SUMS

Reply #56

hypersoniq — Sat, 03 Apr 2004 06:29:01 GMT

Unless, of course, you use sums as some sort of final inspection of a pick, sort of like Quality Control

Reply #55

lottoarchitect — Fri, 02 Apr 2004 23:18:41 GMT

About SUMS: Suppose we have a game 5 of 45. A draw then is A,B,C,D,E. Total sums is then A+B+C+D+E=SUM. This value depends on 5 different variables (which may be related dvents but we don't have a way to determine this). Thus trying to predict the sum of a draw, we face the problem of analysing the participation effect of 5 distinct numbers in this example. This makes the whole attempt very difficult. As I always say, try other filters first that are more steady than sums. Differences (A-B/B-C e... [ More ]

Reply #54

Andrew — Mon, 29 Mar 2004 16:54:55 GMT

Regards SUMS,The most common sums will be drawn more often, as compared to the least common sums (0 and 27) for pick 3.

Reply #53

Andrew — Mon, 29 Mar 2004 16:45:22 GMT

Quote: Originally posted by apagogeas on March 26, 2004I believe RNG is useless to predict lottery games and I cannot understand why people keep using it. Indeed, previous history draws of a lottery game is the best source of random numbers (supposing that lottery games are random).Another problem of using the RNG (Or PNRG for that matter) is that you cannot re-run a previous test. There is no way to prove or substantiate your claims of success, because once you ahve a successful prediction one... [ More ]

Reply #52

Clipper — Mon, 29 Mar 2004 03:14:58 GMT

Regards SUMS.....I have used them for years in my Pick 3 tracking and they work quit nicely. After an analysis of a years data in Texas, it was clear that combinations that had Sum Totals 11-20 played 2 1/2 times more often the combinations that Sum 0-10. Also that Combinations that had Sum Total 21-27 played only 1/3 as much as combinations that Summed 0-10.I therefore think there is a clear advantage in considering Sum Totals as part of the process for determination of what to play. You can... [ More ]

Reply #51

lottoarchitect — Fri, 26 Mar 2004 15:12:48 GMT

Quote: Originally posted by Andrew on March 22, 2004Quote: Originally posted by hypersoniq on March 21, 2004Regardless of how you slice, dice, optimize and call RND or set multiple timers, the PRNG built into Access/Excel/Vb is a pre-determined list...A point Colin failed to point out. Also notice, he didn't post his 15 lines of code he claims to have written. Makes you wonder.....You are right Adrew, no matter how you use the RNG function, it remains deterministic so it is absolutely inadeq... [ More ]

Reply #50

Andrew — Fri, 26 Mar 2004 11:07:22 GMT

Quote: Originally posted by Colin F on March 22, 2004Not much to do with a predetermined list there old son. A lot to do with getting your facts right and having egg on your face.You missed his point Colin. He was simply pointing out that the RND function uses the SAME predetermined list. Randomize how ever much you want, it is still the same list. With that in mind, it is just a Pseudo-Random Number Generator.Hmmm... Predetermined? I don't see random there.

Reply #49

plumbusa — Thu, 25 Mar 2004 05:08:32 GMT

Win The Lotto

Reply #48

Colin F — Thu, 25 Mar 2004 04:46:54 GMT

Hypersoniq and AndrewIf a person came on to this forum and in his/her first Post they proved me wrong on a point I would concede that point and learn from it. This is a concept that generally civilized people and gentlemen or gentlewomen adhere to.It is a concept that is totally foreign to both of you. Both of you were done like a dinner but any acknowledgment, no. As I am a person of mature age with principles that are obviously at odds with yours I have to ask myself are you both people that I... [ More ]

Reply #47

hypersoniq — Wed, 24 Mar 2004 22:50:11 GMT

Who said anything about multiplying by 49 and trimming with INT?....you can run the same experiment with JUST the decimal values... if you put a number into the seed instead of the TIMER argument, you will get a list of numbers, if you put the same number in, you will get the same list of numbers...Again, it's irrelevant because no RNG or PRNG is capable of modelling the type of entropy created by the ping-pong balls...

