Not to worry; this won't last says Myrtle

[CaptainCrunch]

Originally Posted by Henry Fool

The most significant statement in the article is that the Flames are "very poor at controlling play." Is that based on possession stats or watching the games, I wonder. There are many ways for teams to "control play".

If defensively you allow a team to run around the perimeter and then they finally get a shot off from a non dangerous area.

the team on defense has just played very good defense, but in terms of possession advanced stats they're taken a hit.

If the Flames who are very much a rush team goes down the ice in 3 seconds and scores on a breakaway and then the opposition comes back and holds the puck on the cycle for 30 seconds without getting a good shot the stats will likely show that the Flames have been dominated.

Moneyball stat tracking is flawed in a fluid sport.

[Nehkara]

There is also an error in this article. The Flames sit 8th overall in the NHL, not 10th.

[Tinordi]

I think one of the reasons there's so much antipathy to "advanced stats" is that they refute long held tropes, biases and stories we tell ourselves about team X or player Y. We don't like that, we like the story of "heart" and a player being "clutch." It creates a causal and hence relatable nature to the game. Watching the game and drawing conclusions is available to anyone who wants to do it whereas those who want and are able to evaluate data is a much smaller segment of the population. That's seen as an attack by many, that somehow there knkowledge of the game isn't as valuable as new knowledge, of course there would be almost a visceral pushback.

But that's the precise value of the stats, they provide a way to 'ground-truth' your biases and assumptions. Are they perfect, no but should that preclude them from being used? Are they any less perfect that Joe fan sitting on the couch complaining about how terrible Phaneuf is?

[Bandwagon In Flames]

Originally Posted by CliffFletcher

Then we're watching different coverage. 90 per cent of the commentary about the Flames this year has been highly positive. The universal praise of the work ethic of the team. The excitement over Gaudreau. All the praise for Giordano.

But one or two commentators make one or two remarks that the team success might not last, and it's time to circle the wagons against all those mean guys in the media who hate the Flames.

Christ we have a thin-skinned fanbase.

And let's be honest guys - if advanced stats analysis earmarked the Flames as one of the better teams in the league, this forum would be nodding in agreement at the predictive quality of advanced stats.

Stay Golden and I are watching different coverage than you I guess. Kypreos, Damien Cox, Glen Healy are three guys with lots of air time that constantly spew crap and seem opposed to Calgary putting in effort.

The nice thing is that Calgary does have supporters now, so somebody else on the panel typically speaks up and backs the Flames. This is a new and refreshing anomaly.

[Street Pharmacist]

Originally Posted by JiriHrdina

The basic problem is not the stats themselves, but how they are used. At times it seems like reporters have a theory and then they go out to find the numbers that support it. It happens in business too. The better approach is look at the stats and decide what narrative they are actually telling you.

Mirtle's niche is to be a stats guy. He tweets stats all day every day. I have very little doubt it was the numbers first that lead to his theory, not the other way around. I'm not convinced the numbers mean write what he thinks, but I highly doubt he had a negative flames view first and went on a search to prove it.

[Henry Fool]

Originally Posted by Gaudfather

A big issue with advanced stats - the "garbage in - garbage out" factor.

Here is a quote from Brad Treliving today: "That’s a challenge in our sport - to get real data. Meaningful data."

We also have to remember that the stats need to be interpreted against everything else we know about a team's or player's current sitution and not all of that can be put in meaningful statistical form.

I think what irritates some people about articles like this is that they only quote a couple of stats as if they were peering beyond appearances to the real truth. You'd have to watch a team carefully and see what those stats mean for that team. Sometimes a save percentage is unsustainable, sometimes it isn't. Sometimes +/- tells you about a player, sometimes it doesn't. That doesn't change when you use more complicated (and more vague) data.

[Nehkara]

Originally Posted by Tinordi

I think one of the reasons there's so much antipathy to "advanced stats" is that they refute long held tropes, biases and stories we tell ourselves about team X or player Y. We don't like that, we like the story of "heart" and a player being "clutch." It creates a causal and hence relatable nature to the game. Watching the game and drawing conclusions is available to anyone who wants to do it whereas those who want and are able to evaluate data is a much smaller segment of the population. That's seen as an attack by many, that somehow there knkowledge of the game isn't as valuable as new knowledge, of course there would be almost a visceral pushback.

