hypersoniq's Blog

Before moving in a positive direction with my excel work against the PB, I gotta let out the negativity, so here goes...

First, let's start on the same page, with the simplest of all games (nice round numbers), the straight pick 3. (not worth the time to get into box play)

In pick 3, you are playing 3 sequential games of 1:10, you MUST match all 3 games correctly to have a straight hit.

odds of first correct ball = 1:10

odds of also selecting the second correct ball = 1:100 (or 10 possible first game combos times 10 possible second game combos)

odds of ALSO selecting the third correct ball = 1:1,000 (10x10x10)

the range is 0-9 for each game, 000 to 999 for the entire process. To further prove the math I'm about to lay down, most serious pick 3 players know there are 720 combos that contain NO doubles or triples... the derivation is to imagine 10 balls and they pick 3 of them

first ball 1:10

2nd ball 1:9 (one ball is gone)

3rd ball 1:8 (two balls are gone)

10 x 9 = 90,  90 * 8 = 720

All in agreement so far?

now let's look at what the MUSL advertises as the odds for a powerball jackpot win... 1:146,107,962    ... and it's flat out wrong. False on purpose? who knows.  NOW let's do the actual math...

the game starts with one group of 55 white balls and one group of 42 red balls. the odds of matching the first number...

1:55  simple enough. on to two numbers

55x54= 1:2,970 (55 for the first ball, then 54 remain) 

three numbers (a prize they claim has odds of 1:290) 55x54x53 = 1:157,410

four numbers (they say 1:14,254 ... I say) 55x54x53x52 = 1:8,185,320

five numbers 55x54x53x52x51 = 1:417,451,320 (wait, it gets better...)

odds of winning the powerball jackpot, 55x54x53x52x51x42 =

1:17,532,955,440

you read correctly, that is 17 BILLION to one!!! Makes that 15 million minimum look a bit low, yes?

you may slice/dice/analyze this as you see fit... but it's true.

you start with 55 balls, the first one could be any of the 55... the second could be any of the 54 remaining...

This is what we're all up against.... and I STILL accept the challenge!

the bright side is the 42:1 odds of matching a red ball... that is where I choose to begin anew. Also, all 17 billion are not unique, many sets contain the same numbers. BUT trying to create a system that uses draw order data does require you realize the TRUE odds.

As I mentioned last post, I am right now only concentrating on the 1:42 red ball in PB. I am working on an excel file called PROfile that presents several bits of info...

1. RAW odds

for example, the number 5, RAW data would indicate that there are 4 numbers lower, 1 repeat, and 37 higher. using percentages, you can see raw odds of the next number drawn being higher/lower (5 favors higher)

2. Observed history

same scenario, only using observed history.

3. Confidence index

which is simply #1 * #2

4. Binary odd/even data

what has been the count of odd vs. even followers

and finally

6. HEXades

more on that later...

I pretty much dropped off the lotto prediction software game last year when I lost BOTH hard drives in my PC (one was just for backups) to a fried IDE controller card. 2k6 was a miserable year, lottery wise... stuck to playing one ticket with hardly any small hits.

I have since rebuilt my PB sheet and plan to go in some new directions.

My goal remains a one-line prediction... however the methods are getting way out there. 

brief rundown of the first one...

the periodic system...

synopsis: assign each drawn number to it's corresponding element on the periodic table and try to use some chemistry formulas to gain the next number.

So far: one excel sheet, 3 worksheets...

A: the full PB history back to 1997

B: the jackpot calculator (for motivational purposes only)

C: the conversion page

for instance, last night's draw

01-05-26-32-38 + 30  becomes

 H- B-Fe-Ge-Sr  + Zn

info kept includes the element's valence number and periodic family, also whether it is a nonmetal or transition metal... all done with Excel's LOOKUP() function. I'm hoping to possibly use lewis dot notation and valence info to unearth some hidden info... not convinced it has a chance, but nothing else has worked, so why not? at least I will be one who can say I have tried EVERYTHING ;-)

I have been sustaining the PB "habbit" with two QPs per draw, and will continue to do so until I have a single pick from some home brew system to replace one of the QPs. The nice part is that regardles off winnings, I know the maximum total cost of 2007 will be $416 (because I go for the PowerPlay, even less tho because I only play 2 QP's when the jackpot is over 40M). I've sworn off all other lottery games for good. I also spent much less in 2006.

