Welcome Guest
You last visited December 9, 2016, 2:19 pm
All times shown are
Eastern Time (GMT-5:00)

# Powerball Wheel Design discussion...

Topic closed. 31 replies. Last post 13 years ago by hypersoniq.

 Page 1 of 3
Pennsylvania
United States
Member #1340
April 6, 2003
2450 Posts
Offline
 Posted: April 2, 2004, 11:01 pm - IP Logged

I have been over the prediction track for some time, and I think it's time to move on into trapping...

In my many failed experiments with prediction, I have come to the conclusion that elimination is a big mistake, so rather than wheel 20 or 30 numbers, I would rather catch all of them a given number of times...

Here's what I am working on for Powerball so far...

the UberWheel...

Lines = 42 (this is the basis for the wheel, play all 42, or target 3 for elimination---riskier--- and play 39, but that remains to be seen)

Guarantee = one \$3 winner... that's it... anything else is up to luck

This wheel will be for wednesday drawings only, and will be grouped as follows...

5 lines each containing ONE number from last wednesday's drawing.
5 lines each containing ONE number from two weeks ago.
5 lines each containing ONE number from FOUR weeks ago.
10 lines containing TWO numbers from last wednesday's drawing.
1 line containing THREE numbers from last weeks drawing and TWO numbers from 2 weeks ago.
21 lines containing ZERO numbers from the last 4 wednesday drawings (wheeling all of the remaining 33 to 48 non-repeats).

Total is 42 lines, powerballs played from 1 to 42...

now this is still in the planning stages, but the dvents I am tracking (repeats from 1 to 4 weeks ago) are significant enough to warrant some kind of attention, I am hoping to catch either the occurrence or NON-occurrence of these dvents. Distribution will be based on an 80% to 90% rule based on previous draw history (the numbers in each position had to hit in 80 to 90 percent of the entire history).

I am still checking the data to see if any of the sets can be modified or combined, but how does the idea sound so far?

Playing more than one ticket per game is betting against yourself.

Pennsylvania
United States
Member #1340
April 6, 2003
2450 Posts
Offline
 Posted: April 3, 2004, 1:35 am - IP Logged

another, simpler wheel would involve 44 lines and be open for both draws (Wed. & Sat.)...

made up of four eleven-line groups that contain all 53 numbers (all 53 can be caught in 11 lines with 2 open selecctions), all 42 powerballs and 2 picks for powerball of your choosing...

Guarantee = \$3 + having all 5 white balls 4 times

Playing more than one ticket per game is betting against yourself.

MD
United States
Member #1701
June 18, 2003
8363 Posts
Offline
 Posted: April 3, 2004, 1:47 am - IP Logged

I never tried it like you have it posted. I think it's a good idea because some numbers do repeat and leaving them out narrows your chances greatly. In the same respect if no repeats show up your dependent on the remaining numbers to show either way you still have a chance.

I tried breaking the spread into three groups so I could wheel the whole group of 52. And wheeling them. Two groups had hot and cold numbers and the third group had hot cold and lukewarm numbers. Sometimes I capture three white balls doing it this way. Cost to much though 51 bucks. I used the power balls stats to select the power ball.

Pennsylvania
United States
Member #1340
April 6, 2003
2450 Posts
Offline
 Posted: April 3, 2004, 1:13 pm - IP Logged

One of the big problems I am facing right now is how to populate the first 25 lines in addition to the numbers that are repeated... I need to more closely analyze the trends... like... if the First number repeats, does one from 2,3 or 4 draws back accompany it more than 50% of the time?... I am basically trying to forecast dvents (repeats) rather than actual numbers, by using the wheel format I can account for each number repeating.

I have the lines built up on a spreadsheet, when I enter the last 4 wednesday results ,it populates the list with the chosen numbers, but there are picky little details like doubles (and even triples) to contend with... the way I see it, the wheel will need to be worked by hand before each wednesday... I have no plans on actually playing an expensive wheel like this until the jackpot gets really big again... and even then only if "dry runs" prove profitable on "paper" first... But I have no more ideas for prediction so it's time for a change in strategy ;-)

Playing more than one ticket per game is betting against yourself.

