Calgarypuck Advanced Stat Summary

[Rerun]

nm

[Enoch Root]

Couple things, formulate...

1) before I was willing to accept that a player was 'driving possession', I would want to see a track record - a history - of evidence. Having good corsi numbers on one team, or in one specific role, could be due to multiple factors. But if a guy was putting up good corsi numbers for years, in different situations and on different teams, well then he probably actually is driving possession.

2) similarly, the numbers you're using for the Flames are probably too small sample sizes to get very excited about

3) most importantly, I still think your argument puts the cart before the horse. While it's certainly true that a good team is likely to have good possession numbers (yes, there would be exceptions, but good teams usually do), it does not follow that trying to improve your possession numbers will make you a better team. In fact, there is no proof of this at all, and teams like the Oilers, who seem to be attempting to employ that strategy, seem to be failing miserably.

I always come back to the same argument: worry about improving your team - by acquiring better players and filling holes in the lineup - and the possession numbers will take care of themselves.

[formulate]

Cheers.

1) Absolutely, and very well said
2) Again, absolutely. I mostly posted that lineup because I found it funny how close the actual lineup had become. Most of the numbers have very little meaning, but the coincidence was rather strong.
3) You're right about this too, but I don't think that's really what I'm doing here. My point is to maximize chemistry in the lineup (something that is usually rather difficult to quantify). I was using possession stats of multiple players together as a very rough proxy for chemistry. This wasn't so much about improving possession numbers to get better, but to match up players who play well together as an attempt to improve, which would then increase possession numbers, as you've said.

[Enoch Root]

good stuff

[Resolute 14]

Speaking of the Oilers, the growing sample size is causing their Corsi to regress as well. They are now drowning - 49% CF, 22nd in the NHL.

[Enoch Root]

Originally Posted by Resolute 14

Speaking of the Oilers, the growing sample size is causing their Corsi to regress as well. They are now drowning - 49% CF, 22nd in the NHL.

Or maybe under their new coach, they're not playing corsi-ball anymore

(just throwing that out there)

[PeteMoss]

Originally Posted by Enoch Root

Or maybe under their new coach, they're not playing corsi-ball anymore

(just throwing that out there)

When Eakins was fired 50.9% CF.
MacTavish/Nelson combo was 44.6% CF.
Nelson alone has been 46.8% CF.

[Bingo]

Feb 9th Update.

Penguins game one of the worst of the season as you'd expect.





Corsi
1 BUF Dec 11
2 BUF Jan 27
3 NYR Dec 16
4 TOR Dec 09
5 NSH Oct 14

53 CHI Oct 15
52 DAL Dec 19
51 EDM Oct 09
50 VAN Jan 10
49 CHI Dec 14

Fenwick
1 BUF Jan 27
2 EDM Jan 31
3 BUF Dec 11
4 OTT Nov 15
5 MTL Oct 28

53 CHI Oct 15
52 VAN Jan 10
51 PIT Feb 06
50 L.A Dec 29
49 FLA Jan 09

PDO
1 CAR Oct 23
2 ANA Nov 18
3 EDM Oct 09
4 EDM Dec 27
5 ARI Nov 13

53 BUF Dec 11
52 PIT Feb 06
51 ANA Nov 25
50 CAR Nov 10
49 ANA Jan 21

[driveway]

Thanks for doing these snapshots. Are you using total corsi and fenwick, or 5v5 close?

[Imported_Aussie]

Not sure if this is the best thread to place this, but TSN is using "advanced stats" here to beat up on the Flames big time. Same story - Flames are lucky, not good (outside of Giordano + Brodie)
http://www.tsn.ca/flames-trying-to-b...-odds-1.204334

[Phaneuf_Rules]

Being someone who has applied advanced stats to Manufacturing, Construction, and Oil and Gas that article is a joke.

Im going to use that with my boss next time stats dont explain something... must be luck!

[Resolute 14]

Originally Posted by Imported_Aussie

Not sure if this is the best thread to place this, but TSN is using "advanced stats" here to beat up on the Flames big time. Same story - Flames are lucky, not good (outside of Giordano + Brodie)
http://www.tsn.ca/flames-trying-to-b...-odds-1.204334

I was actually about to start a thread on that article. (Bingo or a mod can split If they like, I suppose), but I will comment here instead.

I look at that article as showing the good and the bad of hockey analytics.

First, the good. The Corsi chart shows that what the Flames are doing is unusual, and why people are skeptical. If the post-season began today, the Flames would qualify with the second-worst Corsi against/60 of any playoff team since 2007. We appear to be about 15-17th worst in Corsi for/60. People keep waiting for the other shoe to drop because the other shoe usually drops. It doesn't always, however, and I think Yost does show that with that table. So a point to the good for him.

