The 2025 Offseason thread.

[Strange Brew]

Originally Posted by PepsiFree

Wrong.

If you tried that last year, your error rate would have been 12.3.

And 10.4 the year before that if AI can be trusted.

I don't see much evidence that his model is anymore accurate than the very basic single data point inputs being suggested by posters on here.

[PepsiFree]

Originally Posted by Strange Brew

And 10.4 the year before that if AI can be trusted.

I don't see much evidence that his model is anymore accurate than the very basic single data point inputs being suggested by posters on here.

His model was 9.9 the year before.

[the-rasta-masta]

Originally Posted by Goriders

At least they are consistent

It's wild that Backlund has been with this team since 2008/09 season in which he made a one game debut.

He's only been in the playoffs 5 times (including the play-in from 2020)...

[Enoch Root]

Originally Posted by PepsiFree

His model was 9.9 the year before.

My model says every team will get 90 points - how did I do last year?

[getoverit]

Now he knows and can appreciate what it feels like to be a flames fan lol

Originally Posted by the-rasta-masta

It's wild that Backlund has been with this team since 2008/09 season in which he made a one game debut.

He's only been in the playoffs 5 times (including the play-in from 2020)...

[PepsiFree]

Originally Posted by Enoch Root

My model says every team will get 90 points - how did I do last year?

You failed because it’s statistically impossible.

[GioforPM]

Originally Posted by PepsiFree

You failed because it’s statistically impossible.

Yeah but his MOE is ok.

[PepsiFree]

Originally Posted by GioforPM

Yeah but his MOE is ok.

13.7! Terrible!

[D as in David]

[Enoch Root]

Originally Posted by PepsiFree

You failed because it’s statistically impossible.

Originally Posted by PepsiFree

13.7! Terrible!

Not impossible.

And the average error is 11.8

[PepsiFree]

Originally Posted by Enoch Root

Not impossible.

And the average error is 11.8

Sorry, astronomically unlikely and realistically impossible, though technically mathematically possible.

1/4000000000000000000000000000000000000000000 or thereabouts.

Hope we live to see that many seasons!

[Enoch Root]

Originally Posted by PepsiFree

Sorry, astronomically unlikely and realistically impossible, though technically mathematically possible.

1/4000000000000000000000000000000000000000000 or thereabouts.

Hope we live to see that many seasons!

42 zeroes to try and not seem wrong... ironic. Or perhaps pre-determined!

[PepsiFree]

Originally Posted by Enoch Root

42 zeroes to try and not seem wrong... ironic. Or perhaps pre-determined!

That’s the actual number.

[The Cobra]

Originally Posted by PepsiFree

You failed because it’s statistically impossible.

Why is it statically impossible?

The average team gets about that many points.

[UKflames]

Well we've all made it to september. There's been verbal fistycuffs a plenty this Summer, there's been sadly very little hockey news to distract us, but we made it.

Training camp, prospect and NHL, is just around the corner. Hang in there folks we're nearly there

[Kipper_3434]

Has there been any leak on what the Flames are offering Wolf to re-sign? I really hope the team is being aggressive trying to lock him up. Been thinking an 8 year 60 mil deal would be very hard for him to turn down. That's a big risk from the Flames for a young goalie, but I don't think we want to see what the number will be if he pulls off a big season. 7.5m aav would put him 10th in goalie salary (negating Carey Price ltir).

[bettercallbettman]

wrong thread

[PepsiFree]

Originally Posted by The Cobra

Why is it statically impossible?

The average team gets about that many points.

As noted, realistically impossible though not technically statistically impossible.

Every team would need the exact same number of wins, losses, and overtime losses against a very specific set of opponents with little to no room for variation.

The most generous way to achieve it, which is the 1/4x10^42 likelihood, requires ignoring a bunch of factors anyway. More realistic scenarios make it far less likely than any NHL season repeating exactly (ie each team getting the exact same number of wins, points, against the same teams, etc.)

Which just illustrates the point further. Even coming up with a realistically impossible season standings that adheres close to the average would not be more accurate than JFresh’s model, so as unimpressive as a 10.4 variation seems, it’s hard to achieve.

[The Cobra]

Originally Posted by PepsiFree

As noted, realistically impossible though not technically statistically impossible.

Every team would need the exact same number of wins, losses, and overtime losses against a very specific set of opponents with little to no room for variation.

The most generous way to achieve it, which is the 1/4x10^42 likelihood, requires ignoring a bunch of factors anyway. More realistic scenarios make it far less likely than any NHL season repeating exactly (ie each team getting the exact same number of wins, points, against the same teams, etc.)

Which just illustrates the point further. Even coming up with a realistically impossible season standings that adheres close to the average would not be more accurate than JFresh’s model, so as unimpressive as a 10.4 variation seems, it’s hard to achieve.

I was simply referring to when you said statically impossible.

[PepsiFree]

Originally Posted by The Cobra

I was simply referring to when you said statically impossible.

And I answered, you’re welcome!