Reply #46

Colin F — Mon, 22 Mar 2004 23:56:16 GMT

The use of Rnd(-something) and Randomize to use the System Timer was assumed. Nobody has been talking about anything else. If you then multiply the usually 7 digit decimal number produced say .081755 by say 49 you get 4.28995. If you then use the Int function to strip away the decimal part you get 4 and so on.Not much to do with a predetermined list there old son. A lot to do with getting your facts right and having egg on your face.The trap is that with the speed of modern computers the process... [ More ]

Reply #45

hypersoniq — Mon, 22 Mar 2004 22:17:53 GMT

The list is fixed, ask Bill Gates why...Try a call to RND(-1)then RANDOMIZE(5)then loop a call to RND unbounded 10 or so timesprint the variables to a file...run it again, same output.TIMER is used to pick seemingly random entry points into the list, if you don't use the TIMER word explicitly, RND has nothing to do with the system timer.don't take my word for it...http://www.vbexplorer.com/VBExplorer/random/random_numbers_1.asp

Reply #44

Andrew — Mon, 22 Mar 2004 13:38:30 GMT

Quote: Originally posted by hypersoniq on March 21, 2004Regardless of how you slice, dice, optimize and call RND or set multiple timers, the PRNG built into Access/Excel/Vb is a pre-determined list...A point Colin failed to point out. Also notice, he didn't post his 15 lines of code he claims to have written. Makes you wonder.....

Reply #43

hypersoniq — Sun, 21 Mar 2004 14:47:47 GMT

In the end, Andrew is right...Regardless of how you slice, dice, optimize and call RND or set multiple timers, the PRNG built into Access/Excel/Vb is a pre-determined list...The only way to get random numbers when using the MS apps is to call a 3rd party .DLL that gets a source of entropy from somewhere else, or a call to some outside source for random numbers like random.org (if you believe the premise that atmospheric noise is random to begin with). In the end, either attempt is futile because... [ More ]

Reply #42

tdempsey — Fri, 19 Mar 2004 19:35:53 GMT

Back to sums...I realized in the beginning that all lotto draws have a discreet sum and to me that is one, if not the most important, value to building a system.I always play draws based on sums to help distribute the sum range. If the sum of the next draw is 160 and I played all my sums less than 155, I would never hit the jackpot.I also base several other rules on the sum.Tom Dempsey

Reply #41

Andrew — Tue, 16 Mar 2004 11:50:18 GMT

BTW,Here is a simple VB Random number generator that should suffice. (snipped from http://www.codetoad.com/vb_random.asp)Public Function RandomNumber(Upper As Integer, _ Lower As Integer) As Integer 'Generates a Random Number BETWEEN the LOWER and UPPER values Randomize RandomNumber = Int((Upper - Lower + 1) * Rnd + Lower)End FunctionInterestingly enough, even the programmer that I linked to said this:This function will generate an ARRAY of TRULY random numbers. By default they are... [ More ]

Reply #40

Andrew — Tue, 16 Mar 2004 11:39:27 GMT

Colin F,I will address these issues, and I expect straight answers for each. No dodging the issues here buddy.Quote: Originally posted by Colin F on March 11, 2004At first glance I don't think youv'e made enough calls to the Rnd function.And just how many calls to RND are enough?Isn't RND random enough with one call?Without the calls to Randomize , RND will always return the SAME sequence of random numbers test per test. With this in mind, wouldn't the more true response to #1 above be More... [ More ]

Reply #39

Colin F — Fri, 12 Mar 2004 10:35:41 GMT

AndrewI've looked at your code. It's fixable but it's also commercial,(Invoice Numbers)- nothing to do with Lotto. You've drawn the wrongconclusions and your calculations for either Permutations orCombinations are both wrong.A better idea would be for you to post your code (your the open sourceman, not me) for producing Random Numbers and I or others will getit working nicely. That way everyone can use it and there will be somestandardisation.Colin

