Not to worry; this won't last says Myrtle

[bubbsy]

I love it. The flames are in year 2 of the rebuild and have come out and had an unbelievably great start to the season. Can they keep it up and be in the playoff picture next month, into the new year, or come spring? Who the hell knows, expectations were low again this year, just hoped for effort. Flames fans we can sit back and enjoy the ride.

the author compares the situation to the leafs, but the big difference was that the leafs were a cap team intending to compete and push to be an eastern contender, whereas the flames were supposed to be a bottom 5 team.

[Bingo]

Actually interesting to look into the last season's final third coupled with this year's first quarter

That's 51 games and a record of 30-19-2 for a win percentage of .608, this year they are 11-6-2 for a win percentage of .631

The PDO stats are the discussed 1.028 for this year, but if you take the 51 games together for a larger sample size its 1.019, so they've actually been carrying a pretty lofty PDO for 51/82nds or 2/3 of a hockey season.

The Flames have only been out shot in 24 of those 51 games.

[edslunch]

Originally Posted by saillias

I don't understand why PDO exists. Possession I get, shot attempts I get, those are quantifiable things. But PDO seems to me, someone got stoned and thought it would be brilliant to add a team's save percentage and shooting percentage.

It exists because it combines two meaningful measures (shooting and save percentage) into an easy to understand benchmark. The league average PDO is 100 by definition (since every non-blocked shot is either a goal or is saved), so numbers above 100 indicate that you have better than average shooting and/or saving and vice versa.

It seems like a useful indicator, but then we're told that saves and goals are both based on luck and everything reverts to the mean (100) so I guess it ends up being kinda meaningless.

[EldrickOnIce]

Originally Posted by Resolute 14

The thing about PDO is that it tends to closely match the NHL standings. Most teams end up + or - three spots difference between their PDO rank and their overall rank. The significant exceptions last year were the Leafs (6th in PDO, 23rd overall), Washington (8th PDO, 17th overall), San Jose (16th PDO, 5th overall)

Exactly.
PDO will closely mirror standings, in hindsight.
Possession numbers will closely mirror standings, in hindsight.
Are these predictive? No - but one can assume they will be, on average, by years end. So the Flames will either improve in those underlying stas (as they have) or they will not sustain anywhere close to current winning percentage.
This is no slight on the Flames. No bias. It's simply the truth.

Also interesting that these much maligned advanced stats (around here) are the same ones that have Giordano and Brodie as the leagues top defensive pairing. One can't play the game both ways

[Lanny_McDonald]

Originally Posted by Tinordi

You know I'm trying to be nice here but the effect you're describing is essentially meaningless. People have looked into the 'rink effect' and found that it may impact team corsi by a hundredth of a percent over a season.

This gets back to my original point that you have alot of criticisms of 'advanced stats' without a commensurate understanding.

Commensurate understanding of statistics? I was polite enough to explain to you why these stats are crap, and identified the very reasons they are junk, but the best you have is to come back and insult someone by telling them they don't have commensurate understanding of the subject, without any explanation of your own. That takes some real brass ones. I suspect you don't have the comprehensive knowledge of statistics you like to advertise and the claim of people not understanding them is nothing more than your defense mechanism to prevent that disclosure. If this isn't accurate, please prove otherwise by giving me that 600 level class in advanced statistics I must have missed in grad school. Save the obnoxious response and finally try to explain the validity of the work you place so much faith in.

[calumniate]

Just wait until Paul Byron becomes money on the breakaway. He'll annihilate this PDO stat

[waner]

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 would say the biggest antipathy towards advanced stats from the average hockey fan is that the people "promoting it" to such a piss poor job of it.

Too often it seems there is an your either with it or against it attitude from advanced stats folks and if you don't agree with them it is because you are an idiot, stuck in old ways or too stupid to understand the stats. There seems to be little desire or ability to acknowledge that there can be flaws in the system. Too often I see advanced stats guy fall back on the fact that the "stats don't lie, your eyes" do BS argument.

That is also the other thing I find many stats guys also act as though they are the be all and end all and they do a better job than any human could. I doubt stats guys are right in this regard but whether they are or aren't nobody likes to be told this.