It felt good to fire up excel once more, gotta shake off the rust! outside of flaky systems, I have goals in mind for excel this year, including finding matches regardless of position and more accurate backtesting statistics. I will set some kind of backtest goal before replacing one of the QPs, much higher than in the past. Also I will be working separately on JUST THE POWERBALL. Separate sheets, formulae, programs etc... My programmer's new years resolution is to work on that pesky red ball. They pick ONE from 42, why is it so hard just to pick that one? I hope to answer that question. the pb only progs have priority #1 for my limited time behind the keyboard.

so what are you planning this year? 

I basically suck at picking numbers... latest PB challenge I entered 15 numbers, not a match...

for powerball I had a 13 and 15, 14 comes up grr..............

my betting strategy does keep the wallet happy tho, 1 ticket per $100,000,000

I spent $3 for last night (I gave up the powerplay unless it can be paid for by the previous draw's winnings, and I am in what can best be described as a "dry spell" )

nothing seems to be working for me, I am tempted to just pick up 3 QPs for wed. I think I will...

should be interesting for a winner faced with 2 options

1. Annuity: after tax consists of 30 payments of $3,900,000 they will deduct the 27% and the winner will need to pay $480,000 each year to hit the current tax rate (will go up as the top tax rate goes up)

2. Cash: after tax one-time payment of $56,680,000 (you owe $6,976,000 during the next year in taxes and would have $56,680,000 left free and clear)

the end value of the annuity depends entirely on the tax rate, could fluctuate but will probably never dip below the current 35% (40%+ next liberal president).

If you choose the cash, where do you put it? it is a one-time wire transfer... FDIC insurance is 100k per account PER FINANCIAL INSTITUTION (100 savings accounts at the same bank don't protect you)... highest progra I have ever seen for FDIC protection is a CD-ARS account but that has a cap of $20,000,000...

wonder if that is by s.s.n. ? then a married couple could have 2 accounts at 2 different banks and get 40 million of insurance... putting the balance (16 million) into investments/real estate/municipal funds/trusts for the kids... so many options, too easy to be taken advantage of... I would definitely opt for at least one account in the CDARS format... ideally it will work by s.s.n. then I could have one, the mrs. and each of the 3 kids can have 4 mill (the trust funds) all totally protected by FDIC insurance.... sounds like a definite conversation starter if I ever need to consult a financial planner...

ah, yes but there remains that pesky detail of actually winning...

but I am at least aware of some of my options if I do... come on MUSL, let me have a shot at solving those type of problems...

there appears to be no one formula for solving problems that we need to solve. there are, however different approaches to doing it.

1,2,3,4,?,6

of course it's 5, and to describe the relationship is

n+1

for n+5 with a start number of 10 we have

5,10,15,20,25,30

-------------------------------

for THE FOLLOWING

1,4,10,22,90,182

(n+1)x2

--------------------------------

I am looking at trying something on the powerball (just the red one, not the white 5 for now)involving backsolving an equation thatthe data can fit into... I can't see anyother way around it. I have googled for days (weeks even) and according to mathematics as we know it, there is no general formula for this type of thing. I am going to try some different things out including interpolation (to find midpoints in draw data) and then extrapolation (to find the next draw basedonresultsfrom interpolation).

I'm otherwise out of ideas again, and I hate being out of ideas with a PB jackpot over 100M

sorry for anyone playing along, but I am no longer developing or updating any of the systems I have been working on and sharing until I go back to basics and get concrete steps defined for the process of prediction.

website will stay up, sheets there now will remain... but I am tired of spending time crunching numbers with no idea how to really interpret the results. I'm even going back to QP only for PB...

hopefully I can share a newer system soon with some real power behind it (probably not, but you never know)

what do we DO with draw history?

we could convert it any number of ways (high/low)(odd/even)...

but then what?

we have a ton of filter tutorial threads yet not one single post on how to interpret the results...

why? why not?

deltas... converts the data into changes from the last draw... but then what?

skips... some swear by them, why? how are they to be interpreted?

we have a board full of people capable of gathering and presenting data in wonderful ways, able to extract every statistic possible... but then what?

time to move everything to the next level... time to define once and for all what prediction is, what a trend is, what a pattern is... we need interpretation tutorials... data mining instructions are good, but what to do with the results would be priceless...

this is our "missing link", this is why NONEof the systems can pass backtesting with good results.

any ideas? any links? anything?

so far, the coding is going well... need to get some variable seed sweeps going, gotta learn more about excel to pull that off.

