Didn't the New York State lottery draw 9 1 1 on 9/11? Lotteries are weird.

Just remember, since every ball draw is independent of all other ball draws, and the probability of a certain ball being drawn is uniformally distributed among all balls, these stats have no bearing on future performance.

According to John Von Neumann's game theory, the best strategy for lottery is not to buy lottery. That maximizes your expected winnings, because it makes them zero, and with any other strategy they will be smaller than zero.

next time you morans fill out your "stupidity tax" forms, remember to not circle 13.

Yes, but what is the likelihood of drawing the number 42?

/DNRTFA

count the dots

LOLZ

count the dots

LOLZ

Excuse me, the alarm on my Skinner box is ringing.

Actually, law of averages says 13 is the most likely number to come up in the future (over time).

Tanqueray

You're interpreting the law of averages incorrectly.

As supporting evidence, from Wikipedia:

"I just got 5 tails in a row. My chances of getting heads must be very good now" is an example of a false perception. It was unlikely at the beginning that one would get six tails in a row, but the probability of six tails was the same as five tails followed by a head: 1/64. Looking forward after the fifth toss, these probabilities are still equal. The only difference is that there are no other possibilities, so the probability of either outcome is 1/2. This error can be devastating for amateur gamblers. The thought that "I have to win soon now, because I've been losing and it has to even out" can encourage a gambler to continue to bet more. This is known as the Gambler's Fallacy.

I'm not a big fan of the lottery. If I'm going to gamble away my money, I'd rather do so playing a game which allows me to influence the outcome in some way.

Like betting on which 60-minutes anchor will die next.

Yes - and it drawed 666 the next day.

Illinois had 666 on Christmas a few years ago.

Actually, law of averages says 13 is the most likely number to come up in the future (over time).

There is no such thing as the law of averages. The number 13 has the same odds of being picked now as ever. In pure luck situations, the past has no effect on the future.

666 meant "Caesar Nero" back when John The Nut Job wrote the book of Revelation, no more, no less. The association that evolved over the centuries... that it has something to do with the end of the world, doomsday, etc.... is complete BS.

HomoHabilis

While I know that to be true, my love of irony is still fond of the fact that the numbers believed to be associated with the devil came up on a day believed to be associated with the birth of the son of God.

Physicists -- is this part of that whole quantum entanglement thingamabob?

Dear FARK: my lucky balls have been quantum-entangled. Please help.
/chair with wooden etc.

You guys make the assumption that each ball has an equal chance of being drawn than the others from the beginning. With enough trials, the law of average does say that every ball should be drawn as many times as the others. As some previous poster says, it doesn't mean that it is "bound" to appear more often in the future, as the future results do not depend on the past ones. _However_ any good Bayesian will tell you that the past may be an indication of the true odds at play. Let's say that we _don't_ know what are the odds that a certain ball will be picked. What would you do? I'd collect some data (watch the loto), count the number of times a certain number appears and divide it by the total number of trials.

One could make the case that some balls are less likely to be picked due to some factors that may remain unknown (such as one ball weighs more than the others or something) if they were so inclined. One could make a hypothesis test with H0 ball 13 has as much chance to be picked than the others and H1: 13 has a lower chance to be picked. find the p-value of that test and if it's not too low then it is indeed just by chance that 13 was picked so few times compared to some others. It all comes down to the number of trials and what you think is an acceptable p value, really.

/ amateur statistician
// Bayes ftw!
/// My first slashies!

I'd collect some data (watch the loto), count the number of times a certain number appears and divide it by the total number of trials.

I think such methods stopped being useful a long time ago. Someone broke the bank at Monte Carlo by noticing that a certain Roulette table had a bias, since then the tables at casinos get changed around every so often, as do the balls in the lottery. I think there's about a half dozen sets of balls that get rotated in and out at random intervals.

Hehe I wasn't actually serious about it :P. As someone said before, the expected gain from lottery is much less than 0 if the game is fair. If it wasn't so, the lottery organisers wouldn't gain money, now would they?

About the supposed bias: if the lotto people let a machine have a bias they'd be out of money pretty quickly as people start to realise it. That's why they employ statisticians to do what I just described. It's most probably just another statistical abnormality. They do happen. But hey, it's still fun to figure out the odds.

Developer comes up with 'intelligent combinations' for winning 6/49 lottos.

Oddly enough, still looking for investors.