But that's the precise value of the stats, they provide a way to 'ground-truth' your biases and assumptions. Are they perfect, no but should that preclude them from being used? Are they any less perfect that Joe fan sitting on the couch complaining about how terrible Phaneuf is?

That's beside the point. Of course advanced stats are indicators, and legitimate ones... but one piece of information taken on its own with no context is useless.

For example, the Flames have difficulty with possession right now but we are missing 5 NHL regular starting forwards in our lineup including our 2nd and 3rd line centers and another player who can play center. If you are missing your regular centers, you lose more faceoffs... which hurts possession.

Yes, the Flames might drop back down to Earth in terms of luck but they should also improve in possession when the veterans return to the lineup... especially at the center position.

[rage2]

I've been tracking our team's trends for the last few years, and I added PDO this year. FWIW, using last 5 average sample points, our PDO has been trending downwards, while our Points per game has been trending upwards. There was a big outlier on Oct 23rd (CGY 5 - CAR 0) that really skewed the numbers.

Doesn't really mean much, other than the fact that trends are agreeing with the article in that it's coming back down to earth, but we are still winning games and earning more points than in Oct where our PDO numbers were truly unsustainable.

[Street Pharmacist]

Originally Posted by CliffFletcher

I'm pretty sure Myrtle watches far more games than you, or 90 per cent of fans, do. A person can watch lots of hockey and analyze stats.

But yes, there are limits to how much hockey a person can watch. So where does that leave us when it comes to hockey commentary?

A) If you want to have any credibility as a league-wide hockey analyst, you need to spend 15 hours a day watching recorded hockey games.

B) There should be not be any such thing as league-wide analysts, since A is impractical. We should only have individual team reporters reporting on hockey. Or better yet - enthusiastic and highly partisan fans!

Yeah. There's ~135 hours of hockey to watch per week. They can't watch it all.

Stats have their flaws for sure, but they eye test is heavily flawed due to preformed opinion bias, outside noise, inability to watch all players all the time and last-event bias.

[Resolute 14]

Originally Posted by Burner

simply a story from Toronto where they cant stomach someone elses success while they lose to Buffalo and continue to suck.

If you think that, you need to read a lot more of James Mirtle's work.

He was the one guy who told the world last year that Toronto wasn't going to make the playoffs. And he did so using the same arguments he is using here.

[Henry Fool]

Originally Posted by CaptainCrunch

If defensively you allow a team to run around the perimeter and then they finally get a shot off from a non dangerous area.

the team on defense has just played very good defense, but in terms of possession advanced stats they're taken a hit.

If the Flames who are very much a rush team goes down the ice in 3 seconds and scores on a breakaway and then the opposition comes back and holds the puck on the cycle for 30 seconds without getting a good shot the stats will likely show that the Flames have been dominated.

Moneyball stat tracking is flawed in a fluid sport.

The Flames are very fast up the ice. The one thing I like the most about them probably. The forwards are relentless, constantly going wide past the defense or threatening to. I don't know how relevant this is statistically but they sure don't spend much time standing around with the puck. It's forward, forward, forward all the time.

[Street Pharmacist]

Originally Posted by GioforPM

The lack of experience at centre = lower faceoff percentage = less initial possession in the d-zone, = more shots against. However, those same players are not bad defensively + Brodano = less quality shots. Really -how many 5 bell saves do you remember the goalies having to make the last few games, especially where the game was on the line? The last on I can recall is Hiller's against the Caps.

Lots of goals aren't scored on 5 bell opportunities. Rebounds, tips, etc are the main offense for most teams.

The reason Gio and Brodie are awesome defensively is because they allow less shots towards the net than the they get. Not because they make them take worse shots

[PeteMoss]

Originally Posted by Tinordi

I think one of the reasons there's so much antipathy to "advanced stats" is that they refute long held tropes, biases and stories we tell ourselves about team X or player Y. We don't like that, we like the story of "heart" and a player being "clutch." It creates a causal and hence relatable nature to the game. Watching the game and drawing conclusions is available to anyone who wants to do it whereas those who want and are able to evaluate data is a much smaller segment of the population. That's seen as an attack by many, that somehow there knkowledge of the game isn't as valuable as new knowledge, of course there would be almost a visceral pushback.

