New York — number analysis
What the New York lots have actually done. These are records of the past; they are not forecasts, and the test at the foot of this page shows why.
Appeared most often — over 800 draws
Appeared least often
Longest since last seen
Turned up together most often
Digit frequency
Tens digit
Ones digit
Model ranking
Our model ranks numbers by recent frequency, overall frequency, how long since each appeared, and which numbers have turned up alongside the last draw. It is a ranking of past behaviour.
Tested, and no better than random.
We replayed it over 266 past draws, each time letting it see only what came before. It picked 5 numbers; it hit 40 times (15.0%). Picking 5 numbers at random hits 14.4% of the time. The difference is within chance.
Is this draw fair?
A chi-square test over 2400 recorded lots gives p = 0.30. A large value means the spread is exactly what chance produces — the draw shows no bias. The number at the top of the list above is high for the same reason the tallest person in a room is tall: somebody has to be.
Every number has the same chance in every draw — 1 in 100 for each lot, whatever it did last week. Play only what you can afford to lose.