Wild Card Race (Western Conference)

[CorsiHockeyLeague]

Originally Posted by Enoch Root

It is far from absurd. A team higher in the standings has, by definition, performed better. To draw the conclusion from that that they are better is not at all absurd. Premature, as with any other stat, sure. Absurd? No. The fact that you think it is, shows again your overconfidence in your opinion.

As for the Oilers, you might want to take a bit of your own advice and consider sample size. Any team, over a short number of games, can win the vast majority of them. Sample size applies on that front too.

But that's exactly the point! What sample size do you need to show that games won or lost actually demonstrates that a team is better? If I can say that after 82 games, team X has 100 points and is therefore better than team Y with 96, why cant I do the same thing after ten games? If the standard being used by some that standings position is the final answer on what teams are better at hockey, sample size shouldn't even matter. If it does, it NECESSARILY implies that something else is affecting your results that needs time to work its way out of your sample.

[Dan02]

Originally Posted by CorsiHockeyLeague

^That chart does not give you reliability, which the one I posted before does.

No your previous chart ignores the stat both my chart and myself are suggesting is most predictive.

I will offer a challenge to you, run the math on goal differential, if it proves to have less reliability then you will have gone a long way to changing my mind. I'm not someone opposed to changing my opinion when the math proves things out.

But over and over I see fancy stat guys completely ignore it, or shift the goalposts (aka 5v5 GF%)

Just because something is new, doesn't make it better.

[Street Pharmacist]

Originally Posted by CorsiHockeyLeague

Eh, but subjective descriptors can still have at least some truth value to them. If someone tells you they think the Oilers are a better hockey team than the Blackhawks, it won't strike you as being equally true as the inverse.

Yes, but there's so much middle ground. You can say the Kings are better than the flames and be right based on your own criteria, and I can say the reverse. It's a stupid argument because we're not wrong the same thing.

You're saying it's more likely the Kings would win a game/multiple games based on statistics, they're saying the flames are "better" with no qualifier. It's like you arguing the apple is red and they say no the apple is delicious.

This is where those who are opposed to statistics like these without really understanding their role get so upset. You have drance saying crazy things like the oilers are better than the flames in November:

@ThomasDrance: @YYCHockeyfan911 I'm not at all. The Oilers are better, they control games better at evens. They will finish with more points.

That detracts from what usefulness there is because now people don't want to hear about it

[Enoch Root]

Originally Posted by CorsiHockeyLeague

But that's exactly the point! What sample size do you need to show that games won or lost actually demonstrates that a team is better? If I can say that after 82 games, team X has 100 points and is therefore better than team Y with 96, why cant I do the same thing after ten games? If the standard being used by some that standings position is the final answer on what teams are better at hockey, sample size shouldn't even matter. If it does, it NECESSARILY implies that something else is affecting your results that needs time to work its way out of your sample.

You are taking their point to the extreme to try and prove it wrong.

You agree that Corsi isn't a perfect predictor and sample size matters, but when it comes to the standings it has to be absolute or people are wrong to refer to it?

[troutman]

Originally Posted by CorsiHockeyLeague

I don't know if it factors into sportsclubstats's model either.

I assume that it does. When it simulates the rest of the season, remaining schedules are part of the simulation.

It knows the season schedule and scores for past games. As games are played it grabs the new scores from the internet (or gets scores sent in from fans) and simulates the rest of the season by randomly picking scores for each remaining game. The weighted method takes the opponents record and home field advantage into account when randomly picking scores, so the better team is more likely to win. The 50/50 method gives each opponent an equal chance of winning (or tying if the sport allows it) each game. When it’s finished "playing" all the remaining games it applies the league’s tie breaking rules to see where everyone finished. It repeats this random playing out of the season million of times, keeping track of how many "seasons" each team finishes where. Finally it updates the site with the new results for you to read with your morning coffee.

[CorsiHockeyLeague]

I'm not taking the point to the extreme, that is literally the point that several people were making. I.e., I say I think the Kings are better at hockey than the Flames, Jets, and Canucks, and someone says, "you're wrong; look at the standings". That is not the final word.

Wins and losses of course have some value in predicting future wins and losses; they're just not as useful a predictor as possession is. And possession, in turn, is not even close to infallible.

[Enoch Root]

Yes, them saying 'you're wrong' is just as bad as your 'absurd' assertion.

These are opinions, referring to that which is a) always in flux, b) based on different and individual definitions of what constitutes 'better'.

Why people have to try so hard to be right and wrong about them is baffling to me.

[CorsiHockeyLeague]

But that makes no sense. Them saying "you're wrong" is absurd precisely because it's so adamantine and absolute.

Whatever, I don't know why I keep getting into this nonsense.

[Scornfire]

All I know is my eyes have had a better prediction rate than Corsi when it comes to the Flames this year.

[EldrickOnIce]

Originally Posted by CorsiHockeyLeague

Let's take another spin on it, shall we: by your argument, the Vancouver Canucks were the best team in the National Hockey League for two years running in 2011-2013.

Yes, they obviously were (assuming the schedule was balanced enough). Translating that supremacy into winning a short series is not a guarantee, fortunately for anyone not a Canucks fan. Hockey is different than some other sports in that respect. The EPL is an example - where the Champion is crowned by season record, without playoffs.