But that's the precise value of the stats, they provide a way to 'ground-truth' your biases and assumptions. Are they perfect, no but should that preclude them from being used? Are they any less perfect that Joe fan sitting on the couch complaining about how terrible Phaneuf is?

I can't speak for everyone - but I don't care about heart or clutch... they are just attributes that get applied to players on winning teams (i.e. Iginla's reputation here when the team won in the mid-2000s versus when they lost at the end of his run here).

These stats are not advanced. Added together save percentage and shooting percentage is not advanced. Tracking shots on net is not advanced. They are just taking what we know already and putting numbers on it.

There are advanced stats out there that make a difference - dump-ins versus carrying the puck in, which players work best together, match-ups, etc.

At this point you are using an 19 game sample and trying to draw conclusions from it.

The Flames have been better of late. Which is the real Flames - the team at the beginning of the year that getting outshot widely or the team now that is playing better? We don't yet know.

[transplant99]

Originally Posted by CliffFletcher

I'm pretty sure Myrtle watches far more games than you, or 90 per cent of fans, do. A person can watch lots of hockey and analyze stats.

But yes, there are limits to how much hockey a person can watch. So where does that leave us when it comes to hockey commentary?

A) If you want to have any credibility as a league-wide hockey analyst, you need to spend 15 hours a day watching recorded hockey games.

B) There should be not be any such thing as league-wide analysts, since A is impractical. We should only have individual team reporters reporting on hockey. Or better yet - enthusiastic and highly partisan fans!

Why on earth would you think this?


Does Myrtle have more eyeballs than most people? More TV's?

I just don't understand why this would have anything to do with...anything.

Especially advanced stats, because unless he is the guy collecting the data and tracking it all, he doesn't need to watch a single second of hockey to come to his conclusions...its all laid out in this stuff called PDO/Corsi...whatever.....right?

[Cali Panthers Fan]

Originally Posted by Tinordi

I think one of the reasons there's so much antipathy to "advanced stats" is that they refute long held tropes, biases and stories we tell ourselves about team X or player Y. We don't like that, we like the story of "heart" and a player being "clutch." It creates a causal and hence relatable nature to the game. Watching the game and drawing conclusions is available to anyone who wants to do it whereas those who want and are able to evaluate data is a much smaller segment of the population. That's seen as an attack by many, that somehow there knkowledge of the game isn't as valuable as new knowledge, of course there would be almost a visceral pushback.

But that's the precise value of the stats, they provide a way to 'ground-truth' your biases and assumptions. Are they perfect, no but should that preclude them from being used? Are they any less perfect that Joe fan sitting on the couch complaining about how terrible Phaneuf is?

I get what you're trying to say, and I agree that statistical analysis is incredibly valuable, but I think all others are saying is that you can go too far and become a robot that is only interested in data. This is still a game played by humans where emotion and psychology play as much of a factor as the actual skill itself. That makes it wildly unpredictable over a short term, and really this team's data from this season is still too small to make assertions of what will definitely happen to them. You could use data from last year as well, but there are different players on this team (Gaudreau, Raymond, Hiller, Bollig, Engelland, Jooris) than last year even, so it's really only 3/4s of last year's team that is the same, making some of that data irrelevant.

I don't have a problem with what Mirtle said, it's just that he's ignored a lot of other data that is still relevant, namely goal differential. Also, when you look at the experience level of our centers, it's not surprising that we lose a lot of face-offs and don't have as much possession as a result. Also, if you were to watch this team, they don't look the same as the Leafs he mentioned, or the same as the Avs of last year which other posters have mentioned. There's a harmonious marriage between analytics on paper and analytics with visuals, and neither should be considered more important than the other.

[Street Pharmacist]

Originally Posted by rage2

I've been tracking our team's trends for the last few years, and I added PDO this year. FWIW, using last 5 average sample points, our PDO has been trending downwards, while our Points per game has been trending upwards. There was a big outlier on Oct 23rd (CGY 5 - CAR 0) that really skewed the numbers.

Doesn't really mean much, other than the fact that trends are agreeing with the article in that it's coming back down to earth, but we are still winning games and earning more points than in Oct where our PDO numbers were truly unsustainable.

This is my argument too. PDO might regress, but that doesn't have to mean losses as the sample size is small.

