Well, we have the data from all 5 drawings in this silly experiment and it is unimpressive. Here’s the raw results for each set of numbers in the 5 drawings.

Number Set | 8/16 | 8/19 | 8/23 | 8/27 | 8/30 |

Family Numbers | No matches | No Matches | No Matches | Two Matches | No Matches |

“Hot” Numbers (=>3x in last 20 draws) | No matches | No Matches | No Matches | One Match | No Matches |

“Cold” Numbers (=1x in last 50 draws) | One match | No Matches | No Matches | Two Matches | No Matches |

Quick Pick | No matches | One match | No Matches | No Matches | One Match |

Let’s take the sum of these to see if we can see any differences in the way we chose numbers and the results.

Number Set | Total Matches | Difference from Average | % of all numbers drawn (30 total) | Dollars Won |

Family Numbers | 2 | No Difference | 6.6% | $0 |

“Hot” Numbers (=>3x in last 20 draws) | 1 | -1 | 3.3% | $0 |

“Cold” Numbers (=1x in last 50 draws) | 3 | +1 | 10% | $4 |

Quick Pick | 2 | No Difference | 6.6% | $0 |

Average Total Matches | 2 |

There is a small argument to be made that cold numbers did get drawn more often, but I would put this into the area of “too close to call.” It’s not a material difference in my mind. To be a material difference I would have liked to see the dollars won be higher. We could have matched three numbers in the cold draws on different days and still not have won any, so the odds of us winning (1 in 89) were just “luck of the draw” at best.

If you are really interested in these probability type problems, I recommend checking out Kevin at the Vsauce 2 channel on YouTube. I particularly like his “The Perfect Illegal Lottery” video that talks about the history of the daily numbers.

My friend Ryan pointed out early on that this is doomed to fail based on the “Gamblers Fallacy” and he’s 100% correct. Let’s think of it on a simpler basis, flipping a coin. Assuming it’s a “fair” coin over time it should come up heads half the time and tails half the time. Just because it flipped “heads” 5 times doesn’t mean the odds on the 6th flip are greater than 50/50 that it will come up heads because the coin has no memory. So there are really no “hot” or “cold” numbers in the lottery.

Here’s some background on how the lottery drawing is executed. From working at a TV station that helped put on a state lottery drawing back in the day I know there are random sets of balls used, it’s not the same set in every drawing. There’s not only one drawing either, there are practice drawings held before the real thing. There are procedures done to weigh the balls and make sure they all weigh the same. Even the camera crew is vetted prior to the drawings, so you can’t sneak someone in.

So, there is no ‘best way” to pick your numbers for MegaMillions. Does all of this mean I’ll quit playing the lottery when it gets big? Probably not, because there is still a chance and all you need it $2 and a dream.