Guess the number i bet you lose

[antonio glez]

here you have 1.000.001 pseudo random secuence of digits:

wait for it to load fully:

http://www.eveandersson.com/pi/digits/1000000

the game is put a secunce of four numbers trying to guess it right like for example 3498 and count how many times the page searcher finds it for this case it would find it 108 times so i would have won 8%

odds say you should get right one hundred in a million of numbers:

probaility to get right a for digit decimal number:

1/(10*10*10*10)=1/10.000 and you try a million times so by probabilty you should get right 100

so if you get an avearge 110 in your four digit number search youve won an average of 10% but if you find an avearge of ninety% in your searches you lost 10%

please tell how do you do i think most will lose
IMPORTANT: TAKING INTO ACCOUNT THAT WHEN THE LINE BREAKS 3/50 SECUENCES ARE NOT COUNTED YOU HAVE TO ADD 6 TO EVERY RESULT YOU OBTAIN TO PLAY FAIRLY

 

[WhoWee]

I'll take your word for it.

 

[fedaykin]

2298 is found 109 times
9617 is found 81
7841 is found 97
8283 is found 90

Looks like I'm losing so far.
Is Pi not conjectured (but unproven) to be normal? If Pi is normal, people should average nearly 100 finds if they were to conduct a bunch of trials.

It's an interesting game. If someone were ever to get an average higher or lower than 100 in a statistically significant sample size consistently, I would love to meet him/her, but I doubt that will occur. That would be one badass human calculator with the ability to search for a sequence very quickly.

Edit: Just FYI, I am aware that there's strong evidence that Pi is normal.

 

[gerben]

I had the computer count the number of occurrences of each sequence of 4 numbers, 0001 up to 9999, in the data linked to (after removing the line breaks). It shows a distribution nicely centred around 100.

4866 sequences occur less that 100 times
4741 sequences occur more that 100 times
(392 sequences occur 100 times)

 

[raam86]

I had the computer count the number of occurrences of each sequence of 4 numbers, 0001 up to 9999, in the data linked to (after removing the line breaks). It shows a distribution nicely centred around 100.

4866 sequences occur less that 100 times
4741 sequences occur more that 100 times
(392 sequences occur 100 times)

[PLAIN]http://winblog.org/wp-content/uploads/2010/10/star-wars-halloween-costume-win.png