Lotto mathematics as a time series

[taaroa]

Hi,

I'm interested in analysing the lotto as a time series. I must therefore have an idea of this time series caracteristics, in sorted order or unsorted order.

Is the lotto time series linear or non-linear? Is it the same for drawings in sorted order or unsorted order?

Others caracteristics are also necessary. Periodic or not, deterministic or not, chaotic or not, stochastic or not,etc...

Is there an adress where to find those informations about lotto?

I'll appreciate answers about those subjects.

Thank you.

taaroa

 

[Icewynd]

Interesting question, Taaroa. I haven't seen anything that addresses those isssues, although mathemeticians have written quite extensively on the lottery.

Have you checked out Ion Saliu (saliu.com) to see if he has anything relevent? Also, you might try your search in Google Scholar (click 'more' at the top left of the screen) to see if any scholarly papers or publications look into some of these issues.

Good luck with your search, and keep us posted on the results

 

[taaroa]

Interesting question, Taaroa. I haven't seen anything that addresses those isssues, although mathemeticians have written quite extensively on the lottery.

Have you checked out Ion Saliu (saliu.com) to see if he has anything relevent? Also, you might try your search in Google Scholar (click 'more' at the top left of the screen) to see if any scholarly papers or publications look into some of these issues.

Good luck with your search, and keep us posted on the results

Thank you.

After more 20 years of searching with the tradional analysis method like skips, mean, median, etc... my conclusion is that those methods are too simplistic and do not yield good results. We must go farther. I decided to search for others methods. I read many articles found on the web. The most used methods for analysis are linear or nonlinear regression and neural networks, all used in the stock market business. Some lotto players do use these programs. But, what I learned is that you must model your time series and know his caracteristics before choosing the correct tool. This why I asked my questions in my post.

Till now, I have used all methods available. Saliu's programs are good ones, based principally on skips and median. But none of these methods give good results i.e better than probability. Maybe I do not used them the right way, who knows?

The problem for me is that I don't have enough mathematic knowledge to clearly understand the concepts lying behind time series analysis and neural network. Understanding the lotto model will help me a lot. I'm learning those mathematics but don't know how to find if lotto is a linear or a nonlinear time series depending you use it in sorted order or unsorted order.

I hope some mathematic guy will see this post and give some help about it. I'll post on mathematics to see if I'll got some help there. By the way, I didn't saw many articles on lotto by mathematicians. Just some basic statistics, nothing else. I know that most of them think that you cannot predict lotto. Maybe they are right, maybe not. I have the intuition that if you use and combinate some methods, you can improve your results in a significative way. This is I begin to do. It's a long way......

taaroa

 

[bloubul]

Hi taaroa

For starters I'm not a mathematic guy, but find my on blitzed web page and I can give you Robert Perkins email address, maybe just maybe he can help.
You must include your BB name in your request please.

BlouBul

 

[Gonga]

Lotto Time-Line

E.A.L has a plugin that plots each Lotto Ball on a Time-Line graph.
http://www.slx.za.net/rtfm/eal/plugins/timeline.html

Regards
N.G van der Westhuizen

 

[taaroa]

Hi taaroa

For starters I'm not a mathematic guy, but find my on blitzed web page and I can give you Robert Perkins email address, maybe just maybe he can help.
You must include your BB name in your request please.

BlouBul

Thanks.

I think I can contact him directly. I have bought his Winyourlottery soft.

Thanks again.

taaroa

 

[taaroa]

E.A.L has a plugin that plots each Lotto Ball on a Time-Line graph.
http://www.slx.za.net/rtfm/eal/plugins/timeline.html

Regards
N.G van der Westhuizen

I do the job with Excel. Can also fit the graph to get the polynomial function.
I think that there is not an easy way to find the caracteristics of the lotto matrix. You must apply some formulas and get some statistics. I have calculated parameters like variance and covariance, standard deviation, maximum lyapunov exponent. I want to know if I take the sorted lotto drawings history, that lotto can be considered as a linear system or not. Same thing for non sorted drawings. As I found a positive lyapunov exponent around 2, lotto can be considered as a chaotic system, I think. I also consider that lotto is a stationary system as variance and mean doesn't change. Now, I don't master these elements and need confirmation.

Thanks.

taaroa

 

[Icewynd]

Yes, I would agree that the lottery fits the characteristics of a chaotic system. Linear methods explain a very small amount of the variance.

I saw something recently about lottery and chaos theory -- if I find it again I'll post it here. A google search for 'lottery chaos fractals' turns up some very interesting links.

Sounds like you are making some progress with your investigations.

 

[CMF]

Lotto: A stimulant for the brain if you treat it right

Some who comment intelligently on Lotto never actually play the game, others do and I fall into the latter category. Every Saturday I make my voluntary $11.00 contribution to the government coffers and get back on average a bit under 30%.

While the majority of players who go to the trouble of considering what numbers to play do so by referencing the recent history the basis for doing so is nothing more than intuition. While not wishing to denigrate the beneficial effects of exercising one's brain little more than that is obtained.

The 18 lines I play are based solely upon the following three principles which I have written extensively on: -

1 Use all of the Pool of integers.
2 Keep your paying subsets unique.
3 Have the Coverage as good as possible without going over the top.​

Generally, any mathematician with some claim to authority, rightfully dismisses any consideration of history simply because a Lotto draw is an independent event and anything you perceive as beneficial is pure coincidence.

