Active TopicsActive Topics  Display List of Forum MembersMemberlist  Search The ForumSearch  HelpHelp
  RegisterRegister  LoginLogin
Superseded Normal Distribution Lotto Record Covers
 LottoPoster Forums : WORLD RECORD LOTTO COVERS (WHEELS WITH GUARANTEE) : Best Yield Normal Distribution Lotto Covers : Superseded Normal Distribution Lotto Record Covers
Message Icon Topic: Record 289 Lines Cover 6/45 Lotto Unique 3's Post Reply Post New Topic
Author Message
Colin F
Lotto Systems Tester Creator & Analyst
Lotto Systems Tester Creator & Analyst
Avatar
To dream the impossible dream ...

Joined: September 30 2004
Location: Australia
Online Status: Offline
Posts: 678
Quote Colin F Replybullet Topic: Record 289 Lines Cover 6/45 Lotto Unique 3's
    Posted: April 08 2009 at 9:37pm
 
Record 289 Lines Cover 6/45 Lotto Unique 3's™
by Colin Fairbrother
 
Pick 6 from Pool 45 Lotto games are played in Australia and Singapore where the former pays with two bonus integers and the latter with one. You can obtain a set of numbers to play using this structured template from a random order of the 45 integers from LottoToWin for only a $5.00 annual subscription.
 
The odds for getting a Three win with a Bonus integer can be calculated from 1 in 45c6 / 6c3 x (39c3 - 37c3) which is 1 in 8145060 / 20 x 1369 or 1 in 297. The table below shows the 3 prizes which you need to get before getting the paying prize which must have the bonus ball. The odds for getting a Three only which is not paid on is roughly 1 in 45. To put that in perspective a good set of 22 structured numbers should average out to giving you a Three for every second draw played and the same set should give you a 3 + Bonus about every 14 or 15 draws. If you're not getting this then the explanation is that you are using stupid System plays such as System 8 which instead of using the 45 integers only uses 8 and is the second worse way to play! See my  Comparison of 15 Lotto Structured Number Sets.
 
Essentially this is a partial 3if3 Cover which from 1 to 289 combs is as good as any other, each comb contributing a coverage of 20 from the 14,190 possible Threes. A 289 line 3if3 set with any distinct Threes is going to give a close to 99.9% 3if6 coverage. However, if the Threes are chosen on the basis of maximizing progressively the coverage of the Sixes a proportionality is maintained especially up to around the 90% mark or 60 combs from where they taper off to about the 132 mark or 99% and from thereon it is practically a straight line. By contrast a 3if3 without maximization of the Sixes at 60 Combs may only give about a 64% coverage a difference of around 26%.
 
A distorted 3if6 Cover with repeat Threes can be done in 131 combs. The methodology in constructing this way is to force low yielding combinations into the set at the expense of higher yielding ones. For 131 Combs played one should expect per probability calculation close to 3 prize Three wins per draw. The grouping of the possible winning combinations for the set played should reflect this by not having the group which delivers 1 Three win as dominant as is the case with the distorted Cover but the group that delivers 3 wins as is the case with my set. Even after sorting the 131 comb set by highest dependency the top 45 combs only give a coverage of 49% compared to my set with 57% and similarly only 37% for the top 22 compared to mine at 46%. In fact for the groups delivering greater than 1 win my set is ahead at every stage. Obviously, the obsession with guaranteeing a Three win every draw is pure folly and to advocate it as delivering better on average than Random Selections or my Unique 3's set with optimized coverage is no less than fraudulent.   
 
 
 
 
 
Regards
Colin Fairbrother
IP IP Logged
Millsy
Registered Member
Registered Member


Joined: April 26 2009
Location: Australia
Online Status: Offline
Posts: 2
Quote Millsy Replybullet Posted: April 26 2009 at 7:43am
Hi Colin,
 
I hope your well and congratulations on a great site.
 
Would it be possible for you to post the 289 lines used for your record on this forum, or send them to my email address.
 
I would love to backtest this set against previous Australian Lottery draws.
 