But the bad is very bad, and highlights why so many people are so skeptical of analytics itself. Travis Yost correctly points out that one of the reasons why we are where we are is that we have one of the best 5-on-5 shooting percentages in hockey at 8.9%. He notes this is significantly better than the 7.8% we shot last year. That is all fair game. But for some reason, Yost declares that a "betting man" should assume we will be closer to that 7.8% over the next 30 games than the 8.9%, but he gives no actual reason for why that should be.

The most reasonable explanation is the kind of circular logic that advanced stat people are extremely prone to falling into: We will fall back because he believes we should fall back.

But that argument opens up some rather important questions. First, why a false dichotomy? When we have gone 54 games at 8.9%, how does one justify that we will fall to the 7.8% over 82 games last year? (Implied: why is a 54-game sample insignificant compared to 82?). If he wants to argue the roster is "basically the same", then why wouldn't he combine the two years and pull a figure for 136 games that should be the "reasonable expectation"?


His comment about overtime is both apt and silly. I agree that OT/SO performance is not a skill a team can reliably reproduce. But pretending that LA and Calgary don't have the records they do to make a point is silly. We do have the records we do, and that is why we are in a playoff spot and LA isn't. Again, abusing the data to try and fit a preconceived belief rather than taking the data for what it is.

The third point is self-evident: Gio and Brodie drive this team. No surprise there.

[Street Pharmacist]

Originally Posted by Imported_Aussie

Not sure if this is the best thread to place this, but TSN is using "advanced stats" here to beat up on the Flames big time. Same story - Flames are lucky, not good (outside of Giordano + Brodie)
http://www.tsn.ca/flames-trying-to-b...-odds-1.204334

I disagree. He didn't "beat up" on the flames at all. He simply suggested that the stats show they likely are lucky. That's not mean or unfair. That's what they do suggest. It simply means the flames have come out on the right side of it more often than not. It's fairly clear that the Flames score on more of their chances than other teams. It's also not true that the Flames limit chances from the outside. They don't.

Whether it's sustainable or not is not a question of whether the person answering the question likes the flames or not. People can analyze what they think without "beating up" on the flames. The trend has been for a few years now that a couple/few teams with bad corsi make the playoffs only to have a more difficult time sustaining it the next year.

Maybe the flames do buck the trend and I'm happily watching them hoping they do. What he's saying is it's not that relevant because this is a bonus. The team has the right people in place and more coming.

Squeezing into the first-round of a competitive Western Conference would be a huge feather in the cap of the entire franchise, and it doesn’t disservice them any as they continue to build towards future long-term contention.

I think that he's correct here. We would be squeezing in and that's fine because we aren't expecting to be a contender. Are we?

[Street Pharmacist]

Originally Posted by Phaneuf_Rules

Being someone who has applied advanced stats to Manufacturing, Construction, and Oil and Gas that article is a joke.

Im going to use that with my boss next time stats dont explain something... must be luck!

Goals are a fairly random event, that's borne out by the facts. Is Gagner an 8pt a game player, or was he maybe "fortunate" to have had so many chances for points?

[Bingo]

The verdict is in!

Lambert says the Flames suck http://flamesnation.ca/2015/2/12/fiv...medium=twitter

[Street Pharmacist]

Originally Posted by Bingo

The verdict is in!

Lambert says the Flames suck http://flamesnation.ca/2015/2/12/fiv...medium=twitter

No, Lambert sucks

Edit:
Not only that, but the people with numbers are "smarter" than flames fans

[Resolute 14]

Yep. People like Lambert, MudCrutch and Drance do the advanced stats cause far more harm than good via their attitudes and unfailing belief that their stats tell the whole story.

[FBI]

yeah more of the same... flames do better when they are underdogs anyway.
boy the media sure has a hard on for corsi these days.

[Flames Fan, Ph.D.]

Originally Posted by Bingo

The verdict is in!

Lambert says the Flames suck http://flamesnation.ca/2015/2/12/fiv...medium=twitter

It's interesting watching people tie themselves up in knots because they can't make sense of reality versus the predictions of metrics that don't correlate to wins. Lambert is working hard to make the data fit a rule that is inaccurate.

Arguably, one of the easier "possession" sports to track is soccer since the ball is not turned over as readily as the puck in hockey. Over the last 5 years, the team with the greatest possession in an English Premier League game has won ~50% of the games.

[Bingo]

its stuff like this that drives me nuts

(Author's note: If you think wins are the only arbiter of whether a team plays well and regularly say fancy stats "don't tell the whole story" to refute them altogether, please stop reading here. Enjoy your day.)

That's a person that wishes to discount opposing views rather than just standing on your two feet and defending your position.

I wonder about sustainability all the time. I look into advanced stats, and then I marry that with my eye test and try and come up with what my blended theory of things are. The Flames have bad advanced stats, and that is troubling. However they aren't getting absolutely killed every night or scoring flukey goals from center ice. Anyone who watches the games knows that.