[Street Pharmacist]

Originally Posted by PeteMoss

The Flames have been better of late. Which is the real Flames - the team at the beginning of the year that getting outshot widely or the team now that is playing better? We don't yet know.

I agree we don't know yet. I'm a big believer that full sample size (from beginning of season to whatever date you're at) is almost always a better predictor of future success than current trends. There are exceptions of course, but not often

[Robbob]

Originally Posted by CaptainCrunch

I think that sometimes Advanced Stats are designed for people that need to feel better about themselves and don't take the time to actually watch a hockey game.

Yeah its great when your talking about possession times and PDO, but I have yet to see someone come up with a quality of shot taken, quality of shots allowed stat, or mean passes before a shot.

If a team is getting outshot like the team up north and the shots are coming from key scoring areas, and teams are getting it off of the forecheck because the defense runs around a lot then you would have an argument.

Even the svpct is invalid because it dosen't take into account things like rebounds allowed for goals or good goals versus bad goals.

But its easy to sit back with a spreadsheet and mutter, they're no good without ever watching a game or ever actually thinking about it.

Stats are not a substitute for the real thing.

with the addition of the new tracking system I hope they can extrapolate on the quality of shots. In the first Oilers game sure they got 40 shots on us but only around 10-15 came from quality areas. The facts are NHL goaltenders will stop pucks all day driving up their save % if they face those shots from the outside.

In a perfect world there would be some sort of stat or metric for shots/saves within a defined area.

[_Q_]

Two main reasons why advanced stats are garbage.

1) Possession stats are measured by shot attempts rather than possession time. So if the coyotes skate up the ice, fire the puck from the blue line, two feet wide of the net and then Brodie and Giordano control the play for most of the shift, looking for a good play and get no shot off, advanced stats say that the coyotes have better possession numbers.

2) PDO says literally nothing about anything and is a poor predictor of how a team will fare. "The Flames have the same PDO numbers as the top teams from last year right now, we know they're not a good team so those PDO numbers are definitely going to go down because they're not the Bruins or Penguins." Well surprise surprise, the Flames are top 5 in the West right now and have top 5 PDO.

Anybody who watches the games can tell you that the Flames are actually very good at controlling the play. No amount of shots off the glass or blocked shots by the opponents is going to prove otherwise.

[Lanny_McDonald]

Originally Posted by Tinordi

Just to be clear. You are the person claiming that the data is unreliable. That's the claim. I haven't made any besides insinuating that criticising stats generally comes with a lack of understanding them. So far you seem to be supporting that. If there is a systemic bias in the data then we should observe issues with it in the analysis. So back up your claim. What are the problems with it? I'm not saying it's above reproach, I just want more than platitudes like some guy named New Era on the internet knows that the data is biased or incomplete.

This is pretty simple stuff, even for the Stats 101 set. If statistics are not gathered the same way, using the same criteria for evaluating an action, and maintaining the same level of integrity in judging events, the data then becomes unreliable. I see all sorts of differences in what is considered a shot in the same game let alone different buildings and different data collectors.

When the league can't train its on and off ice personnel to consistently identify what a particular penalty is, and enforce that with any consistency, how can you honestly sit there and smuggly say that off ice officials are also not using similar subjectivity to collect stats? There is no consistency in data collection. Just closely watch one game and compare the stats you collect versus that presented during the broadcast and in the box score afterwards. A great example was last night's game between Edmonton and Phoenix. The Coyotes spent a good minute in the offensive zone and had several close in shots on goal than were never registered on the shot clock. On the other side of the ice a dump in ended up on net was judged a shot on goal. Sure makes those Corsi stats look good for the Oilers! Also makes the advanced stats look bogus. And that is just one game in one building in the league. The same thing happens all over the league, affecting every stat. That is why regular stats are questionable, but when you start making inferences adding multiple stats together to make another statistical category, well the data errors are compounded and the data becomes useless.

To believe that the statistics are meaningful means you have to believe the data collection is accurate and consistent, which has been a problem for as long as hockey has been around, let alone now where teams are hiring these advanced stats guys to dream up new and improved ways to prove their teams are better than they really are. When guys like Mudcrutch, who have made some of the dumbest claims in the history of the game (and has the stats to back himself up) are hired by NHL teams, well, the system is wide open for abuse. I think last nights game was another example of this abuse in action and is a shining example of why these advanced stats are garbage.