You should do this as fast as possible-

1. Think of a number between 50 and 100 where the digits are both even yet different, no fractions or decimals allowed

2. After you have your number answer the hidden text below without posting any of what it says.

Is your number 68?

nope

Mother of God.

No, sorry.

nope

no

I remember seeing this on the TV. I chose the same number and so did the guy being filmed. I then tried it on my friend and he also chose it. I believe it is partly because there are only six possible answers and also how the average mind works. However this thread so far shows either it doesn't work when written down, most puzzler's minds work differently or the theory behind it is flawed.

I said 82.

-Doug

Me too.

My first thought was 68, and because it was my first thought, I changed it to 82!

mine was 82

No.

When I couldn't get 42, the one I usually go for, I doubled it for 84.

No it's not.

I thought of 76 (then 86, once I had finished reading the question).

Does the trick have anything to do with subliminal messaging, in the way the question is phrased?

Another No!
By now it is surprising that we haven't seen a single hit, so far.

BINGO! Drunken luck!

I looked at the first number given chose the second and the next followed. Not a puzzler tonight am I?

Nope.

Yes! It was the first one to pop into my head.

yes but it seems like there are a bunch of "no's"

How are there only six possible answers?
I chose 84, by the way.

I count 8... is zero not considered even?

60
62
64
68
80
82
84
86

Carl

I chose 68

Dave

Wow, 68.

I had 82

I think you are right! Nobody else recognized this

Edit: I will blank it out so as not to spoil anyone's fun!

I guess it's ok to post my attempted explanation for it now as others have disregarded Tony's original guidelines- as you read the question first you pick a number between 50 and a 100 but probably change it when you get the next piece of info, the digits must be even - first even digit most people will meet going from 50 to 100 is a 6 (maybe you think 66) but then the next piece of info is that they must be different and as you were going up from 50 to 100 you continue upwards from 6 to 8 giving you 68 - and yes, i thought of 68!

Your explanation sounds reasonable. Still it is the question, why it did not work here most of the time? Because it was written down?

Not everybody that got the "right" answer will be inclined to post about it.

I got the "right" answer hours ago, but didn't bother to post. Only after reading that some are thinking that "it did not work most here most of the time", I decided to log in and post.

Maybe those who didn't get the "right" answer are more inclined to post about the failure of the experiment. Posting of failure vs posting of success could be, in itself, a nice research project.

Also, would the "right" answer have anything to do with normal distribution?

Skarabajo.

I finished reading the directions before thinking about a number. Then I thought "this is one of those everyone uses a similar thought process ones, how can I choose something unlikely? I know, I'll choose backwards and go for 86".

I was a bit surprised to see 68 which is a transposition of my number. If I had tried to choose forward I would have picked 60 or 62.

It's possible that if I hadn't read all of the directions beforehand I might have been coerced into choosing 68.

I guess there might be 8 numbers if we consider 0 to be even. That however is another very long discussion I won't get into and don't have a strong opinion on (yes there are a few).
Although I think this thread shows it doesn't really work I would point out that my instructions say do it as fast as possible. So the two people or so who changed their minds or spent time considering the question weren't really following the rules. If it does have some factual basis I am sure it's only going to work when people are doing it more instinctively.
For those wondering I think it was either Derren Brown or David Blaine (I always get them muddled up) I saw doing this with people in the street. It's only because it worked on me (watching it on TV) and then on my mate that I wondered if there truly was a bias to how we respond to such a question.

There is definitely a bias to how people pick "random" numbers. I'd love to see a distribution of answers to this question. My guess is that 68 will be the most commonly picked and 60 and 80 the least commonly picked.

Basically (I may be wrong) think that what happens is we're told to answer fast, so we read the question fast so we stick to the rules. When told any number between 50 and 100 there's quite a lot to choose from. As soon as we see "both even" our minds jump to the first even number after 5, that being six, so we think 6X. We carry on reading and discover that both digits have to be different. As we are reading and answering quickly, we easily progress forwards and jump from 6 to 8 instead of backwards. This leaves us with 68. I have no idea if what I've said has made sense, or whether or not its right. I'm certainly no physchologist!

This is exactly what I was thinking (although I chose 82).

56 here I wasn't allowed to pick 55, so.. added 1

However both digits need to be even, so 56 is wrong. Admittedly I went with 78 before I fully processed the conditions, after which I went with 68.

Oh like so.. I guess I read too fast again *giggle*

No

Mine was the prime 37.... instantly!




