# Hack The Lottery

This may well be the most frustrating story I’ve read in a long time. A bunch of mathematics academics in the United Kingdom have been using a system to play the national lottery. They won A\$13 million. Here’s the part that will drive you crazy: they don’t say anything about how the system works.

## 4 thoughts on “Hack The Lottery”

1. I’m not buying it. Each lotto drawing is a discrete event — the past winning numbers have nothing to do with future winning numbers (unless there’s something *seriously* wrong with the lotto, like the balls are weighted differently).

If you toss a fair coin 50 times and it lands heads every time, which side would you bet on for the 51st toss? It doesn’t matter!

2. Ed says:

I’m guessing it doesn’t.

3. Marr says:

The story describes exactly how the system works, they just used a very simple program to ensure they were choosing their numbers evenly from the entire range available. The improved probability this achieves is essentially zero. They’ve won the same way anyone else does, blind, millions to one luck.

4. It actually isn’t difficult. The bad part is it requires time. I just haven’t had time to write the program. (Mostly because I dont’ want to take the time to download 5 to 10 years of lottery results) The lottery website does do a lot of the work for you by having the historical winning numbers on their website. You just throw those in a text file and write a script (I’d use ksh as I’m a UNIX nerd) to count the occurrence of how often a number shows up. From that point you could derive sets of probable numbers. You could get much more complicated with it if you want for example, calculating a z score for the chance that the number will show up…Which is pointless to me because you get your own statistics by counting the numbers. I suppose you could do combinations of numbers for a z score such as how often you will get a 3 if you get a 5 and stuff like that. It would take time but I don’t think it would be that bad. Anyway have a good one.

P.S.
I did a mini paper version of that once with a year of lottery results and I typically got 2 to 3 results to match alot of the times. (2 to 3 results doesn’t win you anything unless you get the last number) Let’s just say you definately increase your chances and I guarantee those mathematicians didn’t win the first time they tried after they derived their equation. The real stickler is the last number, that one is a problem (I don’t know how the UK lottery works) but yeah, it’s completely possible to do. Will it be perfect, no, but it will get you close enough. Now that I wrote this maybe I will get my bottom in gear and write the code. If you don’t know how to code and want to win the lottery by developing a process of your own, I suggest you learn, it would take a dang long time to do it on paper.