United States
Member #41,382
June 16, 2006
1,969 Posts
Offline
You guys both missed ALL of my points, that's pretty rare.
The odds of winning the lottery are what they are, but out of 121 or so games, a 5/0 has come in 4 times, an 0/5 has come in 2 times, but a 3/2 has come in 43 times, and a 2/3 has come in 34 times. (4/1-15, 1/4-22)
I understand what you are saying about odds being the same for one set of numbers vs. another, but I don't see how you can say you have just as good a chance winning by playing an 0/5 as you do a 3/2 ?? Yes, I understand the numbers are inanimate objects that don't know what they are, but the fact remains, a 3/2 has a helluva lot better chance of hitting that an 0/5. And to be clear, I don't really care too much WHICH numbers above or below 27/28 I play, I just care that I spread them out the way most of the games hit.
Enlighten me please.
PS - and no, I don't play any of the combinations I listed above, and I bet you don't, either.
mid-Ohio United States
Member #9
March 24, 2001
20,272 Posts
Offline
Quote: Originally posted by guesser on Oct 25, 2006
You guys both missed ALL of my points, that's pretty rare.
The odds of winning the lottery are what they are, but out of 121 or so games, a 5/0 has come in 4 times, an 0/5 has come in 2 times, but a 3/2 has come in 43 times, and a 2/3 has come in 34 times. (4/1-15, 1/4-22)
I understand what you are saying about odds being the same for one set of numbers vs. another, but I don't see how you can say you have just as good a chance winning by playing an 0/5 as you do a 3/2 ?? Yes, I understand the numbers are inanimate objects that don't know what they are, but the fact remains, a 3/2 has a helluva lot better chance of hitting that an 0/5. And to be clear, I don't really care too much WHICH numbers above or below 27/28 I play, I just care that I spread them out the way most of the games hit.
Enlighten me please.
PS - and no, I don't play any of the combinations I listed above, and I bet you don't, either.
guesser,
Looks like you will have to enlighten the guys by actually picking all 5 winning numbers your way.
What the guys are saying is there may be more combinations of 2/3 and 3/2 than there are 0/5 and 5/0 but every combination is unique and one unique combination's odds of winning are no better than those of another. If a certain type of combination hit more often, it only because there are more of that type of combinations in the group of possible combinations.
Note: I believe that there are combinations with a particular profile that have a better chance of having all the winning numbers for the following reasons:
Ohio had a 5/37 game "Buckeye5" from 04/06/93 to 10/02/04. In a 5/37 game there are 435,877 possible combinations of fives and 66,045 possible combinations of fours. During those eleven years there were 2,723 drawings and 6 combinations of fives came up twice and over 1,000 combinations of fours came up several times. Since less than 0.5% of the combinations of fives had a chance of coming up one time and six came up twice and less than 20.6% of the combinations of fours had a chance of coming up once and some came up more than five times, that suggests to me that some combinations are better than others. The question is how do you figure out which ones are better before the game ends or change its matrix. I've been trying to come up with a way to define the profiles of these "better combinations".
* you don't need to buy every combination, just the winning ones *
New Jersey United States
Member #21,205
September 4, 2005
963 Posts
Offline
Quote: Originally posted by guesser on Oct 25, 2006
You guys both missed ALL of my points, that's pretty rare.
The odds of winning the lottery are what they are, but out of 121 or so games, a 5/0 has come in 4 times, an 0/5 has come in 2 times, but a 3/2 has come in 43 times, and a 2/3 has come in 34 times. (4/1-15, 1/4-22)
I understand what you are saying about odds being the same for one set of numbers vs. another, but I don't see how you can say you have just as good a chance winning by playing an 0/5 as you do a 3/2 ?? Yes, I understand the numbers are inanimate objects that don't know what they are, but the fact remains, a 3/2 has a helluva lot better chance of hitting that an 0/5. And to be clear, I don't really care too much WHICH numbers above or below 27/28 I play, I just care that I spread them out the way most of the games hit.
Enlighten me please.
PS - and no, I don't play any of the combinations I listed above, and I bet you don't, either.