MD
United States
Member #1701
June 18, 2003
8363 Posts
Offline
 Posted: April 3, 2004, 1:27 pm - IP Logged

Do you take a list of numbers from one to 53 and write down how long a number has sat out and or most recently hit. Keep really cold numbers in one list really hot numbers and lukewarm numbers on another list. I think lukewarm numbers are the key and one cold number and one hot number. Try making tickets without wheeling them and see what happens. By selecting the numbers to play from the three groups. Or take the numbers you selected and wheel them. Actually you could do each example separately for a trial run. That's my strategy for both mega and powerball.

United States
Member #3680
February 11, 2004
187 Posts
Offline
 Posted: April 3, 2004, 1:57 pm - IP Logged

Here are a couple other things to maybe look at to narrow down the number of balls:

I was updating my Mega Millions database with last nights numbers and replacing the last date drawn dates and noticed every one of the white balls had been drawn within the past month.  I don't know how often that happens but that may be another way to narrow down the numbers.  It basically follows along the lines of hot/cold numbers.  But to give an example, here are the white balls for Powerball that were drawn within the last month and we'll see how they do tonight (28 numbers total):  15, 31, 34, 39, 50, 2, 17, 30, 18, 20, 43, 47, 7, 24, 11, 44, 45, 48, 10, 22, 41, 49, 9, 46, 51, 5, 23, 53.

I was updating the Powerball Drop Theory spreadsheet (I've put an updated copy of it on the web for download in case anyone is interested - it's in the same download location as mentioned in the previous thread on the theory - http://www.lotterypost.com/thread/83317).  Going back to the beginning of the year, there was only one drawing where a ball from each column was drawn.  So maybe only wheeling four of the columns would be a way to eliminate numbers.

Just some ramblings -

plnwebguy

United States
Member #3680
February 11, 2004
187 Posts
Offline
 Posted: April 3, 2004, 10:12 pm - IP Logged
Quote: Originally posted by plnwebguy on April 03, 2004

Here are a couple other things to maybe look at to narrow down the number of balls:

I was updating my Mega Millions database with last nights numbers and replacing the last date drawn dates and noticed every one of the white balls had been drawn within the past month.  I don't know how often that happens but that may be another way to narrow down the numbers.  It basically follows along the lines of hot/cold numbers.  But to give an example, here are the white balls for Powerball that were drawn within the last month and we'll see how they do tonight (28 numbers total):  15, 31, 34, 39, 50, 2, 17, 30, 18, 20, 43, 47, 7, 24, 11, 44, 45, 48, 10, 22, 41, 49, 9, 46, 51, 5, 23, 53.

So much for that theory :)  Only two of the numbers - 7, 43 - were in the bunch.   Hit two others on my 10 number wheel on paper of popular numbers - 6, 26 - so basically the same prize at a lower cost this time.

Wheeling columns 1, 2, 3, and 5 (column 4 very seldom has any picks, so I would have left it off) of the drop theory spreadsheet though would have hit all five.  In fact, only wheeling 1, 3, and 5 would have done the job.

plnwebguy

Pennsylvania
United States
Member #1340
April 6, 2003
2450 Posts
Offline
 Posted: April 4, 2004, 5:25 am - IP Logged

I have looked at decades and columns based on how the balls are dumped into the machine... every time I think I figured something out, the next draw would always shred the theory to bits (Announcers, Announcer Ranges, Decade+Last Digit, Column+Last Digit, Skips (the hot-cold-warm scenario) etc...)

The only relatively consistent thing I have found are the repeats from 1 to four draws back on a wednesday... completely excluding saturday draw data... The ranges per position are definitely narrower than full draw history... most of the drawing "spikes" are from saturday data.

There are 9 "chains" of 3 draws in wednesday's white ball data (where a number was drawn 3 wednesdays in a row), and only one on saturdays.  On saturdays, there are 31 white ball repeats, on wednesdays there are 51, a significant difference... Even if you were working with saturday data, repeats can not be eliminated, even in the entire draw history, but on the whole history and on saturdays, they are not as frequent as on wednesday data.

Patterns are all over the place, the trick is separating the significant ones... I think I found one so far (wednesday repeats) but there must be more somewhere...

