Cheating

[spankthecrumpet]

What's the confidence level for a Million Dollar Challenge? If, for instance, it were 95% that the results did not arise through chance, it shouldn't be that hard to devise an example (eg about predicting number distribution) that had a 5% chance of appearing psychic.... and on average you're only going to need 10 of these silly attempts getting through stage 1 before the million dollar prize is won. What's to stop this being attempted?

Cheers,

Dave

 

[geni]

Odds for the prelim 1000 to 1

Odds for the million 1000000 to 1

 

[spankthecrumpet]

Well, there goes my cunning plan...

But isn't this a very high standard of proof being required? Or is that just the "extraordinary claims require extraordinary evidence" being taken to its logical conclusion? 99.9% is an extraordinarily high degree of confidence to require in a test in my opinion.

 

[geni]

Well, there goes my cunning plan...

But isn't this a very high standard of proof being required?

No. It is theoreticaly posible to raise any real effect to that level of significance. Predicting 20 coin tosses in a row correctly or eight six sided dice roles in a row would get you up to odds of a million to one

Or is that just the "extraordinary claims require extraordinary evidence" being taken to its logical conclusion?

Partly. And partly it is to make sure the chance of the test being passed by chance is very small.

99.9% is an extraordinarily high degree of confidence to require in a test in my opinion.

I depends what you are doing.

 

[spankthecrumpet]

No. It is theoreticaly posible to raise any real effect to that level of significance.

Sure, but for some claims this is impractical; for instance the claim that someone could predict coin tosses with 50.01% accuracy - if this were indeed the case then it is surely still an achievement of sorts. I agree these sorts of claims are silly:- I'm just surprised no one has tried them yet. If 99.9% certainty is required for the inital test, it makes a lot more sense (although, it would surely make the application much more long winded for any claim such as "I believe that objects are attacted to each by a force that is proportional to their mass and inversely so to their distance")

 

[geni]

Sure, but for some claims this is impractical; for instance the claim that someone could predict coin tosses with 50.01% accuracy - if this were indeed the case then it is surely still an achievement of sorts.

Doable. Just look at the rate vagas lets people gamble

I agree these sorts of claims are silly:- I'm just surprised no one has tried them yet.

No one makes such claims for real.

If 99.9% certainty is required for the inital test, it makes a lot more sense (although, it would surely make the application much more long winded for any claim such as "I believe that objects are attacted to each by a force that is proportional to their mass and inversely so to their distance")

Failes to predict the precession of Mercury's orbit correctly. Claim rejected.

 

[spankthecrumpet]

Doable. Just look at the rate vagas lets people gamble

I don't think it's pracitcally doable by one person in a reasonable time frame, although I obviously haven't gone over the maths here - mind you, I think we're both in agreement here, and the information has helped a fair bit; I suppose I'll just have to invest in magic beans instead.

Failes to predict the precession of Mercury's orbit correctly. Claim rejected.

 

[ChristineR]

I have to agree, 50.01% (on a 50/50 coin toss) would probably be rejected by the JREF as an impractical test. But it could still be done. These sorts of tests are done by scientists all the time. There are experiments of years durations. You'd just have to run a lot of trials. It would be worth it if it proved the existence of telepathy, and not getting Randi's million wouldn't matter much in the long run.

I think most casinos would take .5001 odds, if the game were popular. Vegas considers a "serious" blackjack players to be one that makes bets of a minimum of $50 for at least four hours a day. That works out to be $12,000 a day, of which the house gets $12. Blackjack is complicated by the fact that only the most skillfull players will reach these odds, and hundreds of less skillful players will lose much more.

 

[Admiral]

No one makes such claims for real.

Not true- what about Donald Young?

http://www.internationalskeptics.com/forums/showthread.php?t=49003

Also, consider that predicting the result of a coin flip to a chance of 50.01% isn't paranormal if you know which side the coin starts with, as Stanford professor Persi Diaconis showed:

http://news-service.stanford.edu/news/2004/june9/diaconis-69.html

I read about him somewhere else in this forum- don't remember where.

In fact, this fascinating guy has trained himself to flip heads ten out of ten times. It's a bit of a digression, but I find it interesting.

 

[ChristineR]

I've mentioned this a few times...I once met a guy who routinely got ten of ten flips. He used a weird looking flip, but there's no denying he did it.

 

[rjh01]

If you ever need to decide anything based on a coin flip and the other person tosses the coin make sure that the coin ends up on the floor. It is an easy trick to catch the coin and put it right side up on the back of the hand. It does not matter which way the coin started.

 

[Spindrift]

I've mentioned this a few times...I once met a guy who routinely got ten of ten flips. He used a weird looking flip, but there's no denying he did it.

That proves an ability to flip coins so it comes down the way one desires, not the ability to predict which way it comes down when anyone else flips the coin.