mid-Ohio United States Member #9 March 24, 2001 19825 Posts Offline

Posted: August 5, 2015, 2:33 pm - IP Logged

Quote: Originally posted by dr san on August 5, 2015

Hello, RJOH, we have a problem !!!, the law of benford, does not seem to work very well in lotteries Because a given period of sweepstakes, seeks to balance the output of the digits of each number the digits 0 through 9. In lotteries seems be more balanced, that the law of benford, benford the law on accounting data to have very beginning of a series, is different from lotteries. In lotteries there is balance, so we have to look at rates and average, by position, then the average will command

I don't have a problem, I simply looked at the results of Ohio's Classic Lotto drawings as I recorded them with the lowest number first and the highest number last and posted the results as compared with the benford's law. Make of the results what you want.

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

mid-Ohio United States Member #9 March 24, 2001 19825 Posts Offline

Posted: August 5, 2015, 2:41 pm - IP Logged

Quote: Originally posted by CARBOB on August 5, 2015

RJOH,

Do you have the draws in draw order? Would the percentage be the same?

No I don't, I simply record drawings results as they are posted on the websites. Since these results are based on the leading digits, changing the leading digits would probably change the results unless lower numbers are drawn first more often. Besides as dr san points out the benford's law was meant for accounting data not lottery results.

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

ORLANDO, FLORIDA United States Member #4924 June 3, 2004 5896 Posts Offline

Posted: August 5, 2015, 3:08 pm - IP Logged

Quote: Originally posted by RJOh on August 5, 2015

No I don't, I simply record drawings results as they are posted on the websites. Since these results are based on the leading digits, changing the leading digits would probably change the results unless lower numbers are drawn first more often. Besides as dr san points out the benford's law was meant for accounting data not lottery results.

Thanks. here is a link to one of Thoth's post about Benford and the Lottery.

bgonÃ§alves Brasil Member #92564 June 9, 2010 2122 Posts Offline

Posted: August 5, 2015, 3:36 pm - IP Logged

Hello, seems to be a tendency to lower the digits can have higher output, ainde should be balanced, good can can create wheels upon that of 0,1,2,3-digit trend (higher output,, but the average as is?

bgonÃ§alves Brasil Member #92564 June 9, 2010 2122 Posts Offline

Posted: August 5, 2015, 3:46 pm - IP Logged

Note that there are various types of random numbers. If a number drawn eh randomly in [10,100) (with uniform probability distribution) then the first digit has the same chance to be 1, 2 or 7, with 1/9 for each.

Now, the probability distribution of the numbers that appear in various not random events eh THIS way. If the numbers have a density probability that the C / x type (where x varies continuously from 1 to 100, say), then:

a) Integral C / x x = 1 to x = 100 has to 1; with it, you discover where C = 1 / (ln (100) -ln (1)) = 1 / ln (100) b) The probability that the first digit is 1 h the full C / x of 1 to 2, more the full 10 to 20, which Dah (Ln (20) -ln (10)) / ln (100) + (ln (2) -ln (1)) / log (100) = log (2) (here, h log base 10); analogously, the probability that the first digit is 6 will be (Ln (7) -ln (6) + ln (70) -ln (60)) / log (100) = log (7) -log (6). And so on, the lower digits are less common, and the distribution is the difference between successive logs.

The same distribution appears if you assume that x varies continuously 10 ^ 10 ^ m n with probability density type C / x. Ta, the matter did not eh only that (for example, my select range of 10 m to 10 ^ h ^ n means specifies; but if x vary from <10 ^ m <n ^ 10 <b, with several orders of magnitude between a and b, then the "tails" of probability between a and 10 ^ 10 ^ ne me between b should not greatly affect things; the minimum value for a = 1, things improve a lot!), but I think the spirit is contained in this that I wrote. In short, if the numbers that you are looking at has a distribution that is higher numbers are rare (more or less proportional to density 1 / x) and has at least several orders of magnitude possible in the data, then the first digit satisfy the distribution of Benford. Eh why histograms with lengths of rivers and populations of cities must first digit fulfilling this distribution, but the first digit numbers not lottery (lotteries, there is no preference for smaller numbers, all are equally likely).