Playing more than one ticket per game is betting against yourself.

United States
Member #3680
February 11, 2004
187 Posts
Offline
 Posted: April 4, 2004, 7:03 am - IP Logged

I went back and looked and the 6 and 26 had been drawn the last week in Feb., so going back 5 weeks would have hit 4 out of 5.  I have noticed something about the powerball that may help you cut down on the ones to not wheel - it doesn't seem to repeat over a month's time.  The number 6 was the exception for March, being drawn twice, but all the rest were unique.

I did go back and look at Wednesday's draw back to the first of the year and see what you mean about repeats, 17 comes to mind.  It even seemed like the 4th column in the ball drop pattern had most of it's hits on Wednesdays.

plnwebguy

Pennsylvania
United States
Member #1340
April 6, 2003
2450 Posts
Offline
 Posted: April 4, 2004, 10:35 am - IP Logged

My end goal is to use every number, and have a reason to use it, then it could simply be an excel thing to fill in what is needed and have the lines fall into place automatically... Because I am trapping the absence or occurrence of dvents rather than numbers, it should be general enough, and since it uses actual draw history, the source of entropy should be good enough too, it will be modelled on the history itself...

Time will tell... still about 103M away from a live go (provided it rolls that long, otherwise I will have plenty of time to refine before trying for real)

Playing more than one ticket per game is betting against yourself.

Pennsylvania
United States
Member #1340
April 6, 2003
2450 Posts
Offline
 Posted: April 4, 2004, 11:08 am - IP Logged

there are situations where the 25 lines based on repeats can be consolidated, this week is a prime example, as the number 20 was in a chain 2-3 and 4 weeks ago... it could have been eliminated last week (because chains seem to ALWAYS break at 3) and this week it shows up in 2 playable groups of 5.

Playing more than one ticket per game is betting against yourself.

MD
United States
Member #1701
June 18, 2003
8363 Posts
Offline
 Posted: April 4, 2004, 11:57 am - IP Logged

Hypersoniq wrote: My end goal is to use every number, and have a reason to use it, then it could simply be an excel thing to fill in what is needed and have the lines fall into place automatically... Because I am trapping the absence or occurrence of dvents rather than numbers, it should be general enough, and since it uses actual draw history, the source of entropy should be good enough too, it will be modelled on the history itself...

Time will tell... still about 103M away from a live go (provided it rolls that long, otherwise I will have plenty of time to refine before trying for real...

I think that is what I said in my first reply to your original post. Playing all the numbers. By doing it this was you should get at least 3 and a power ball. And possibly by a fluke get all four or five white balls.

mid-Ohio
United States
Member #9
March 24, 2001
19830 Posts
Offline
 Posted: April 4, 2004, 1:07 pm - IP Logged

In the last 50 drawings all 5 white balls have occurred at least one during the previous 7 to 59 drawing averaging 23.6 and a median of 20 draws.  More the 50% of the time all 5 numbers have been in the previous 23 drawings.  This fact may only eliminate 3-6 numbers at the most, but every little bit helps.

RJOh

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

Pennsylvania
United States
Member #1340
April 6, 2003
2450 Posts
Offline
 Posted: April 4, 2004, 1:23 pm - IP Logged

oops, I made a mistake, there would only be 16 lines containing non-repeaters, not 21....

Playing more than one ticket per game is betting against yourself.

Pennsylvania
United States
Member #1340
April 6, 2003
2450 Posts
Offline
 Posted: April 5, 2004, 3:57 pm - IP Logged

Wednesday data charted, finally... I see by doing that that I missed the tail end of the PB trend which went on for 6 drawings... PB then PB-1... the 35 last wednesday busted it up, again always too late...

I figure I should save enough \$\$\$ by not playing saturdays anymore to give the wheel a try this run if it rolls that high... Still establishing guidelines for the number placements, and luck is still a major factor, but I think I can at least be in the position to have the numbers in the right "form" for each scenario that involves repeats, and the last 16 "garbage collector" lines will do better if I was wrong in the top 26... the next vexing challenge is the position of the repeat(s), they vary... and with no particular reason...

Playing more than one ticket per game is betting against yourself.

 Page 1 of 3