I can make it pick fixed numbers with the rnd(-1) call, and randomize using any seed I want without changing after a recalc... now to find the right seed or seed progression to make some hits appear...

I have a backtest sheet that uses simultaneous picks, it will pick 1 three-digit number and then 3 individual digits to try both methods.

sheet for pick4 is ready just as soon as the pick3 shows some potential.

this is only good for trying individual seeds, to sweep I think I need straight VB

SB is a no-go, can't get any improvements.

on to the next set of projects...

up now is the PRNG synch for PA mid 3/4, and some paper testing of a "focused forecast" for powerball... too complex for backtesting right now, lost a bit of ambition on the SB bust...

I have to start PA mid testing clean-sheet because I don't normally track CPU draws. seed fortesting will start at the number value of the date+time (also known as NOW() in excel) adding 13:10 as a start and working forward.

I will be using VBA for the RNG, calling it in the sheet as a custom function. I will use several functions at a time for streamlined testing, not holding my breath for any great results, but gotta try

SB is a no-go, can't get any improvements.

on to the next set of projects...

up now is the PRNG synch for PA mid 3/4, and some paper testing of a "focused forecast" for powerball... too complex for backtesting right now, lost a bit of ambition on the SB bust...

I have to start PA mid testing clean-sheet because I don't normally track CPU draws. seed fortesting will start at the number value of the date+time (also known as NOW() in excel) adding 13:10 as a start and working forward.

I will be using VBA for the RNG, calling it in the sheet as a custom function. I will use several functions at a time for streamlined testing, not holding my breath for any great results, but gotta try

I put up a powerball sheet on my page

important... IT DOESN"T DO ANYTHING YET, it uses data all the way back to 1997 and it's small in size

the only "pick" it even attempts to make is the raw result of FORECAST

there are 4 sheets...

Worksheet 1. Draw Data

  column A = draw date

  columns B-F = white ball draw data SORTED (WS1-WS5)

  column G = powerball 

  column H-L = white ball draw data DRAW ORDER (WD1-WD5)

  column I = day of the week (wed./sat.)

  column J = draw number

  old draws are at the top, add new draws to the bottom...

Worksheet 2. Frequency

  column A = number (1/55 covers PB 1-42 also)

  column B = total count (how many times appeared in the game whether PB or WB)

  column C-G = sort order freq.. how many times each number hit in each position (SORT ORDER)

  column H = powerball frequency

  column I-M = draw order freq..

      row 58 contains the largest number of each column, this is for conditional formatting to allow the highest one to be highlighted.

Worksheet 3. Statistics

a look at some statistics for the game history

Worksheet 4. Cash Value Calculator

enter the data requested in the green cells (MUSL advertised cash value), see what you can actually expect if taking the cash value

don't know what else it's good for at this point, but it might be a good start for something...

uses formulas extended to row 2000 for fast recalcs (no need for F9 in this one), after row 2000 on history sheet, formulas will need to be changed... that's a LONG way off

enjoy

still looking for improvements, not finding results...

also wondering how to apply something similar to a 5/39 gamelike PA Cash5

and ultimately to the only game I actually play, powerball...

maybe I need a completely new idea... I'm out of them at the moment...

to copy from the old cliche...

I have good news and bad news...

the GOOD news is that ALL formulas have been fixed with some help and the keen eyes of other LP excellers

the BAD news is the hits went away... first sheet corrected went from 64 hits down to 8...

this fails both long term cost/winnings AND cyclic hit rates.

Usually I would stop here, write off the system as YAFA (yet another failed attempt), but there must be more progress that can be made... the concept still seems valid... glad I never got to the part where I buy tickets for it... I'm riding this one out a bit longer, going down with the ship...

I will save a version with the bad formulas, false hope is better than none ;-)

these picks are from each of the 4 SB sheet variants...

good for tonight in PA...

L2H_FH = 4-4-9
H2L_FH = 4-4-9
L2H_RH = 4-4-0
H2L_RH = 4-4-9

see how this works... the 4 systems picked the same first 4 digits, and the same last digit 3/4 times....

you could play this twice and be done.... play the most frequent(449), play the oddball in the group (440), play heavier on the 449, or skip it altogether.

the number of repeat picks like this ensure the system will remain cheap to play ;-)