Welcome Guest
You last visited July 22, 2017, 4:58 am
All times shown are
Eastern Time (GMT-5:00)

# Choosing a range

Topic closed. 10 replies. Last post 13 years ago by Pick-4_Master.

 Page 1 of 1

United States
Member #5599
July 13, 2004
1196 Posts
Offline
 Posted: July 16, 2004, 8:55 pm - IP Logged

In general, when your about to anaylze the numbers in a variety ways, one of the first questions that comes up is how many games do I go back. What sample size is going to best reflect what's going on. One of the problems with the pure mathematical approch is that it assumes a huge sample size, our  great grandchildren will probably be dead and gone before the sample size begins to reflect a pure mathematical model. Not to mention, what path will be taken to acheive the end math model. ie 100 games of low values, 100 games of high values, then a few 100 games of average values. Then again, it may be a  small incremental mix of all during the course of traveling to the pure math model.

My self, when detemining the amount of history I want to use, I'll do an initial analysis of several values. In the case of looking at occurences, I found the one range that had the least variance from the expected average value. For fantasy 5 it ended up to be about 23 games for an approximate value of 3 picks per number over that range.  I see large variances as the enemy. When doing other comparisons the game range changes often changes to yeild the most reliable and consistent data.

Does anyone have another method to the madness in choosing the amount of history to look at when analyzing numbers?

New Member
Jeffersontown, KY.
United States
Member #5683
July 16, 2004
1 Posts
Offline
 Posted: July 16, 2004, 9:05 pm - IP Logged

We will see if it works but I go back 10 draws. Using evening numbers of pick 4.

I look for the string connecting Mid day with evening then make my picks.

I place the numbers as if it were a slot machine selecting the position based

on the most favorable odds of each slot.

Then again sometimes I just make a guess at the position if there are dual

numbers. Since my limit is max of \$5.00 on any one game I don't lose sleep over it.

mid-Ohio
United States
Member #9
March 24, 2001
20097 Posts
Offline
 Posted: July 16, 2004, 10:13 pm - IP Logged

I have a routine in my program that count the number of previous drawings you have to include to cover all the numbers in a drawings.  In MegaMillions 75% of the time that is 28 drawings or less, so I go back 28 drawings.  MegaMillions have 52 numbers so it requires more drawings than Ohio649, which I also play, to cover 75% of the time.  Buckeye5 with 37 numbers take even less drawings to do 75% coverage.

RJOh

* you don't need to buy more tickets, just buy a winning ticket *

United States
Member #1759
June 29, 2003
1156 Posts
Offline
 Posted: July 17, 2004, 4:40 pm - IP Logged

In FL I use (14 draws for Pick-3/4),(30 draws for Pick-5

and (52 draws for Pick-6).

United States
Member #5599
July 13, 2004
1196 Posts
Offline
 Posted: July 17, 2004, 9:26 pm - IP Logged

Pick-4 Master,

Is there a method to the madness of why you picked 30 games for pick 5 and 52 games for your pick 6?

Rjoh

I came up with 28 games for the 6/49 also. I can usually expect an occurrence average of 3 when analyzing the individual numbers.

mid-Ohio
United States
Member #9
March 24, 2001
20097 Posts
Offline
 Posted: July 17, 2004, 9:40 pm - IP Logged

JKING,
When I use the same routine on Ohio649P for the last 400 drawings, 23 covered 75% of the drawings and 28 covered 85% of the drawings.  It all depends on the amount of coverage or how wide you want to set you parameters.

RJOh

* you don't need to buy more tickets, just buy a winning ticket *

United States
Member #5599
July 13, 2004
1196 Posts
Offline
 Posted: July 18, 2004, 3:30 am - IP Logged

This is exactly why I like reviewing and re-reviewing basic concepts. Because, given a basic question, the interpertation and implementation varies from player to player. And which one is yielding the best results?

I think you'll find, as you increase the coverage, specifically to this topic, you'll see a dramatic increase in the variances of data your analyzing. Increased variance mean increased error margins. I ran ranges that yield occurance values of 1 thru 10 and found the optimum value to be 3 or 23 games. I'm not saying I'm perfect or 100% right. But from the way I see and anlyze data, it just seemed to be the best result.

Bangalore
India
Member #2444
October 4, 2003
234 Posts
Offline
 Posted: July 18, 2004, 3:59 am - IP Logged

Dear RJoh:

Just a curious thought: With so many features in your program, as discerned from your numerous mentions in posts here, I was wondering why haven't you thought of marketing it yet, after writing a modified version of it?! Or is that you developed it solely for your own research purposes and NEVER to market it?! After reading your posts, I feel it is natural for anyone to salivate to have such a program in their kitty, you know...LOL.

BTW, what has been your biggest win so far with your own program?! Or do you win SMALL and win OFTEN?! Do the winnings offset/override the costs of your wager?! And...WHAT name have you given to your program?! Thank you.

Good Luck!

Just Monkeying Around With Numbers...!

mid-Ohio
United States
Member #9
March 24, 2001
20097 Posts
Offline
 Posted: July 18, 2004, 10:14 am - IP Logged

DreamLotto,

My program is a work in progress and I have added features over the years that I couldn't find in any commercial programs. It only works on 5 and 6 digit games with or without a  bonus ball and can handle up to 60 numbers. It's written in GWBasic so it's not of marketable quality. While it does tells me a lot about past drawings, I haven't won a jackpot yet, the closest I've come is 5 of 6 in the Ohio lottery a couple of years ago.  I do win some small prizes once in a while, but they don't cover the cost of playing, just keeps me interested in playing more.  The final compiled version of my program is called "LOTA", but it has had different names in the past.  Years ago I worked in an Industrial Engineering department and did a lot of work samplings that were used to develop production standards and thought that some of the same principles could be used in playing the lottery and started work on my program and it has become kind of a hobby since.  Some of the information I've shared here is the type of information I use in a RNG program to generate my picks within the parameters of past drawings.

Good luck to you too.
RJOh

* you don't need to buy more tickets, just buy a winning ticket *

United States
Member #1759
June 29, 2003
1156 Posts
Offline
 Posted: July 19, 2004, 12:29 pm - IP Logged

"JKING" I forgot to add something to my history cut-off range so here it is.For the Daily games (7 draws minimum

14 draws maximum) for the Pick-5 (15 draws minimum 30 draws maximum) and finally for the Pick-6 (26 draws minimum 52 draws maximum).

United States
Member #1759
June 29, 2003
1156 Posts
Offline
 Posted: July 19, 2004, 12:36 pm - IP Logged

I got this from various Lottery Books and magazines I have that contain number analysis information and ideas.

 Page 1 of 1