I think stats have a role as a secondary way to judge what is going on on the ice. They will always be behind watching the actual games but used in conjunction with observations they can help to point out some biases that may exist in certain people's assessments when watching and point out an area that is being missed. I think if Advanced Stats (that is such a garbage and misleading name as well) advocates did a better job of explaining how they could be useful in working with traditional views that they would be much better received than the current approach of telling people they are too dumb to understand it if they don't agree that Corsi + is the greatest thing since sliced bread.

[_Q_]

Originally Posted by Bingo

Actually interesting to look into the last season's final third coupled with this year's first quarter

That's 51 games and a record of 30-19-2 for a win percentage of .608, this year they are 11-6-2 for a win percentage of .631

The PDO stats are the discussed 1.028 for this year, but if you take the 51 games together for a larger sample size its 1.019, so they've actually been carrying a pretty lofty PDO for 51/82nds or 2/3 of a hockey season.

The Flames have only been out shot in 24 of those 51 games.

So what you're saying then is the crash will be extra rough when it happens because the Flames have been playing above their head for 2/3rd of a season now.

Get ready boys and girls, the regression to the mean is gonna hurt and it's gonna hurt real bad. The advanced stats have spoken.

[EldrickOnIce]

Originally Posted by edslunch

It exists because it combines two meaningful measures (shooting and save percentage) into an easy to understand benchmark. The league average PDO is 100 by definition (since every non-blocked shot is either a goal or is saved), so numbers above 100 indicate that you have better than average shooting and/or saving and vice versa.

It seems like a useful indicator, but then we're told that saves and goals are both based on luck and everything reverts to the mean (100) so I guess it ends up being kinda meaningless.

Meaningless... but on average closely mirrors standings.
Putting these together aren't new. The Junior A team I worked with combined those and combined special team stats the same way. It wasn't called anything fancy though.

They wanted over 100 combined shooting/save % and over 100 PK plus PP %

[MissTeeks]

I think our GM said it best today:

“@CraigCustance: Treliving on meshing old-school and new-school: "People are ignorant on both sides."”

[Flame19,289]

I don't think that any of us really expected the Flames to be in this position with 25% of the season in the books.

But since "The Brawl" in January, the Flames are 30-20-2.
That translates to a 98 point pace over an 82-game season.

Maybe there's something more to this team than advanced stats are capable of measuring.

[Jay Random]

Originally Posted by Tinordi

The finding was that yes while some arenas over or under count shots, it was an additional 0.5 shots for every 100 shots taken by the home team. In other words insignificant.

Thanks.

Actually, on reflection I don't have a problem with the way the analysis was done (as reported), even though it does essentially amount to second-order statistical analysis and not a controlled study of data collection. Most of the hinkiness I have noticed over the years in SOG counting came in situations where there were repeated rebounds and heavy traffic in the crease; which usually means on the power play.

(Did player X get three whacks at the puck, or only two? Was that second whack a SOG, or did someone else's stick stop it before it reached the goalie? Sometimes hard to tell through so many bodies; and some scorekeepers, as I used to notice, seemed to be tempted to fudge the counting.)

Since PDO is an even-strength stat, these effects would tend to be much smaller.

By the way, it would be interesting to see what PDO would make of a team with a coach like Bob Johnson. Johnson's strategy, as he quite openly proclaimed, was to play .500 hockey at even strength and win games on special teams. (He was the first coach in the NHL, I have read, to make every player on the roster play either PP or PK in practice.) The strength or weakness of a Johnson-style team would tend to lie in an area that PDO doesn't even measure.

[Bandwagon In Flames]

Originally Posted by Flame19,289

I don't think that any of us really expected the Flames to be in this position with 25% of the season in the books.

But since "The Brawl" in January, the Flames are 30-20-2.
That translates to a 98 point pace over an 82-game season.

Maybe there's something more to this team than advanced stats are capable of measuring.