bgonÃ§alves Brasil Member #92564 June 9, 2010 2122 Posts Offline

Posted: August 5, 2015, 4:43 pm - IP Logged

Hello, rojh, then we can use him to piramede odds And do not use the 0.1-digit, nor 8.9 but use the center lane 0-9 = 2,3,4,5,6,7, these 6 digits plants It seems to appear in greater numbers on 49/6 when separated by initial and final digit Ie the medium or the center of the line 0-9 = the digits 2,3,4,5,6,7, always seem vain In larger numbers (sums) that 0.1 and 8.9 in lottery

mid-Ohio United States Member #9 March 24, 2001 19825 Posts Offline

Posted: August 6, 2015, 6:41 am - IP Logged

Quote: Originally posted by RJOh on August 5, 2015

Using the last 1336 combinations of Ohio Classic Lotto, according to WikipediA the benford's law suggests the leading digits should be distributed in the following way:

bgonÃ§alves Brasil Member #92564 June 9, 2010 2122 Posts Offline

Posted: August 6, 2015, 7:33 am - IP Logged

Hello, roll, smaller intervals, eg 10, 20 or sweepstakes? The problem is how the numbers come out more and less leave Mathematically may or may not help in predicting when it comes to unique event

mid-Ohio United States Member #9 March 24, 2001 19825 Posts Offline

Posted: August 6, 2015, 10:42 am - IP Logged

Quote: Originally posted by dr san on August 6, 2015

Hello, roll, smaller intervals, eg 10, 20 or sweepstakes? The problem is how the numbers come out more and less leave Mathematically may or may not help in predicting when it comes to unique event

In the link CARBOB posted, Thoth did a pretty good job of explaining the limitations of benford's law in predicting lottery results. Never the less I thought the distribution their drawing results by the leading digit was interesting.

I didn't remember him but a check of some of his threads showed I responded to some of his posts.

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

Dallas, Texas United States Member #4549 May 2, 2004 1689 Posts Offline

Posted: August 6, 2015, 4:46 pm - IP Logged

Quote: Originally posted by RJOh on August 5, 2015

Actually I've been looking for something similar to the benford's law concerning the formations of lottery combinations that suggest that certain formations are more likely than than others. I think RL with his lexi program and Jadelottery with his three sum math are looking for something similar.

This wouldn't prove that random isn't so random but that random doesn't eliminate the natural laws of orders.

I'm interesting in finding SOMETHING; a concept, idea, law, principle, whatever you want to call that can be used or coupled with math to produce a reasonable possibility of in some way lowering the odds, evening the playing field, or whatever yuo want to call that.

I don't believe there will ever be a mathematical formula to produce the winning combination. But I do believe math will play a major roll in formulating that process just as it does in researching what the numbers are doing.

The NFL has it's QB rating system (which nobody understands) but everybody goes WOW! when a QB has 113.6 day.

And back in the day our number might be named Mantle (or 7) and the stats would say .305/23/74. Everybody understood that was batting average, home runs, and total RBIs. And they say Wow! He's having a good year.

Today we see .339/.410/.749 and people say Ohhhhhh-kay, whatevah. Then somebody mentions VORP.

Just saying baseball stats have exploded to the far corners over the years.

Maybe we need a VOPP for lottery numbers. It sounds Star Trekkian, scientific, and lotteryish all at the same time. I can see a post "The VOPP on 9 tonight is .989/.764/.418."

Yes, some numbers/digits/pairs/combinations do show better in certain positions. A set of 3 or more digits will show a tendency to come out more in one way than another, say for instance, 1,2,3,4 may show more times as 3,1,2,4 than 2,4,3,1. To get an idea why this is, we need to go back and look at how combinations fall in the matrix.

After all that we can develop our VOPP (Value on Placement Percentage) to reflect the percentage of hitting each position.

My greatest accomplishment is teaching cats about Vienna Sausage. When I need a friend, all I need do is walk outside, pop open a can, and every little critter in the neighborhood drops by to say "Hi!"

mid-Ohio United States Member #9 March 24, 2001 19825 Posts Offline

Posted: August 7, 2015, 7:01 am - IP Logged

Quote: Originally posted by garyo1954 on August 6, 2015

I'm interesting in finding SOMETHING; a concept, idea, law, principle, whatever you want to call that can be used or coupled with math to produce a reasonable possibility of in some way lowering the odds, evening the playing field, or whatever yuo want to call that.

I don't believe there will ever be a mathematical formula to produce the winning combination. But I do believe math will play a major roll in formulating that process just as it does in researching what the numbers are doing.

The NFL has it's QB rating system (which nobody understands) but everybody goes WOW! when a QB has 113.6 day.

And back in the day our number might be named Mantle (or 7) and the stats would say .305/23/74. Everybody understood that was batting average, home runs, and total RBIs. And they say Wow! He's having a good year.

Today we see .339/.410/.749 and people say Ohhhhhh-kay, whatevah. Then somebody mentions VORP.

Just saying baseball stats have exploded to the far corners over the years.

Maybe we need a VOPP for lottery numbers. It sounds Star Trekkian, scientific, and lotteryish all at the same time. I can see a post "The VOPP on 9 tonight is .989/.764/.418."

Yes, some numbers/digits/pairs/combinations do show better in certain positions. A set of 3 or more digits will show a tendency to come out more in one way than another, say for instance, 1,2,3,4 may show more times as 3,1,2,4 than 2,4,3,1. To get an idea why this is, we need to go back and look at how combinations fall in the matrix.

