ~14% together for 2 months, or the team wins a slew of one-goal games in the playoffs) and can skew results enough to change an organization's fortunes even if the team truly is no bueno. Translation: teams with strong CF% but poor results can be expected to get better results without doing too much different; teams with weak CF% but winning can be expected to come back down to earth when their goalie goes back to normal luck in a couple of weeks. It's really just a matter of looking at the underlying foundational elements of the results to see how sustainable the macro results are.
Teams with strong xG% over the season tend to win. As Bender points out though, hockey is randomer than most people are comfortable with. Basketball is pretty random too, but there are soooo many scoring events that it really reveals the good teams over time (like 3 games or week). Take a game of heads or tails. Play 5 rounds and you get wonky ratios like 5-0, 4-1 most of the time. Play 100, and it gets closer to 50-50, unless one side is weighted (talent, skill). GF/GA are rare in hockey (0-8 total goals per game, most commonly 4-6); CF/CA are very frequent (unless it's NYI vs CBJ), thus a much more stable indicator of talent/skill/system efficacy.
ZBBM
Active member
So you're saying, at this stage in stats development, that xG% is currently the best single indicator of success, even though it's far from perfect. Yes?
Another question: what do you think will be the big breakthrough tech advance that revolutionizes hockey stats? I think it could be real-time continuous player movement tracking ? both of the player's location on the ice and where his stick is located at any given moment, in relation to the puck.
Side comment about randomness: are some sports really less random than others? I have my doubts. All sports have clear rules and result in an artificially binary outcome: W/L rankings over a set time period, the "season" (which assimilate ties in sports that still have that as a game outcome). Within those boundaries the games can proceed in essentially infinite ways; no two games are ever even close to being exactly the same. So on the face of it, saying a sport is more or less random than another doesn't say much because the range of possible ways each game can unfold is so large.
Another question: what do you think will be the big breakthrough tech advance that revolutionizes hockey stats? I think it could be real-time continuous player movement tracking ? both of the player's location on the ice and where his stick is located at any given moment, in relation to the puck.
Side comment about randomness: are some sports really less random than others? I have my doubts. All sports have clear rules and result in an artificially binary outcome: W/L rankings over a set time period, the "season" (which assimilate ties in sports that still have that as a game outcome). Within those boundaries the games can proceed in essentially infinite ways; no two games are ever even close to being exactly the same. So on the face of it, saying a sport is more or less random than another doesn't say much because the range of possible ways each game can unfold is so large.
Highlander
Active member
or Axe throwing.Bullfrog said:
herman
Well-known member
Zanzibar Buck-Buck McFate said:
No. It's just a more refined way of looking at the events of a game to see what the teams accomplished than raw shot attempts/shots on goal/final score. It's not the single best indicator of success but it does illuminate whether the game results can be relied on to predict future results to a degree.
Player/puck tracking for sure will help, so we'd be able to see and evaluate off-puck play more objectively over large sample sizes. Teams that haven't invested in data scientists, db infrastructure by now will be swamped with waaaaay too much information and have a very hard time discerning any signal in the noise. What they've shown on screen at the all-star game is largely noise: who skates the fastest on any given night, who had the hardest shot, etc.Zanzibar Buck-Buck McFate said:
Zanzibar Buck-Buck McFate said:
Watch this video on why underdogs do better in hockey than basketball.
Randomness is inherent to everything that happens (quantum mechanics!) and usually only apparent when zoomed into a small sample of events, but smoothed out on a macro level.
https://twitter.com/espn/status/1320219524210282496
Hockey is more susceptible to the effects of randomness than other sports because
a) there are LOTS of possible interactions at once:
- 10+2 players in a confined space
- puck does not bounce as predictably as a sphere
- boards, stantions, sticks, knobs, legs, skates are all fair game for the puck to bounce off of and into or away from the net
b) goal events are rare, and they're the only thing that matters; like playing heads or tails with a small slice of buttered French bread in the middle of a flock of seagulls, you'll need many, many, many slices of bread just to get a couple of results.
https://twitter.com/NextSportStar/status/935888372572282881
Guilt Trip
Active member
What a beauty by Polak lol.