Pantazis

I thought of Johnny Ball.

Before I picked, I knew 68 would be the common number you were looking for, but I wanted to try this on my wife. So I wrote down 68 on a piece of paper. I told her to quickly think of the number. She had no idea why I wanted her to pick, but she said 68, as soon as she said it, I unfolded the paper and showed it to her and she was completely dumbfounded.
I then explained that really there are only a few possibilities and I just got kind of lucky that she used a normal thought process.

-d

That's pretty much what I did with my mate. He is a total sceptic like me and was asking how I did it. I could give him all the twaddle I wanted knowing it was mostly luck.

Speaking of the way people pick numbers, there was a massive dump of ATM / Bank PINs.

Somebody did a fantastic analysis of the data here: http://www.datagenetics.com/blog/september32012/

The bias to 68 is probably somewhat related to how folks pick other "random" numbers.

Wow, what an interesting article. I always wanted to know what the least interesting PIN was. I guess it's 8086 which all of a sudden makes it more interesting.

Back in the day my PIN used to be 0911 (In the states, 911 is the phone number to call in case of emergency). The funny thing where I used to work, there was an ATM. The alarm on it went off one day and the manager knew the way to get it to turn off was to key in the secret code which happened to be 0911. I of course couldn't tell anyone this crazy ironic thing. Since then, I have changed my PIN and I can almost guarantee it's the least interesting PIN since 8086 is no longer the least interesting

But yeah, point is some numbers are more random than others when people or bad generators make them.

-d

Indeed, a great article. In fact I've always wanted to see a similar analysis of what lottery numbers people choose, since all numbers have equal probability of winning, but the advantage would go to the rare numbers that nobody else would choose, because then you don't have to share the prize if you win. This is especially important for mega rollover jackpots, where the jackpot (less tax!) multiplied by the probability of winning is actually greater than the price of a ticket. In this case playing can be considered an investment rather than gambling, because the expected (probability-adjusted) net return is positive, provided you have good reason to believe that you won't share the jackpot with anyone else.

We just had a laugh with this in our house, I wrote 68 on a piece of paper and tried it on my wife too, she picked 68 and was amazed when I unfolded the piece of paper to see it written there, so she went off to try it on our daughter,
and she picked 62, lol

I've always wondered this for the lottery. How could you work out the math to determine if it's worth buying a ticket or not. For simplicity sake, let's say the jackpot is the only prize and is $2,000,000. Say there are 1,000,000 possibilities. Obviously if you played each number once, you would come out $1,000,000 ahead. But if multiple people play and the prize is split amongst all of the winners, how would you know if you are ahead.

My gut is telling me you would need to know the exact number of people playing. But once you have
that, how would it be figured out? Maybe it is somewhat related to the birthday paradox.

Sorry, this is starting to get even more off topic.

-d

I remember reading an article, some years ago now, that said that if the UK lottery ever gave out the numbers 1 2 3 4 5 6, then the jackpot would be shared by around 6000 people.

You want the Poisson distribution. Normally this is about rates of occurrences happening over time but I think you can look at one lotto session and sorta cancel out the time element of it.

Using R it would be ppois(k, lambda)

So if there are 1M possibilities and you buy one of each ticket then you have a 100% chance of winning. Assuming everyone else plays randomly you need 0 other winners. Supposing 1 million other tickets besides yourself were sold then:
> ppois(0, 1000000/1000000)
[1] 0.3678794

So you have a 36.7% chance of being the only winner.

If you break even if you share the jackpot with one other winner:
> ppois(1, 1000000/1000000) - ppois(0, 1000000/1000000)
[1] 0.3678794

And you lose money if there are 2 other winners:
> ppois(2, 1000000/1000000) - ppois(1, 1000000/1000000)
[1] 0.1839397

And in general there is a 26.4% chance you lose money:
> 1 - ppois(1, 1000000/1000000)
[1] 0.2642411


So suppose 5M tickets are sold instead. You have a 0.67% chance of being the only winner:
> ppois(0, 5000000/1000000)
[1] 0.006737947

And 3.3% chance of breaking even:
> ppois(1, 5000000/1000000) - ppois(0, 5000000/1000000)
[1] 0.03368973

On the topic of zero being even.

Zero must be even because if it is not, then it breaks our rules of mathematics.

sooo...

60
62
64
68
80
82
84
86

eight

Anyways, tony, you totally got me. My first guess was 68.