It's a pity this site is not more active as you have alot to offer on this site.
 
Best Regards
 
Millsy
IP IP Logged
Colin F
Lotto Systems Tester Creator & Analyst
Lotto Systems Tester Creator & Analyst
Avatar
To dream the impossible dream ...

Joined: September 30 2004
Location: Australia
Online Status: Offline
Posts: 678
Quote Colin F Replybullet Posted: April 26 2009 at 8:47am
Originally posted by Millsy

Hi Colin,
 
I hope your well and congratulations on a great site.
 
Would it be possible for you to post the 289 lines used for your record on this forum, or send them to my email address.
 
I would love to backtest this set against previous Australian Lottery draws.
 
It's a pity this site is not more active as you have alot to offer on this site.
 
Best Regards
 
Millsy
 
Hi Millsy
 
Thanks for the comments.
 
This is a record which mainly is of interest to those who program in the area of Coverage. The analysis data provided where the set of numbers is tested against all possibilities is what people who are expert in this area are looking for. You don't see much response because most with the expertise have been in the past, if not now, involved in the creation of deliberately distorted Covers which may have a valid use outside the Lotto field, where the size of something to do something is more important than how wasteful it is. In a way I'm poking fun at them because they have been so stupid conceptually, even if clever in achieving their lower bounds in applying those distorted Covers to the Lotto field, where percentage return in the short term is the most important consideration. (First prize maximization is easily achieved by simply keeping all your lines different.) Generally, I think most of them have moved on after noting they didn't get that one right but there are still plenty of recalcitrants around who just hope I'll fade away. People that have been involved in something since the '80s are loathe to admit that they got it all wrong and that they wasted a huge chunk of their life.
 
Let's not forget the simple probability premise that it is all based on. If you have 10 identical balls in a box, 8 of which are red and 2 blue then after jumbling all the balls and picking one sight unseen, then 80% of the time you will get a red. That you got 10 reds consecutively does not make a blue more likely to come up in your next pick a very hard fact for Lotto history buffs to swallow. 
 
I have had my own VBA program since 2000 where I can run the full set or part of it over how many draws I want and get back performance data whether it be history or from my own RNG. It has a whole bunch of tricks including playing the same numbers or randomizing from a template. I authorize no one else to test my sets against history as it is too easy to pick the worst performing sets albeit it more difficult with my Unique 3's™ as an integral part of their design is their consistency and they perform very well as the analysis data indicates they should. In other words it is possible to produce sets with 1st or 2nd or 3rd prizes from the histories. Notice the "or" because with my sets no repeat Threes means there are no repeat Fours, Fives and Sixes. Sets with repeats can be manipulated to give multiple high prizes. Basically what I'm saying is the tests I do are honest but I can't vouch for those done elsewhere.
 
I'm not advocating that someone should spend $179.00 per draw playing the full 289 line set although I am aware of people that vainly and unfortunately spend $1,000 to $4,000 per month on Lotto. They mainly do system plays, the 2nd worst way of playing Lotto and because so and so got first prize playing say a System 12 they won't have a bar of anyone doubting its merits. My sets in the main are directed to people like myself who spend $11.00 per week or those that fork out 2 or 3 times that. However, I am open to approaches from bona fide syndicate organizers wanting to play the full set.
 
If you know how to test then you should have no problem in constructing a set of 50 to use as a template for randomizing (the first 7 lines can be done by just keeping all the digits different) - failing that then I'm always glad to receive your $5.00 for a 12 month subscription to LottoToWin.
 
If you have programming skills you can have a go at posting a record.
 
Regards
Colin Fairbrother
Lotto Draws have no relationship to one another; the integers serve just as identifiers. Any prediction calculation on one history of draws for a same type game is just as irrelevant as another.
IP IP Logged
Post Reply Post New Topic
Printable version Printable version

Forum Jump
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot delete your posts in this forum
You cannot edit your posts in this forum
You cannot create polls in this forum
You cannot vote in polls in this forum



This page was generated in 0.039 seconds.