Wild Card Race (Western Conference) - Page 40 - Calgarypuck Forums - The Unofficial Calgary Flames Fan Community

puffnstuff · 03-09-2015, 07:48 AM

So how do the adv stats adjust for strength of schedule? Say two teams that both have good corsi numbers are going to play, how does the prediction account for that? They cant both have the puck (simplification, i know) more than 50% of the time.

Igster · 03-09-2015, 08:03 AM

I know some people like it and consider it "worthwhile", but that corsi chart is ridiculous IMO. Flames have proven it wrong time and time again.

White Out 403 · 03-09-2015, 08:17 AM

Quote:

Originally Posted by Igster

I know some people like it and consider it "worthwhile", but that corsi chart is ridiculous IMO. Flames have proven it wrong time and time again.

Well, its a predictive tool... not a crystal ball. Of course there will be teams that beat the odds.

mrkajz44 · 03-09-2015, 08:19 AM

Quote:

Originally Posted by puffnstuff

So how do the adv stats adjust for strength of schedule? Say two teams that both have good corsi numbers are going to play, how does the prediction account for that? They cant both have the puck (simplification, i know) more than 50% of the time.

I'd imagine that the corsi numbers affect the simulations in predicting who is more likely to win. This would be similar to goal differential. Two teams with a positive goal differential can meet, but only one can end up with a positive goal differential at the end of the game, yet we can use goal differential as a predictor for who is more likely to win the game.

D as in David · 03-09-2015, 08:35 AM

It wouldn't surprise me if that Corsi predictor still had the Flames at less than 50% even once they've clinched a spot.

Itse · 03-09-2015, 08:35 AM

Our odds at sportclubstats are now at 82.9, and they see us winning probably between 8 and 11 games of the last 16.

Hockey.reference.com has our chances at 80%.
Playoffstatus.com has them at 84%.

So essentially they all agree.

(Sportsclubstats also thinks it's more likely we go 16-0 than lose more than 9 games. A pretty strong vote of confidence if you ask me.)

troutman · 03-09-2015, 08:35 AM

Jay Random · 03-09-2015, 08:36 AM

Quote:

Originally Posted by Resurrection

Well, its a predictive tool... not a crystal ball. Of course there will be teams that beat the odds.

But it's not a matter of beating the odds. When you beat the odds repeatedly, an intelligent bookie will change the odds to reflect reality. All the Flames are doing is beating the odds that they would be facing if there were no other elements in the game of hockey except puck possession and luck.

White Out 403 · 03-09-2015, 08:52 AM

Quote:

Originally Posted by Jay Random

But it's not a matter of beating the odds. When you beat the odds repeatedly, an intelligent bookie will change the odds to reflect reality. All the Flames are doing is beating the odds that they would be facing if there were no other elements in the game of hockey except puck possession and luck.

The Flames haven't beat anything yet; its still March. No team in this wild card race or race for 2nd or 3rd in Pacific have earned a playoff spot. This tool is predicting that the Flames have a 40ish % chance of playoffs when all is said and done. That's all.

Jay Random · 03-09-2015, 08:54 AM

Quote:

Originally Posted by Resurrection

The Flames haven't beat anything yet; its still March. No team in this wild card race or race for 2nd or 3rd in Pacific have earned a playoff spot. This tool is predicting that the Flames have a 40ish % chance of playoffs when all is said and done. That's all.

Actually, the Flames have been beating the predictions based on possession stats for over 100 games now. The point is not whether they make the playoffs or not; the point is that the self-appointed oddsmakers are not doing a very credible job of calculating the odds.

White Out 403 · 03-09-2015, 08:58 AM

Quote:

Originally Posted by Jay Random

Actually, the Flames have been beating the predictions based on possession stats for over 100 games now. The point is not whether they make the playoffs or not; the point is that the self-appointed oddsmakers are not doing a very credible job of calculating the odds.

I'm not talking about a game to game corsi prediction. I'm talking end of season where are the teams now type of discussion. I really dont know if were just debating 2 different things here, or, your being purposefully difficult.

Corsi is a strong predictor of where teams future success will go, but its not perfect.

This table is predicting Flames have x% chance of making the post season based on corsi.. but it's not a scroll written from god. Why this is getting some people riled up I get it, its your team, but it's just one predictor for gods sakes.

