I have had a naive, yet novel idea over some months and need to let it off my chest.  Since I began wondering how random numbers work, I decided to check them out.  So, over many months' deliberating, I came to the conclusion that randomness may exist but since we have the databases available, then my idea would be to use these pseudo-random numbers to best estimate and predict my megamillion winners.  Later on, I ventured into checking out other lotteries, with six and fivers.  Mine is less mathematical but more logical, even though I have made statisitcal assumptions as the basis of simple projections that can mint my ''my millions'', as I travel around my world of mathematics and logical extensions.  How? 

By adjusting my mathematical assumptions to mimic graphical movements of my imaginary logarithm, which travers the entire database, based on the simple logic that nothing ever tried could predict lottery winners accurately.  So here are my basic assumptions: 

1) That there are only 6 balls drawn at each game; and that there are spread over 6 columns (Excel-based). 

2) Each column has a maximum of the total number of main balls (75 in megamillions and 59 in powerball, for an example), each with equal significance.

3) I have used Excel in my model, and within that, the assumption is that my logarithm defines the positions of six main intervals, spread over 30 rows.  The starting point of my model begins with the basic spread of the 75 balls downwards, from column A to F (or 1 to 6).  The moneyball position has not been given a special significance and is treated as being equal to the rest of the numbers. 

4)At the end, my model has been based on logarithmic variations, with the main assumption being that at any given time, the game can only draw on the set of intervals at the beginning of my database, on top, made up of 30 rows, and of these, only 6 rows are active, at all games. The intervals are in total 6, which repeat over and over thoughout the database, but the first set is significant. These intervals are specific to a panel (made up of 6 rows, whose values differ).

5)So far, after testing and adjusting my method, over many months, I came up with the following: That each number drawn has variants defined by its original value, at the interval (I call it a 'hinge'), plus its present value and adjusted by its future position. 

6) Excel has been the main tool I use,  to simplify the method.  In doing so, each and every row has a specific cell address, as defined by my intervals, hinges.  To define them, I have tried to demistify randomness by assigning cell values with letters and numbers.  In this model, my starting point being '1', and it nullifies the idea of randomness.

How does this look like on my baseline (an outline of real hard numbers from 1 to 75)?

7)When laid out in each column, the first numbers (which defined my baseline position) are as follows:

        1, 2, 3, 4, 5, 6    and thereafter, this row defines the entire argument, so that vertically, the argument continues logically, i.e. 1, is followed by 2, 2 by 3, 3, 4 and so on, for the entire set of values that make up the sum of balls used in a game, 75, 59, or whatever else.  These are then divided up into panels, which create a 'home' to each of my hinges, with variants in baseline positions.

Now, am attempting to establish the parameters that define the game winners in compliance with my belief that randomness in my database, can never exist.

To be continued...