I thought Nate Silver was supposed to be smart. It appears he doesn't understand the most basic aspects of probability

Not really.. wrote:

carmine9 wrote:

If there is a 50% chance of a coin landing in heads? What is he chance the 1st 2 flips will result in heads?

Coin flips are independent events. The house and senate races are not.

carmine9 wrote:

Not really.. wrote:

Coin flips are independent events. The house and senate races are not.

Of course they are.

prognog wrote:

carmine9 wrote:

Of course they are.

You’re advertising your ignorance.

carmine9 wrote:

Not really.. wrote:

Coin flips are independent events. The house and senate races are not.

Of course they are.

carmine9 wrote:

prognog wrote:

You’re advertising your ignorance.

Feel free to explain.

With the GOP at 82% to win the House and 46% to win the Senate, we need to odds of the GOP winning the Senate and the Dems winning the House to be 0% for his numbers to work.

Please show how the math gets us to 0%.

prognog wrote:

carmine9 wrote:

Feel free to explain.

With the GOP at 82% to win the House and 46% to win the Senate, we need to odds of the GOP winning the Senate and the Dems winning the House to be 0% for his numbers to work.

Please show how the math gets us to 0%.

First admit that they aren’t independent events like a coin flip as you tried to claim. If you can do that we can proceed.

indep wrote:

carmine9 wrote:

Of course they are.

Just to be clear, what you're saying is that whether someone votes republican for their house member can't be predicted *at all* by whether they've voted republican for their senate candidate.

If you want, I can write it out. It's a single application of the chain rule in probability and then thinking about it for a ~10 seconds. You can find the math in the first chapter of almost any statistics book.

carmine9 wrote:

prognog wrote:

First admit that they aren’t independent events like a coin flip as you tried to claim. If you can do that we can proceed.

Of course they are.

Silver assigned an 82% probability to one event and a 46% to the other.

Period.

randomist wrote:

carmine9 wrote:

If there is a 46% chance the GOP wins the Senate and an 82% chance it wins the House; the odds it wins both is .46 x .82 = 37.72%. Not 0%. Simple stuff.

Rojo is right. Silver is a moron.

If there is a 50% chance of a coin landing in heads? What is he chance the 1st 2 flips will result in heads?

I know, folks around here are so co-dependent whiny. Here’s another example: 51% chance that the Dow jones index goes up tomorrow, and 11% chance that none of the 30 stocks in the DJI goes up tomorrow, so the chance that DJI goes up tomorrow and none of those 30 stocks goes up tomorrow is .51x.11 = .0561.

Say with me: we want independence!

prognog wrote:

carmine9 wrote:

Of course they are.

Silver assigned an 82% probability to one event and a 46% to the other.

Period.

We can’t proceed if you can’t get this very basic concept. It explains so much about your world views.

indep wrote:

carmine9 wrote:

Of course they are.

Just to be clear, what you're saying is that whether someone votes republican for their house member can't be predicted *at all* by whether they've voted republican for their senate candidate.

If you want, I can write it out. It's a single application of the chain rule in probability and then thinking about it for a ~10 seconds. You can find the math in the first chapter of almost any statistics book.

carmine9 wrote:

prognog wrote:

We can’t proceed if you can’t get this very basic concept. It explains so much about your world views.

God help us

carmine9 wrote:

Last try.

An appeal to common sense.

If there is an 18% chance the Dems win the House and a 46% chance the GOP wins the senate, does it make any sense at all that the possibility of both happening is 0%?

knox harrington wrote:

50% of Americans are female. 10% of Americans have a beard. Therefore, 0.5*0.1 = 5% of Americans are bearded females. Not zero %. Simple stats.