Heartley

[Resolute 14]

Originally Posted by Jay Random

This is why I really don't understand those people who claim that PDO is largely an indicator of luck. The people who are saying the Flames are due for a fall based on PDO are basically saying, 'This team isn't good enough to have a PDO that high, and the proof of it is how high their PDO is.' This is circular reasoning with epicycles and eccentrics. It begins with the assumption that the PDO should regress to 100 in all cases; whereas the fact is that it regresses to a value indicative of the team's overall quality.

I don't think people who know PDO say it is "largely an indicator of luck", but rather that teams whose PDO deviates far from 100 can be said to be lucky or unlucky. And as I noted in an earlier reply the bias of the commentator will affect that 'luck' statement.

And not unfairly, I think. Lets look at the Flames. We were 26th in the NHL last year in PDO with a 99.0. Finished 27th in the standings. This year we are third in PDO at 102.7 and 8th in the standings. Key differences: we lost Mike Cammalleri and gained Jonas Hiller. Our goaltending at even strength has actually fallen to around league average (14th), but our shooting percentage is ungodly high - in Pittsburgh and Tampa territory. And lets face it, guys like Bouma, Hudler and Jooris are less likely to maintain ridiculous shooting percentages than guys like Crosby and Malkin and Stamkos are.

So when looking at teams likely to be a little 'lucky', it's hard to complain when Calgary is included in that list.

Moreover, there seems to be an assumption that the one team you want to criticize is the only one whose PDO will regress. Just from eyeballing the data offered up at various points in the season (and from a little common-sense knowledge of basic statistics), it would seem that the standard deviation of PDO over a small number of games, league-wide, will naturally be larger than the standard deviation over a whole season. That being so, everyone's PDO can be expected to regress somewhat towards the league mean as the number of games increases. But that can happen even if there is no change at all in the relative rankings.

What I mean: A team could have a PDO of 103 at this point of the season, let us say, and be 8th in the league by that measure. By the end of the year, the team could have a PDO of 101.5 and still be ranked 8th, because the grouping is that much tighter due to the larger sample size. To say that the team can't sustain a PDO of 103 is quite correct; to infer from this that the team doesn't deserve to be 8th, and is doomed to fall down the standings, is unjustified.

Agreed. However, the regression to the mean works both ways. So while everything tightens up, that means Calgary is likely to go down, but Chicago, Vancouver, Colorado, Minnesota and Dallas are likely to come up. That will squeeze the Flames in the overall standings.

Taking PDO exclusively, the Flames are actually probably a little unlucky to be 8th as we have the third best PDO. But when one looks at how high some of the shooting percentages are, and how those are likely to regress, and how we aren't controlling the puck as much as we need to be (the last two games notwithstanding) and the fact that we are dead last in offensive zone faceoffs but face the third most defensive zone faceoffs, and the combination of factors makes an argument: If we do not improve in many of these areas, chances are we fall farther than some other teams.

And I find it hard to disagree with that. If nothing changes, the stats predict we tumble down a little. Fortunately, much can change. Even something as simple as getting a guy like Backlund back would be a major boon.

[CliffFletcher]

Originally Posted by Flame19,289

Maybe there's something more to this team than advanced stats are capable of measuring.

Of course there is. Who claimed advanced stats measure and predict everything?

[Henry Fool]

I've never encountered a hockey fan who was against stats. However too often the new stats are presented as a final answer rather than a starting place for discussion. I imagine this is frustrating for many stat enthusiasts as well.

It's good to judge for yourself whether some stat actually adds to our understanding of a certain situation. Some of them are useless, some of them are interesting. Often it's not at all clear whether something is relevant information or not.

[Jay Random]

Originally Posted by Resolute 14

Agreed. However, the regression to the mean works both ways. So while everything tightens up, that means Calgary is likely to go down, but Chicago, Vancouver, Colorado, Minnesota and Dallas are likely to come up. That will squeeze the Flames in the overall standings.

Not necessarily. Winning percentage is another stat where the standard deviation tends to decrease as sample size increases. What matters is not the absolute numbers of PDO, but the relative rankings.