Reply #38

Colin F — Fri, 12 Mar 2004 03:36:59 GMT

To Allhttp://bulletin.ninemsn.com.au/bulletin/EdDesk.nsf/All/922B3394142CB2A1CA256E3F001711DCI couldn't resist letting you know about this. I've been a thong wearer from waybackand still insisted on wearing them when others were getting around in 2 ton sneakersand it definitely was not de rigour. But I have to let you know that I never wore Ugboots and I just can't understand how any self respecting American or Canadian wouldbe seen dead in them let alone think of them as a fashon item. They are... [ More ]

Reply #37

Colin F — Thu, 11 Mar 2004 23:45:52 GMT

To AllI've had a quick look at Andrew's Code and even though I think it is more appropiateto an Access or VB forum I will comment later. At first glance I don't think youv'e madeenough calls to the Rnd function. If anyone is chaffing at the bit to have a go be my guest.Don't forget Andrew ... If it walks like a Duck, If it quacks like a Duck, If it paddles like a Duck, If it has feathers like a Duck, If it has webbed feet like a D... [ More ]

Reply #36

Andrew — Thu, 11 Mar 2004 11:28:45 GMT

Colin,Your missing my main point. There is a vast difference between pseudo-random and pure random.http://random.mat.sbg.ac.at/~ste/dipl/node10.htmlBUt I will leave that argument up to the math heads. My point was given the fact that your RNG is actually a PRNG, and your prediction engine was performing better than random expectation, then you can deduce that your engine appears to indicate that any lottery tested (which is assumed to be purely random) is similar to a PRNG. You even state tha... [ More ]

Reply #35

Colin F — Tue, 09 Mar 2004 12:24:34 GMT

AndrewYou keep going over points that I've already answered. It doesn't matter whether I use my RNG or a Lotto History; the results are pretty much the same.Now all this talk about atmospheric noise etc it's all baloney as far as I am concerned. We know the method that is accepted to pick Lotto Numbers by applying compressed air to balls in a confined space to make them dance around in a haphazard way then extract one. Then for the balance of the balls we make them dance around then extract one.... [ More ]

Reply #34

Andrew — Tue, 09 Mar 2004 11:12:45 GMT

Quote: Originally posted by Colin F on March 09, 2004 Anyway, why are y'all picking on my RNG for; how about Todd with his Quick Picks? I read in a thread where he mentioned some of these problems to the Chief.Think of it this way... Computer's can't produce random numbers on their own. So, you are using a pseudo-random number generator.Correct me if I am wrong, but you have optimized your methods based upon this pseudo-RNG. Since you are using a pseudo-Random generated history for testing,... [ More ]

Reply #33

Colin F — Tue, 09 Mar 2004 09:23:14 GMT

I'll stick with being a realist which means Murphy was a pessimist and O'Toole was the optimist and in any case they were both morbid. Always look on the bright side of Life De Dum De Doo Always look on the bright side of life De Dum De Doo Colin

Reply #32

johnph77 — Tue, 09 Mar 2004 08:00:11 GMT

O'Toole's Commentary on Murphy's Law - Murphy was an optimist.<g>gl

Reply #31

Colin F — Tue, 09 Mar 2004 05:10:15 GMT

Bob PThe point your making is not unknown to me and as explained to Andrew I've put a Random Timer between Creating Numbers for Draws and Creating Numbers to play.Without doing this it is possible to get a coincidence in the run through times where there are no tryagains such that the decimal setting of the starting time coincides and if then there are no tryagains you can have a six come up. This is eliminated by putting a variable timer between them.The other point I would make is that if ther... [ More ]

Reply #30

BobP — Tue, 09 Mar 2004 04:27:45 GMT

Quote: Originally posted by Colin F on March 08, 2004 Johnph77You're quite right. I was mixing up two methods of measuring. I work with typical payouts and thats the figure I usually monitor for the Percentage Return over Cost. So for A++ I'm looking to go from 37% to 46%.I'm beginning to conclude that not much can be done to improve the 3's and the 4's its all with the 5's. Now we are only talking 1 or 2 or 3 here but if one can consistently get 2 fives and the occasional 3 then in terms of pri... [ More ]

Reply #29

Colin F — Mon, 08 Mar 2004 19:20:52 GMT

AndrewLook at the results; they conform to probability expectations.I have previously said that I test using a RNG and then a Lotto History. They work out pretty much the same. I've even run a random selection from a Lotto History against another Lotto History.I'm not saying that there aren't pitfalls in programming especially when your running created Random Numbers against other created Random Numbers with no calculation in between and thats why I put a random timer there.RegardsColin... [ More ]