It's none of my business how you play the lottery, first of all. You can do what you wish. I think I do understand, at least in general terms, what you are trying to say, but I am telling you that you are kidding yourself.
I'm not sure what you're talking about when you say 0/5, 3/2 etc, but the appeal to past events in probability has no bearing whatsoever on future events, first off.
I'm not sure I can convince you of anything, but you can try this on for size, and see if it helps. (If it doesn't help, that's OK too): Suppose you start playing the Powerball lottery so that you restrict your set of numbers so that one number comes from the set 1-11, one from 12-21, one from 22-32, one from 33-43, and one from 44-55. You are imposing a distribution. Of course when the first number is picked, there is a 100% chance that it will be in one of those groups, but of course, it may not be the number you have picked, unless you have picked all of the numbers. Let's say that the number 31 comes up and that this is the number you have picked from the appropriate group. There are still 10 ways that the next number will be from the same group. If any of those 10 events occur, you will not win the lottery. Granted there are still 44 ways that it will not be in the same group. Thus your odds of losing are now 10/54, while your odds of being right about the distribution will be 44/54. Let's say that by luck, you are right and that the next number picked is 44. Now your odds of being right about the distribution are smaller. There will be 20 ways of the number being in one of the two groups, and only 33 ways of it being in a different group. The next time you go through this iteration - assuming you don't fail on the second one - the odds will be against you being right about the distribution: There will be 30 ways to be wrong and only 22 ways in which you could be right.
If you do this analytically, you will find that any distribution has a certain probability, but that probability assures nothing at all about whether or not the event will actually happen. Any restriction placed on the way numbers are picked automatically results in a probability of less than 100%.
I'm not sure that you get any of this - I'm going to guess that you don't - but these are the facts of the case. Insisting on a particular distribution gives you no edge on the lottery whatsoever.
For the record, I do generally play numbers of my own picking. I do this because I generally buy more than one ticket. I do this because I want to assure that if I buy n tickets, I have reduced the odds against me by as much as 1/n. There is no way, however, to improve my odds beyond that however, unless I buy n+1 or more tickets.
Once in a while though, I pick up a quick pick, just for the hell of it. Hey, you never know.
United States
Member #41,382
June 16, 2006
1,969 Posts
Offline
I never said I could pick the 5 winning numbers my way, or any other way, what I said was, a 3/2 has a better chance of winning than an 0/5.
Yes, I know all about odds, and if it's meant to be, it's meant to be, one losing number set is the same as another losing number set, but if a person could play 100 games of 0/5, or 100 games of 3/2, I think I know where I would take my chances.
Can anyone in all honesty tell me if they have been playing a system, and that system tells them to play numbers 34-35-36-37-38, that the person wouldn't think twice before playing it ? I know a person HAS to play that set if their system arrives at it, if you don't then what makes a person think their system is right any other time, I'd play them, even though I KNOW they have almost no chance of hitting - much much less than a 3/2, or a 2/3.
The other thing to keep in mind is there is more to it than just picking which split to play.
I guess I could shift away from the splits, I will give you all the 6 current longshots, and I will take the other 49 numbers, and I'd bet I have a better shot than you do at winning......... even though the 'odds' say we are equal.
United States
Member #41,382
June 16, 2006
1,969 Posts
Offline
Quote: Originally posted by Prob988 on Oct 25, 2006
It's none of my business how you play the lottery, first of all. You can do what you wish. I think I do understand, at least in general terms, what you are trying to say, but I am telling you that you are kidding yourself.
I'm not sure what you're talking about when you say 0/5, 3/2 etc, but the appeal to past events in probability has no bearing whatsoever on future events, first off.