Itse · 03-09-2015, 08:59 AM

Quote:

Originally Posted by Jay Random

But it's not a matter of beating the odds. When you beat the odds repeatedly, an intelligent bookie will change the odds to reflect reality. All the Flames are doing is beating the odds that they would be facing if there were no other elements in the game of hockey except puck possession and luck.

It's just one site that has completely different odds than all the more popular sites. I recommend putting CorsiHockeyLeague on ignore, you'll miss out on nothing.

Jay Random · 03-09-2015, 09:17 AM

Quote:

Originally Posted by Resurrection

I'm not talking about a game to game corsi prediction. I'm talking end of season where are the teams now type of discussion. I really dont know if were just debating 2 different things here, or, your being purposefully difficult.

It does appear that we are debating two different things.

The thing is, the same predictor that says the Flames have a 60% chance of missing the playoffs also predicted that they had virtually zero chance of being in a playoff position at this point of the season. In this case (and some others I could mention), the predictions have been out of whack with reality for months. Some people still insist that the predictions are correct, and reality itself is out of whack. That's the very opposite of scientific, and that's what bothers me.

heep223 · 03-09-2015, 09:19 AM

Quote:

Originally Posted by Resurrection

The Flames haven't beat anything yet; its still March. No team in this wild card race or race for 2nd or 3rd in Pacific have earned a playoff spot. This tool is predicting that the Flames have a 40ish % chance of playoffs when all is said and done. That's all.

You can't look at this season and say that the Flames are beating the odds?

Think of it this way, odds simply follow where the money goes. At the beginning of the year, a LOT of bettors were betting against the Flames. So odds were very long. Now, far less bettors are betting against the Flames. So the odds are shorter. They have beaten the odds and the odds have adjusted.

Just taking a quick look at Bodog and SportsInteraction, the Flames have better odds of winning the cup than the Jets. I couldn't find a prop line on odds of making the playoffs but it's pretty safe to assume it'd be the same story.

Quote:

Originally Posted by Jay Random

Some people still insist that the predictions are correct, and reality itself is out of whack. That's the very opposite of scientific, and that's what bothers me.

Couldn't have put it better myself. I see this all the time in my industry (investment management). People often insist that it's reality that is "wrong" based on their models, until it persists for so long that they have to adjust their models to the new reality. Reality is reality. Bad teams are bad until they're not.

Street Pharmacist · 03-09-2015, 09:21 AM

Quote:

Originally Posted by Jay Random

It does appear that we are debating two different things.

The thing is, the same predictor that says the Flames have a 60% chance of missing the playoffs also predicted that they had virtually zero chance of being in a playoff position at this point of the season. In this case (and some others I could mention), the predictions have been out of whack with reality for months. Some people still insist that the predictions are correct, and reality itself is out of whack. That's the very opposite of scientific, and that's what bothers me.

It never would've predicted a zero chance to be here. A 60% chance prediction will be wrong 4 out of 10 times
That doesn't make it a bad prediction.

I don't like that chart because the methodology is random. Heavily weighting a rolling 25 game Fenwick over other stats at this point in the year seems arbitrary to me

CorsiHockeyLeague · 03-09-2015, 09:23 AM

This discussion is pretty similar to the one that came up earlier in the thread when that graph was posted, so no point going over old ground. All I'll say is it's based on a methodology with a record of making correct predictions about future team success. That doesn't mean it's always right and it doesn't mean the Flames won't make the playoffs in spite of the methodology. It's based on this, subject to a bunch of adjustments made by Micah McCurdy.

http://www.silversevensens.com/2014/...tions-with-pip

troutman · 03-09-2015, 09:24 AM

Seems to me Flames are far more likely than 44% to get to 93-94 points. They might not get there, but the odds have to be higher than that. Would be a surprise now if they didn't.

CorsiHockeyLeague · 03-09-2015, 09:27 AM

^By the chart's methodology the Flames are projected to end up at 93 points, while the playoff threshold is 95-96 (though this was calculated before last night).

Jay Random · 03-09-2015, 09:29 AM

It does not help that the charts you are posting do not actually predict how many points each team is projected to end up with.

CorsiHockeyLeague · 03-09-2015, 09:30 AM

Quote:

Originally Posted by Jay Random

It does not help that the charts you are posting do not actually predict how many points each team is projected to end up with.

Here you go - this hasn't been updated since yesterday, though.