Reply #28

Andrew — Mon, 08 Mar 2004 11:56:35 GMT

Colin,Will you post your RNG code here? Im wondering how RNG'ish your code really is. Given the fact that your using Access, the validity of your random sequences comes into question. In my own testing of RNG on a PC, I am finding that the RNG is not as random as you might expect. Read this essay on computer RNGs:http://www.random.org/essay.htmlPerhaps in order to prove the validity of your RNG, you should use one of the methods discussed here:http://www.fourmilab.ch/random/There is even... [ More ]

Reply #27

Colin F — Mon, 08 Mar 2004 10:38:20 GMT

Johnph77It still won't make sense unless you take into account that Lotto Corporations are a bit more generous in Australia; House Margin 40%. Tax free too.RegardsColin

Reply #26

Colin F — Mon, 08 Mar 2004 10:12:17 GMT

Johnph77You're quite right. I was mixing up two methods of measuring. I work with typical payouts and thats the figure I usually monitor for the Percentage Return over Cost. So for A++ I'm looking to go from 37% to 46%.I'm beginning to conclude that not much can be done to improve the 3's and the 4's its all with the 5's. Now we are only talking 1 or 2 or 3 here but if one can consistently get 2 fives and the occasional 3 then in terms of prize money there is about 25% improvement. My own system... [ More ]

Reply #25

Colin F — Mon, 08 Mar 2004 09:48:16 GMT

Bob PYou know I like this. All the baloney Malone that sometimes gets thrown around on this Forum but here we are talking about the real thing. This is the nitty gritty stuff. The big prizes in lotto are down to luck. The smaller ones you can do something with but in the long term its only marginal compared to the money you've forked out. So the satisfaction has to come from just getting that slight improvement in the odds by doing clever things. In short it's a brain exercise and if you keep ex... [ More ]

Reply #24

johnph77 — Mon, 08 Mar 2004 05:41:53 GMT

Colin F -Thank you for your kind words - when the swelling in my head goes down far enough where I can fit it through the office door once again I'll go back to work on the site.As far as the grading of systems goes, commercial or not, I'm of the opinion that your ratings might be overly generous. Most systems, particularly those which are sold for profit, claim to be capable of generating positive returns, if not assuring one that they can hit the jackpot. But, in doing the work on my site, I'v... [ More ]

Reply #23

BobP — Mon, 08 Mar 2004 05:41:44 GMT

Quote: Originally posted by Colin F on March 08, 2004 BobPI try not to express things in decimal even though as you know Probability is defined as being between 0 (impossible) and 1 (certain) because it can be a turn-off for some people who otherwise would be interested. If you look at figures for Expected per Probability Formulae you see 54 games/ Win or approximately 2 Wins per Draw or 1 Win per 50 Games. We are playing 100 games per Draw.Max Games Dry is the maximum number of games that occur... [ More ]

Reply #22

Colin F — Mon, 08 Mar 2004 03:42:26 GMT

BobPI try not to express things in decimal even though as you know Probability is defined as being between 0 (impossible) and 1 (certain) because it can be a turn-off for some people who otherwise would be interested. If you look at figures for Expected per Probability Formulae you see 54 games/ Win or approximately 2 Wins per Draw or 1 Win per 50 Games. We are playing 100 games per Draw.Max Games Dry is the maximum number of games that occurred with no Wins. It is important because if you were

Reply #21

BobP — Mon, 08 Mar 2004 01:18:06 GMT

Colin: Can you define the terms Games/Win Max Games Dry 3 4 5Run 1 53 326 884 37 There were 500 draws right? So what does the 53 and 326 mean? Am I correct in assuming my lines had 884 3# wins and 37 4# wins? Not sure what I'm looking at here, sorry. To me it looks like I'm playing at house odds. Please go into some detail what Max Games Dry mea... [ More ]