I'm not sure I can convince you of anything, but you can try this on for size, and see if it helps. (If it doesn't help, that's OK too): Suppose you start playing the Powerball lottery so that you restrict your set of numbers so that one number comes from the set 1-11, one from 12-21, one from 22-32, one from 33-43, and one from 44-55. You are imposing a distribution. Of course when the first number is picked, there is a 100% chance that it will be in one of those groups, but of course, it may not be the number you have picked, unless you have picked all of the numbers. Let's say that the number 31 comes up and that this is the number you have picked from the appropriate group. There are still 10 ways that the next number will be from the same group. If any of those 10 events occur, you will not win the lottery. Granted there are still 44 ways that it will not be in the same group. Thus your odds of losing are now 10/54, while your odds of being right about the distribution will be 44/54. Let's say that by luck, you are right and that the next number picked is 44. Now your odds of being right about the distribution are smaller. There will be 20 ways of the number being in one of the two groups, and only 33 ways of it being in a different group. The next time you go through this iteration - assuming you don't fail on the second one - the odds will be against you being right about the distribution: There will be 30 ways to be wrong and only 22 ways in which you could be right.
If you do this analytically, you will find that any distribution has a certain probability, but that probability assures nothing at all about whether or not the event will actually happen. Any restriction placed on the way numbers are picked automatically results in a probability of less than 100%.
I'm not sure that you get any of this - I'm going to guess that you don't - but these are the facts of the case. Insisting on a particular distribution gives you no edge on the lottery whatsoever.
For the record, I do generally play numbers of my own picking. I do this because I generally buy more than one ticket. I do this because I want to assure that if I buy n tickets, I have reduced the odds against me by as much as 1/n. There is no way, however, to improve my odds beyond that however, unless I buy n+1 or more tickets.
Once in a while though, I pick up a quick pick, just for the hell of it. Hey, you never know.
Well, I do understand you 100%, and I play along those lines. If you have 2 or 3 numbers hit from one of those groups, what is the likeliehood of 2 or 3 from the SAME group hitting in the next game ? It may be 0, it may be 50-50. If they did hit in the next game, what are the odds of it doing the same 3 games in a row ? It happens once in awhile. 4 in a row ? Not very often.
0/5 - no numbers picked under #28, 5 numbers picked over #27.
3/2 - 3 numbers picked under 28, 2 numbers picked over #27.
And so on.
What if #20 hits 3 times in a row, will you pick it to hit in the 4th game ? How about 5 in a row ? 6 in a row ?
I know the odds, and I know we can't predict what will happen, so what *I* try to do is predict what WON'T happen based on recent trends, that's all. Sometimes I'm close, sometimes I'm not, but as I keep saying, a losing ticket is a losing ticket, no matter what.
I don't see how you can possibly say insisting on a certain distribution gives you no edge whatsoever, it has to. It may not get you ANY of the winning numbers, but you have to be in the ballpark to hit it out of the ballpark, and I try to play the odds to get me into the ballpark, at least.
PA United States
Member #22,982
October 6, 2005
2,229 Posts
Offline
What really is bad is that I always beat great odds when they work against me. One time I played 3 last draw eliminations for about $50.00. Because I noticed that very rarely were the same balls drawn again. Sure enough when I decided to play them, 3 balls come up from the last draw alone! Yes, THREE! The odds of that are astronomical, unless of course I play it.
New Jersey United States
Member #21,205
September 4, 2005
963 Posts
Offline
Quote: Originally posted by guesser on Oct 25, 2006
Well, I do understand you 100%, and I play along those lines. If you have 2 or 3 numbers hit from one of those groups, what is the likeliehood of 2 or 3 from the SAME group hitting in the next game ? It may be 0, it may be 50-50. If they did hit in the next game, what are the odds of it doing the same 3 games in a row ? It happens once in awhile. 4 in a row ? Not very often.
0/5 - no numbers picked under #28, 5 numbers picked over #27.
3/2 - 3 numbers picked under 28, 2 numbers picked over #27.
And so on.
What if #20 hits 3 times in a row, will you pick it to hit in the 4th game ? How about 5 in a row ? 6 in a row ?
I know the odds, and I know we can't predict what will happen, so what *I* try to do is predict what WON'T happen based on recent trends, that's all. Sometimes I'm close, sometimes I'm not, but as I keep saying, a losing ticket is a losing ticket, no matter what.
I don't see how you can possibly say insisting on a certain distribution gives you no edge whatsoever, it has to. It may not get you ANY of the winning numbers, but you have to be in the ballpark to hit it out of the ballpark, and I try to play the odds to get me into the ballpark, at least.