What we really ought to be watching, it seems to me, is the z-score of PDO rather than the number itself. That would adjust for sample size and make the numbers between, say, a 19-game stretch and a full season strictly comparable.

Taking PDO exclusively, the Flames are actually probably a little unlucky to be 8th as we have the third best PDO. But when one looks at how high some of the shooting percentages are, and how those are likely to regress, and how we aren't controlling the puck as much as we need to be (the last two games notwithstanding) and the fact that we are dead last in offensive zone faceoffs but face the third most defensive zone faceoffs, and the combination of factors makes an argument: If we do not improve in many of these areas, chances are we fall farther than some other teams.

Very probably true. Though there is still the problem of measuring proxies instead of actual events. I mean, for instance, that the number of DZ faceoffs is not necessarily an accurate indicator of the percentage of time spent in the DZ. There are situations in which a team may be coached deliberately to try to get stoppages in play while hemmed in their own zone; in which case the coach's tactical decisions will contaminate the data. It's a pity that we don't have actual direct measurements of time of possession and time in zone. No doubt that will come with the new tracking technology. Having RFID chips on every player (and ideally the puck too, and just forget about FoxTrax) will make it possible to collect much more direct data instead of mucking about with all these proxies.

And I find it hard to disagree with that. If nothing changes, the stats predict we tumble down a little. Fortunately, much can change. Even something as simple as getting a guy like Backlund back would be a major boon.

This is an important point. 'If nothing changes' is hardly ever a valid assumption in sports (or in life generally). The Flames have been lucky to do as well as they have with all these injuries; but they weren't lucky to have all the injuries.

[Phil Russell]

Originally Posted by Street Pharmacist

The sample size is large enough. War-on-ice.com has over 10 years of data. I would suggest going to that site and looking at different stats and comparing them to what happened. You can even make larger than one year sample sizes. You can then see for yourself if the stat matched what happened. We can't measure future because it hasn't happened.


One very important thing that has only barely been looked at statistically, is advanced stats and playoff success. So far, the only work I've seen was from fivethirtyeight.com showing that playoff success was most strongly correlated with goalie save %. The kicker? Save % for a goalie is only very rarely repeatable suggesting a hot goalie is the most important thing yet almost impossible to predict or procure.

I guess I am asking if anyone has analyzed the predictive power of advanced stats relative to team performance. I am not capable of doing that empirical analysis myself but I would be capable of digesting the results if anyone knows if this has been looked at. Have advanced stats themselves been statistically tested?

[GioforPM]

Originally Posted by Henry Fool

I've never encountered a hockey fan who was against stats. However too often the new stats are presented as a final answer rather than a starting place for discussion. I imagine this is frustrating for many stat enthusiasts as well.

It's good to judge for yourself whether some stat actually adds to our understanding of a certain situation. Some of them are useless, some of them are interesting. Often it's not at all clear whether something is relevant information or not.

This. They are useful for a team that says "Hey - we are losing. What can we improve on?" They are not useful to say "Hey - we are losing. How can we say we are doing well anyway".

[Five-hole]

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?

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.

I think these two comments are really interesting. Tinordi, I get what you're saying, and hockey certainly does have an old school mentality at times (most times). But simply because there is a push back doesn't automatically mean that new knowledge is good knowledge because it threatens the entrenched status quo. It hasn't really proven itself to be good knowledge. There are enough people trying to carve out a piece of the NHL pie that if there was a system that worked, people would be all over it.

The problem I have with advanced stats is basically what Jiri outlines. They're used to explain a narrative that is already in place. What I'd like to see is a blind analysis based on these stats. Take all 30 teams, assign them a blind ID without seeing their team name or their overall record, and try to predict where they would be in the standings based on these so-called stats. It's easy to take a team that is obviously exceeding expectations like the Flames and point out a bunch of stats that should have them lower in the standings. But do that league-wide when you don't have the benefit of history and the standings and show me where the Flames should be.

Maybe it shows the Flames should be 15th to 23rd. Maybe it doesn't. But picking out one team and saying "oh the PDO is way off their winning %" without analyzing the rest of the teams and just cherry picking one stat makes the analysis basically meaningless.