Reply #20

Colin F — Sun, 07 Mar 2004 15:27:53 GMT

GooseJust scroll back to Johnph77 entry in this thread and all the info you want is on, as I keep saying, his excellent site. I think he might teach you a thing or two about cards.But you be careful there cause I think you're in danger of information overload. Just take note that someone as clever as Bertrand Russell still made time to go for 50 mile walks on the weekend. And also remember that Albert Einstein didn't wear socks and only used water when shaving. Now I'm not sure how relevant that... [ More ]

Reply #19

goose — Sun, 07 Mar 2004 08:45:57 GMT

hi colin can you introduce us basic info on Australia lotto(not lottery)?thanks

Reply #18

Colin F — Sat, 06 Mar 2004 10:19:18 GMT

Win DTwo things you need every day for a longer life. A good laugh is one - thanks for that.Now, wheres the Missus ...RegardsColin

Reply #17

BobP — Sat, 06 Mar 2004 05:51:05 GMT

Quote: Originally posted by Colin F on March 05, 2004 Bop PJust another thought. If you would like me to run some tests I'd be happy too.The one I did on Sums/Roots/Even was easy as it was the same each draw. If you wanted me to test on something more elaborate then to protect your Code you would have put it in an Access MDE database which I could reference; I send it the Draw - you calculate the Combinations you want to play and put them in a Table for each Draw.It is my intention dventually to... [ More ]

Reply #16

EXCALIBUR — Sat, 06 Mar 2004 01:11:52 GMT

Colin.By your words, somebody who does not know neither math nor programming like me, would think that you are using some sort of random number generator(s) to find out what the particular random number(s) tendencies are at a given particular given time, as if your random generator(s) were somehow connected and or syncronized with the lottery machines and the random forces or tendencies of both were the same or close enough as if maybe the random forces or laws were universal and would affect al... [ More ]

Reply #15

WIN D — Sat, 06 Mar 2004 00:41:33 GMT

Colin,Noah Webster in his 1828 American dictionary. He insisted English should be spelled as it sounds and other rules. Realize instead of realise, center instead of centre, color instead of colour, tire instead of tyre, wagon instead of waggon, defence instead of defense - these were his ideas. (Poor Canadians...they have tire centres!) Don't know where curb came from...kerb seems more logical! (He also ruled that waistcoat should be pronounced waste-cote instead of weskit. [Notice t... [ More ]

Reply #14

Colin F — Fri, 05 Mar 2004 21:04:10 GMT

Thanks for pitching in John. You want to look up the odds just go to John's excellent site.As you can see BobP to get an expected figure for 3's just divide the total games by the odds figure - 50000/60.91 to give 821.Another important figure to note from those I gave is the maximum dry run which even though we are playing 100 games per draw can still go to 30 draws. This can't be worked out mathematically; it should provide solace for anyone that's on a bad run with their predicting.Another poi... [ More ]

Reply #13

johnph77 — Fri, 05 Mar 2004 14:28:34 GMT

Colin -Although not appropriate for this thread, FYI, I just posted this on another lottery forum.Draw 6 from 49 with BonusHit 6 - 1 possibility, odds - 1::13,983,816.00Hit 5 + Bonus - 6 possibilities, odds - 1::2,330,636.00Hit 5 - 252 possibilities, odds - 1::55,491.33Hit 4 + Bonus - 630 possibilities, odds - 1::22,196.33Hit 4 - 12,915 possibilities, odds - 1::1,082.76Hit 3 + Bonus - 17,220 possibilities, odds - 1::812.07Hit 3 - 229,600 possibilities, odds - 60.91Hit 2 + Bonus - 172,200 possibi... [ More ]

Reply #12

Colin F — Fri, 05 Mar 2004 11:14:17 GMT

Bop PJust another thought. If you would like me to run some tests I'd be happy too.The one I did on Sums/Roots/Even was easy as it was the same each draw. If you wanted me to test on something more elaborate then to protect your Code you would have put it in an Access MDE database which I could reference; I send it the Draw - you calculate the Combinations you want to play and put them in a Table for each Draw.It is my intention dventually to provide a commercial version of my Computer Model and... [ More ]

