Statistical Analysis of the Lottery: Why Consistent Data Matters


Last updated on June 2, 2024

To conduct a meaningful statistical analysis of the lottery, it’s very important to avoid mixing datasets, as lottery games may undergo different changes over time.

Consistent data matters when conducting statistical analysis of the lottery.
Photo by Isaac Smith on Unsplash

Mixing historical results of a 6/36 and a 6/49 game will only lead to inaccurate conclusions.

Allow me to share one of my conversations with one lottery player.

Hi Edvin,

I bought your 6/46 calculator.

I downloaded “the dominant” combination template. However when I checked some suggested composition against the historical data for NJ Pick 6 wins it showed those patterns never won going all the way back to 1980 which contradicts your point of patterns being repeated every 5-6 or even more draws.

So my question is should I check every pattern in your suggestion for historical appearance and only choose ones that showed up in history which means combining combinatorics with statistics that you said not needed to use your calculators.

I am just trying to verify your point of combinatorial composition repeat themselves at least once in while in lottery drawings. The only tool I can use is the search tool for the past winning numbers here

So when I enter any suggested composition from “the dominant” list, it does not show as jackpot won at all for as long back as 1980. Essentially, I am trying to verify your combinatorial math approach by the historical statistics.


Hi Berry,

Thank you for your explanation, and I understand what you are trying to achieve.

Allow me to offer my perspective on this matter.

First, the New Jersey Pick-6 game evolved through several alterations over the years, shifting from a 6/36 format to 6/39, then progressing to 6/42, followed by 6/46, eventually arriving at 6/49, and ultimately reverting to 6/46.1

So, if you’re conducting statistical verification of the Lotterycodex analysis using data as far back as 1980, then it becomes quite challenging to establish any meaningful comparisons due to inconsistent datasets.2

With an inconsistent dataset, you achieve nothing but unstable probability distribution.

In short, you cannot mix 6/36, 6/39, 6/42, 6/46, and 6/49 datasets.

Accurately analyzing the game’s behavior becomes more challenging when the foundational rules continuously change.

I appreciate your diligence in your research. All lottery players should adopt a similar mindset. I actively encourage users of Lotterycodex to verify all calculations with historical results, and I’ve emphasized this on my website multiple times.

As you can observe in many of my articles, I’m actively involved in statistical verification. However, it’s crucial to approach this task with care. For instance, when conducting a statistical analysis of the Powerball game, I must initiate my analysis from October 7, 2015, when the 5/69 format was introduced.

This lottery analysis of the Powerball using low/high numbers composition must cover the period from October 7, 2015 to March 16, 2024.

How to Win Powerball According to Math

The same principle applies to Mega Millions. This game adopted the 5/70 format on October 31, 2017, so my analysis should start on this date to make an accurate data comparison.

This lottery analysis of the Mega Millions game using low and high numbers composition must cover the period from October 31, 2017 to March 08, 2024.

Read more: How to Win Mega Millions According To Math

Powerball and Mega Millions have undergone numerous changes over time. Therefore, extracting the appropriate dataset is imperative to conduct a meaningful analysis.

The reality is that lottery games go through various transformations over time. This is why I exercise great caution in my analysis to provide precise and timely insights.

I hope that helps. If you need further assistance, please let me know.

Stay safe,

Additional Resources

  1. New Jersey Lottery    []
  2. Data Consistency    []


This site uses Akismet to reduce spam. Learn how your comment data is processed.

  • Dear Edvin
    Thanks for your informative site.
    Quick question, do you primarily use odd/even, high/low for the dominant groups? What of adding sum of the line and consecutive numbers to the mix to create even more balanced sets? You could, for instance, choose sets from the 80th percentile for sum of the line and for consecutive numbers choose in a 649 combinations with 1,1,1,1,1,1 (all non-consecutive) and 2,1,1,1,1,1 patterns which I think covers around 50% of all occurrences.
    All the best

    • Hi LJ, once you have balanced odd and even numbers, as well as balanced low and high numbers, you’ll also achieve a balanced sum. It operates under the same principle. Try it yourself. However, let me share an insight, you may not know. The issue with sum range is that not all combinations with an optimal sum have balanced odd/even and low/high numbers. In short, relying solely on sum range doesn’t offer a granular approach. For example, your combination can have a good sum but all numbers are all even numbers. Lotterycodex addresses this problem by creating a unique combinatorial design, providing you with a granular strategy for selecting numbers. So, when you have a dominant composition, it also has an optimal sum. It’s a mathematical certainy.

Recent Posts