After all that we can develop our VOPP (Value on Placement Percentage) to reflect the percentage of hitting each position.

"I don't believe there will ever be a mathematical formula to produce the winning combination."

Neither do I, like you I think my best hope is to find a system that reduces my odds of having all the winning numbers in a single line. That could be done by reducing the number pool, limiting particular numbers to unique positions in combinations along with other limiting parameters. Right now I'm looking at which numbers followed numbers in previous drawings which has been looked at by others in the past but seems to be showing some decent results in the local 6/49 games which has over 3700 drawings to check. My problem is when it's right on I know of no way to calculated the number of the possible combinations. If I reduce my odds of winning from one in fourteen millions to less than one in a million most of the time then I will have accomplished my goal.

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

Texas United States Member #55889 October 23, 2007 5595 Posts Offline

Posted: August 7, 2015, 8:45 am - IP Logged

I used to keep charts of how many times each number would play in each position, and then I could tell what position each decade of numbers played in each position. I lost all that data when my computer caught a virus. The problem I had with the data is that many times the draws wouldn't cooperate. LOL. Most of the time a single plays in the first position, a teen in the second. Maybe a 20 in the third, etc. Well I would play a format like that and a 20 would play in the second position, and a 40 in the third.

Once I lost my data, I said screw it, it wasn't working that well anyway.

mid-Ohio United States Member #9 March 24, 2001 19825 Posts Offline

Posted: August 7, 2015, 10:25 am - IP Logged

Quote: Originally posted by rcbbuckeye on August 7, 2015

I used to keep charts of how many times each number would play in each position, and then I could tell what position each decade of numbers played in each position. I lost all that data when my computer caught a virus. The problem I had with the data is that many times the draws wouldn't cooperate. LOL. Most of the time a single plays in the first position, a teen in the second. Maybe a 20 in the third, etc. Well I would play a format like that and a 20 would play in the second position, and a 40 in the third.

Once I lost my data, I said screw it, it wasn't working that well anyway.

If you keep updated files of the games you play, you can generate new and updated charts in a couple of seconds with your computer every time you're thinking about playing. I've taken to backing up my programs and lottery files to a flash drive once in a while so if my computer goes out I simple put the drive in another computer, go to a website and update my files from the last back-up and continue. The one time that happened to me I only lost my old play files which I seldom looked at any way.

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

mid-Ohio United States Member #9 March 24, 2001 19825 Posts Offline

Posted: August 8, 2015, 1:24 pm - IP Logged

I've noticed that combinations drawn in Ohio's Classic Lotto(6/49) have done better in past drawings than the combinations I pick so I'm going to start picking combinations that have averaged at least 10¢ a draw overall in previous drawings while still staying within the parameters I usually use.

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

Kentucky United States Member #32652 February 14, 2006 7309 Posts Offline

Posted: August 9, 2015, 1:05 am - IP Logged

Quote: Originally posted by garyo1954 on August 6, 2015

I'm interesting in finding SOMETHING; a concept, idea, law, principle, whatever you want to call that can be used or coupled with math to produce a reasonable possibility of in some way lowering the odds, evening the playing field, or whatever yuo want to call that.

I don't believe there will ever be a mathematical formula to produce the winning combination. But I do believe math will play a major roll in formulating that process just as it does in researching what the numbers are doing.

The NFL has it's QB rating system (which nobody understands) but everybody goes WOW! when a QB has 113.6 day.

And back in the day our number might be named Mantle (or 7) and the stats would say .305/23/74. Everybody understood that was batting average, home runs, and total RBIs. And they say Wow! He's having a good year.

Today we see .339/.410/.749 and people say Ohhhhhh-kay, whatevah. Then somebody mentions VORP.

Just saying baseball stats have exploded to the far corners over the years.

Maybe we need a VOPP for lottery numbers. It sounds Star Trekkian, scientific, and lotteryish all at the same time. I can see a post "The VOPP on 9 tonight is .989/.764/.418."

Yes, some numbers/digits/pairs/combinations do show better in certain positions. A set of 3 or more digits will show a tendency to come out more in one way than another, say for instance, 1,2,3,4 may show more times as 3,1,2,4 than 2,4,3,1. To get an idea why this is, we need to go back and look at how combinations fall in the matrix.

After all that we can develop our VOPP (Value on Placement Percentage) to reflect the percentage of hitting each position.

"I don't believe there will ever be a mathematical formula to produce the winning combination."

but a couple of years ago we had an interesting discussion on using recently drawn numbers that produced 5 + 0 MM matches and a jackpot (on paper). It's probably just one of those weird coincidences, but MM changed the matrix shortly afterwards.