Reply #11

Colin F — Fri, 05 Mar 2004 10:55:03 GMT

BobPThere is a big trap that is easy to fall into when looking at Lotto Histories and that is to concentrate too much on one History; I know because I scrapped many hours of work when I realized that I was in fact optimizing for that particular History.So a cardinal rule is to make it work with Random Draws you generate yourself which is what I did for all runs.I can do threes if you want threes; do twos if you want two; even do 4 and Bonus if you want. I just picked a 6 from 49 game with the bo... [ More ]

Reply #10

BobP — Fri, 05 Mar 2004 06:20:47 GMT

Is it that the game you were testing against has no 3# win only 3#+bonus? It is easy enough to get a 3# win or better about 98% of the time in 6/49 with 100 well constructed lines, or am I missing something here? BobP

Reply #9

MikeK — Wed, 03 Mar 2004 14:22:02 GMT

Colin,There are two different goals here:To consistently get better than Random NumbersTo win Big.In the case of Jackpot or Lotto games (Pick5, Pick6 etc.) the first of the goals does not really get you very far. After all, as mentioned earlier, the pay-out amounts are so insignificant that it does not really help. It is also very likely that a system is so terrible that in cannot at least be comparable to a Random Number Selection Approach, at least in the long term.So the A+ System would be th... [ More ]

Reply #8

BobP — Tue, 02 Mar 2004 03:29:45 GMT

I think Sums and other filters are useful if targeting a jackpot win only, as even five out of six number wins are made less likely through the use of filters. If you would be happy with 3, 4 and 5 number wins while waiting to win a jackpot you are better off with an open cover wheel to reduce the number of lines then destroying the value of a wheel (making it a worse open cover) by stripping out lines based on filter values rather then coverage.BobP

Reply #7

WIN D — Mon, 01 Mar 2004 22:59:42 GMT

Hyper, It makes no difference how you look at it.... as long as you play and think consistently. Serious players may not win all the time ....the odds you know.... but we are consistently closer than a guess. The draw is random.......But there is a huge .......I said huge ....amount of forecast that can be made in all these games everywhere and every year. For a random draw .....it's maddening how many consistencies there are. Many! Almost too many to call it random, so many that alot of peop... [ More ]

Reply #6

hypersoniq — Mon, 01 Mar 2004 21:51:51 GMT

Sums are one way to profile a draw after the fact...Sum, odd/even, high/low, vtrac (on pick 3 4)... all operate on data you already have in front of you ,and the only thing they do is give you fewer choices to pick from when making a guess, as even odd/even cannot yet be predicted with anything approaching consistent results. Any of the above abstract transformations of draw history have 2 things often overlooked...1. They must be broken down to get playable numbers, which results in more

Reply #5

LAVERNE MALONEY — Mon, 01 Mar 2004 16:27:30 GMT

Hey there, I use sums everyday as one of my lottery tools to determine what I am going to play.

Reply #4

EXCALIBUR — Mon, 01 Mar 2004 15:45:10 GMT

color=#0000ffhttp://www.lottotutor.com/millennium/100.htmlAt last I got it right, tyhe damn color codes messed me up the last time, the first time was a missing n on millennium.Bye.

Reply #3

EXCALIBUR — Mon, 01 Mar 2004 15:41:28 GMT

Sorry I made an error, but this is it:color=#0000ffhttp://www.lottotutor.com/millennium/100.html

Reply #2

EXCALIBUR — Mon, 01 Mar 2004 15:38:43 GMT

I sugest that You take a look at this:http://www.lottotutor.com/millenium/100.html

Reply #1

goose — Mon, 01 Mar 2004 15:27:30 GMT

you

SUMS

MrSteel01 — Mon, 01 Mar 2004 15:11:38 GMT

SUMS is the most un-used lottery idea ! Every thing has a sum!3 digit 000=0 thru 999=27 Avg sum 14/15 4 digit 0000=0 thru 9999=36 Avg sum 18 5 digit 1 thru 37? sum low 15/high 185! What's the lowest/highest sum that's come up?6 digit 1 thru 49? sum low 21/high 279! What's the lowest/highest sum that's come up? Play the most bonus/mega/or plus number that comes in the most! w/your pick 6 number