In hockey, a season is a finite period - a definitive start and end point. That is an absolute. So yes, in the 2014-15 NHL season, IF the Flames finish above the Kings (who miss the playoffs), the Flames are absolutely the better 2014-15 team. To suggest otherwise, to me, shows a lack of understanding as to what a sport actually is.

[CorsiHockeyLeague]

Yeah man I totally don't understand what a sport is. You've nailed it. I'll just go.

Enoch, see Eldrick for example of someone taking absolutist view as to standings position being equivalent to team's hockey playing ability. I'm not straw man-ing here.

[RyZ]

Originally Posted by Scornfire

All I know is my eyes have had a better prediction rate than Corsi when it comes to the Flames this year.

Actually watching the games has nothing to do with and seems to be actually frowned upon in "advanced" stat circles. Watch numbers, not hockey.

[heep223]

Originally Posted by CorsiHockeyLeague

Yeah man I totally don't understand what a sport is. You've nailed it. I'll just go.

Enoch, see Eldrick for example of someone taking absolutist view as to standings position being equivalent to team's hockey playing ability. I'm not straw man-ing here.

Dude, you've got kind of a victim/defensive/passive aggressive thing going on with your posts but you can't just come in with such a condescending and patronizing tone, and not expect to be criticized:

Originally Posted by CorsiHockeyLeague

That's just so misguided. Some people just want to watch the games and enjoy the wins when they come, and these people will not care what analytics say. Other people want to engage in an intellectual exercise to understand what makes teams successful long-term, and these people will care what analytics say. Both are legitimate ways to enjoy hockey.

What do you expect?

There is a saying in my business, that "the market can stay irrational for longer than you can stay solvent". Sometimes, things can't be explained by our meager humanoid attempt to categorize, quantify and predict. Things should be going differently; that stock should be going up; that team should be losing. The reality is that there is a ton of other factors beyond the simplistic models that we create to try and predict and find meaning, including elements that defy rationality.

My point is that accepting the above IMO leads to more humility and ironically a far more objective eye.

[Hockeyguy15]

He sounds way too invested in analytics.

[bomber317]

Originally Posted by Hockeyguy15

He sounds way too invested in analytics.

I think there's value to analytics. It is still evolving.

The issue I have is when the Corsi playoff chart gets posted every nth post showing how low Calgary is with no comments. It is almost a troll job every time that gets posted. And then somebody always asks "what's the methodology behind this?"

How about thoughts on why Calgary is low on this chart? Or why this methology can't explain Calgary still winning and keeping pace in the playoff race despite some of the advance stats?

Perhaps exploring those questions will bring a more engaging conversation on here that may lead to a better application of some of the advance stat data that we have?

[dissentowner]

Originally Posted by bomber317

I think there's value to analytics. It is still evolving.

The issue I have is when the Corsi playoff chart gets posted every nth post showing how low Calgary is with no comments. It is almost a troll job every time that gets posted. And then somebody always asks "what's the methodology behind this?"

How about thoughts on why Calgary is low on this chart? Or why this methology can't explain Calgary still winning and keeping pace in the playoff race despite some of the advance stats?

Perhaps exploring those questions will bring a more engaging conversation on here that may lead to a better application of some of the advance stat data that we have?

Exactly, that chart RGMG every time I see it.

[The Fonz]

+1 vote for changing this thread title to "Playoff Race"

Central/Pacific

2 Vancouver -78 PTS - 17 GR
3 Calgary - 77 PTS - 16 GR
WC#1 Minnesota - 79 PTS - 16 GR
WC#2 Winnipeg - 78 PTS - 16 GR
----------------------------------
9th Los Angeles - 75 PTS - 17 GR
10th San Jose - 72 PTS - 16 GR

Tonight's games: ANA @ VAN, PIT @ SJ

A regulation loss for San Jose tonight, and I'd say we could drop San Jose out of the race. They'd need a 12-3 record thru their remaining 15 games to reach 96 points.

[polak]

Someone besides us and SJ please lose a game for #### sake.

[Resolute 14]

Originally Posted by CorsiHockeyLeague

But that's exactly the point! What sample size do you need to show that games won or lost actually demonstrates that a team is better? If I can say that after 82 games, team X has 100 points and is therefore better than team Y with 96, why cant I do the same thing after ten games? If the standard being used by some that standings position is the final answer on what teams are better at hockey, sample size shouldn't even matter. If it does, it NECESSARILY implies that something else is affecting your results that needs time to work its way out of your sample.

Standings are a reflection of a moment in time. If the Oilers have 14 points after 10 games and the Flames have 12, then yes, at that point in time, the Oilers are a better team. Even today, the fancystats say the Oilers should be better than the Flames, but are just horrendously "unlucky". Could you, with a straight face, say right now that the Oilers are a better team than the Flames?

[White Out 403]

Originally Posted by Resolute 14

Standings are a reflection of a moment in time. If the Oilers have 14 points after 10 games and the Flames have 12, then yes, at that point in time, the Oilers are a better team. Even today, the fancystats say the Oilers should be better than the Flames, but are just horrendously "unlucky". Could you, with a straight face, say right now that the Oilers are a better team than the Flames?

Once again, fancy stats aren't used to comment on previous games but predict likelihood of future games.