I have taken the topic further and investigated every nook and cranny for something beneficial. Recently, I wrote an article titled:
Analysis of Lotto Draw History - the Final Word
which you can Google if interested.

Colin Fairbrother

 

[garyo1954]

Colin,

Although we both agree that each draw is an independent event, I would add that is in theory. Yes, all theory is true until proven otherwise. Then it becomes basically a rule.

Consider the case of tossing a coin. If we toss a coin and it lands "HEADS" six consecutive times, the Law of Probability tells us we can expect "TAILS" soon.

Consider industrial accident insurance. For every X number of minor accidents the insurance industry expects a recordable accident. For every X number of recordable accidents, they expect a reportable accident. And for every X number of reportable accidents, they forecast a major accident.

Naturally, these don't always align in the manner the insurance industry forecasts, but over time they work out.

Therefore if the only history we have to go by is the sequence of draws, then we shouldn't ignore it. After all, how many times in your lifetime have you heard scientists, astronomers, or NASA spokespersons say, "When the information we have available, our best guess is......?"

G

 

[CMF]

Agreeing on the Law of Independent Events but ...

Gary

Read my article - it is not difficult to find if you Google
Analysis of Lotto Draw History - the Final Word

In the article I conclusively prove that the numbers when given the extra properties of absence and occurrence are demonstratively inferior to that of using the numbers alone. I give the facts and figures to support that ie I prove that it is not beneficial to attribute absence and occurrence to the numbers.

You seem to confuse a hypothesis with a theory. A theory is supported if when applied to some relevant scenario it is not contradictory. The hypothesis you are supporting is that there is a beneficial relationship between the history of draws for a particular Lotto game where the intention of the draw is for it to be applicable to that game. Give me one bit of evidence to support that.

The Laws of Probability regarding coin tossing tell us that after one million tosses the independent event of tossing a fair coin still has the same probability of 0.5 or 50/50 or evens. You may do what is termed Bernoulli Trials which are independent repetitions and you will find that generally it takes up to 10 tosses before a Head or a Tail comes up depending on what you regard as favorable at the time. Following those 9 Tails or Heads the likelihood of a Head or Tail following is still 50/50.

Your post basically confuses statistical data upon which aggregates may be counted or summed with simple independent events such as coin tossing where the event of tossing a fair coin is not dependent on what has occurred with previous tosses ie the occurrence of one of the two possibilities does not effect the likelihood of the other occurring.

As I mention in my article most academics dismiss the study of Lotto history as being quite correctly irrelevant because a Lotto draw is an Independent Event. Over the past 10 years I have gone the extra distance applying my database and programming expertise to investigate just about every possibility that has a whiff of "credibility".

You need to come up with something better than a misguided understanding of probability theory to refute the irrelevance of Lotto history in determining the numbers in the next draw for a particular Lotto game.

Colin Fairbrother

 

[garyo1954]

Kudos, Colin!

"You're to be congratulated for thinking Lotto history analysis as a way to beneficially produce numbers to play is irrelevant and is a gigantic con played out by closet or brazen numerologists or despicable opportunists prostituting their integrity for a few lousy bucks and all using the usual trick of assuming something false is true by pandering to something intuitive but incorrect and then constructing a dung heap on thin air."

Yes, I did read your article. And follow the link to your site which asks for $.

One question: Can you present an inverted Delta Graph (Bell Curve) based on sums, from anywhere in the world? One in which the mid ranges show the lowest totals with the outer edges trending upwards?

Yes, I do appreciate your hard work. And I do AGREE with much of what you present, given the context in which it is presented. As an exchange of ideas it is exemplary.

 

[CMF]

The Irrelevant Sums Chart in Lotto

Gary

Click on my name as above and select all posts by me in this forum. Sums chart was discussed here January, 2007.

Colin Fairbrother
ps There is no fee for reading what is made available on my website. A program I make available online at a different site, which produces numbers to play for most lotteries around the world, has a nominal $5 subscription, which would hardly cover the time spent processing the registration. The program does not use any history of draws to arrive at the numbers.

 

[garyo1954]

Colin,

You bring a lot of good discussion to this board. Some of it ruffles my feathers when you appear so adamantly against the possibility of using available information which may, or may not, bring about a valid conclusion.

Although it is probably impossible for me to elicit the same response in you, personally, I like that you make me think.

 

[CMF]

Gary

If you read and followed the article I referred you to you would have realized I have logically and conclusively shown that no benefit is obtained by using absence or recency and occurrence or frequency with the integers. There is no mathematician, statistician or analyst of merit that believes there is a connection or relationship between the Lotto game history and the next draw - I have gone a step further and proved it!

Will that stop people from using such methods? Simply No! People will use whatever they like to delude or amuse themselves, including throwing bones etc and never let a bit of rational thought get in the way.

However, there are thoughtful people including academics, that are interested in this area of interest - producing sets to play in Lotto to produce a marginal better yield or more dubiously a guaranteed however small prize for whatever outlay.

There are plenty of challenges still in this field - it's just a matter of people moving on as old ideas get debunked or discredited.

Colin Fairbrother