GDT: #14 - 11/4/13 | Anaheim Ducks @ New York Rangers | 7:30 PM - MSG, NBCSN

[Killem Dafoe]

I can stare at that gif for hours.

 

[Tawnos]

I know about the gambler's fallacy. It's actually somewhat based on reality, but people don't understand it. If you play the roulette 10 takes, then winning one time in 10 tries is more likely than winning it if you play it only one time, on the first try. However, if you lost 9 times in a row, your chances don't go up on the 10th time (nor go down).

I think the misunderstanding here is that I was talking about odds, rather than trends. 4 game win streaks aren't common in today's NHL.

 

[BarbaraAlphanse]

I can stare at that gif for hours.

Sign me up when someone starts using a cute Candice Swanepoel gif.

 

[Thirty One]

I think the misunderstanding here is that I was talking about odds, rather than trends. 4 game win streaks aren't common in today's NHL.

They are not that rare for a team that's already won three though.

Look at it in a vacuum cleaner.

 

[silverfish]

 

[Tawnos]

They are not that rare for a team that's already won three though.

Look at it in a vacuum cleaner.

I'll look at the statistics later. If there are anywhere near the number of 4 game streaks as 3, I'll be surprised

 

[Thirty One]

I'll look at the statistics later. If there are anywhere near the number of 4 game streaks as 3, I'll be surprised

Define anywhere near. Logic would suggest there would be around half as many.

 

[SunDin]

Callahan back in the lineup:
http://www.nhl.com/ice/blogpost.htm?id=22254&navid=nhl:topheads

 

[SixGoalieSystem]

Define anywhere near. Logic would suggest there would be around half as many.

We do not like to use that word that around here.

 

[eco's bones]

Callahan back in the lineup:
http://www.nhl.com/ice/blogpost.htm?id=22254&navid=nhl:topheads

Just saw it on TSN.

 

[Tawnos]

Define anywhere near. Logic would suggest there would be around half as many.

Half as many would not be anywhere near. Just keeping my outcome expectations low.

 

[Thirty One]

Half as many would not be anywhere near.

lol ok

 

[Greg02]

Half as many would not be anywhere near. Just keeping my outcome expectations low.

The point is that the outcome of this game is not influenced by the outcomes of the previous game. We're *theoretically* (not really; I'll touch on that in a second) just as likely to win this game having won the previous three as we'd be if we had lost one, or won two thousand, or lost two thousand.

The *theoretically* part assumes that the events are linearly independent, however. Which they're not in this case. For instance, because of scheduling, you might be likely to have a stretch of 3 easy games in a row, or maybe even four, but eventually you have to play a good team.

 

[Tawnos]

The point is that the outcome of this game is not influenced by the outcomes of the previous game. We're *theoretically* (not really; I'll touch on that in a second) just as likely to win this game having won the previous three as we'd be if we had lost one, or won two thousand, or lost two thousand.

The *theoretically* part assumes that the events are linearly independent, however. Which they're not in this case. For instance, because of scheduling, you might be likely to have a stretch of 3 easy games in a row, or maybe even four, but eventually you have to play a good team.

You're talking about odds. I'm talking about trends. One is based on theory, which you correctly point out. The other is based on history, which is a better indicator.

The "gamblers fallacy" assumes that reality remains the same at all times. Different opponents means different "reality".

 

[aufheben]

I think you guys might be looking into this backup thing too much.

The Ducks are a very good team, if we lose, there will probably be many reasons besides lack of Hiller. Isn't Andersen a well known, top goaltending prospect? It's not like Scott Clemmensen is coming to town.

 

[Killem Dafoe]

Callahan is like Wolverine. I love this guy.

 

[5 4 Fighting]

The captain ladies and gents!

 

[Black Tank]

Listen up mother****ers, let me just say one thing. Should Fowler score or even get a point I will be the first one, the very first one to post that Mcilrath sucks, only fights, not worth a top 10, shoulda had a Russian, and Fowler is much, much better. Despite my iron resolve, steely good lucks and lightning fast typing skills, I bet I still get beat by a certain notorious poster.

For the record, Fowler sucks balls and makes Del Zitti look like Scott Stevens.

 

[Fataldogg]

If the team play like they did against Buffalo and Canes they will win.

I call the first shootout of the season with Zucc with the winning SO-goal.

They will need to be BETTER than what we did with the Sabres and Hurricanes. You have to realize, the Ducks are in a different hemisphere than the teams we previously faced. And they're a better team than we are.

We will need to bring in the same level of play against the Sabres / Hurricanes just to compete with the Ducks. And if we do that, we have a shot. Lundqvist will still need to have a great game, in addition to that effort, for us to win.

 

[SnowblindNYR]

I think the misunderstanding here is that I was talking about odds, rather than trends. 4 game win streaks aren't common in today's NHL.

Except that doesn't mean anything either. You don't really have less of a chance of winning the 4th game if you won the first 3. There are 8 possible outcomes in three games. Winning all three is just 1 of 8 outcomes. So no 4 game winning streaks may not be common but that has more to do with (probably) that 3 game winning streaks are uncommon as there are 7 out of 8 outcomes that result in no 3 game winning streak. Now that the Rangers already won 3 in a row, in a vacuum they don't have any less of a probability of winning the 4th game.

 

[SnowblindNYR]

You're talking about odds. I'm talking about trends. One is based on theory, which you correctly point out. The other is based on history, which is a better indicator.

The "gamblers fallacy" assumes that reality remains the same at all times. Different opponents means different "reality".

Absolutely true, but you didn't mention anything about different opponents. That's why I keep saying "in a vacuum". No other factors.

 

[Savant]

Hope Miller plays.

AV loves Pyatt more that Torts loves John Moore.

 

[SnowblindNYR]

And if you mean "math" by theory, then yes. It's theory. All factors being equal math>history. Because you saw something in history doesn't mean the math changes. However, there are more factors in the NHL than the roulette table. I can see psychology and stamina coming to play. However, unless you have stats that show that teams after winning 3 lose more on average than they win, then that's just guess work.

 

[Tawnos]

Absolutely true, but you didn't mention anything about different opponents. That's why I keep saying "in a vacuum". No other factors.

My response to talking about things in a vacuum would be: "who cares?"

 

[Pizza]

One of my favorites. Loved Dubi. I'd take him back in a heart beat.