Well I figured you wouldn't get it. I can say that insisting on a certain distribution gives you no edge whatsoever because it is true.
The probability of the number 20 coming up six times in a row is one in 1,412,466.987, which is relatively probable compared to winning the lottery. On the seventh draw the probability of 20 coming up will still be 9.4% unless some bias has been introduced involving the weight or size of the ball numbered 20. The lottery, however, claims to avoid this by frequently changing the ball sets.
I know you don't believe it, and don't want apparently to believe it, but there is no advantage whatsoever in looking at past performance.
I have heard, so as to believe it, that sequential number plays (for example 31, 32, 33, 34, 35 )are relatively common and are generally covered by more than one player. (This fact increases the probability of the lottery rolling over beyond the Poisson case I post here, which is why I often specify "randomized probability of a rollover.") Since this is the case, it is likely that people playing the numbers in such a fashion will face a reduced share of the jackpot prize should they win it. For this reason, I tend not to play those numbers, although they are as probable as any other set of 5 numbers. This because such a sequence reduces my expectation value, the ratio of the possible prize to the risk (the odds). Of course, if a series of sequential numbers do come up, and they could, those who hold such tickets will be happy and I will have to wait for another day.
New Jersey United States
Member #21,205
September 4, 2005
963 Posts
Offline
Quote: Originally posted by guesser on Oct 25, 2006
You are mistaken: I DO get what you are saying.
You look at the odds one way, and I look at them a different way, that's all, I don't have any problems with it.
The only thing I am befuddled about is the 'odds' of an 0/5 coming up in the last 120~ games is 4 in 120, and the odds
of a 3/2 are like 43 in 120~, and even though the odds tell you there is no difference in which split is likely to hit, it's
right there in black and white that the odds DO tip towards one way. 4/120= 3.3%, 43/120=35.8%
I think we've about worn out this horse, I'm pretty much done with it.
One doesn't "look" at the odds in any particular way. They are determined mathematically and not by having an opinion about them. The mathematics of the situation has been around for centuries and has not changed much. This is how the lottery is able to post odds on the play slips.
I have shown earlier in this thread that there is no "tipping," but that the experimental outcome is well within what one might expect from the sample size. Sometimes the results are surprising, but none the less true. For instance, when RJOH expressed surprise about the 5/39 results, he was relying, I would guess, on the assumption that the odds change linearly. This is quite an honest mistake, and until I actually did the calculation, I would have thought, based on gut feeling, much as RJOH did. Gut feelings however are often wrong, which is why formal mathematics is used to analyze problems.
You are ignoring some very basic laws of probability when you say "black and white," and I'm quite convinced that you do not undertand what I am saying, since you seem to be asserting that there is something special about being a member of an artificially constructed sample size that happens to be larger (and therefore more difficult to cover) than another sample set. There is nothing mysterious in particular distributions coming up, since some sets of distributions have more members than other. For instance the set of 5 numbers between 1 and 5 has exactly one member, while the set of 5 numbers between 1 and 55 has millions of members. This does not mean that if one buys one ticket in the latter set it is more likely to win in the former set. The instant the ticket is purchased it is restricted to a set with one member, just like the set of numbers 1, 2, 3, 4, 5. It is at that point no different.
United States
Member #41,382
June 16, 2006
1,969 Posts
Offline
We look at the same thing in different ways, I play by tendencies of things not to happen 'forever', on the short-term, and you play differently, I'm not ignoring anything. I factor in the splits: you say it doesn't matter, I have no problem at all with your beliefs, my beliefs tell me the same split generally speaking will not hit more than 3 times in a row, and specifically more than 4 in a row, so if I see that happen, I play against it. I may be right, I may be wrong, there is always another day.
NOTHING WRONG with that.
Either way, we both are battling the same odds, we just see them differently, and as far as I can tell, we both have been wrong as far as the jackpot is concerned.
PA United States
Member #22,982
October 6, 2005
2,229 Posts
Offline
It is the luck of the draw mostly. I have also had numbers repeat when I was hoping none would, and got burnt pretty bad using my system. But when all that happens that is supposed to, systems can really help you narrow it down. But it only takes one number to render your whole system useless, and cost you a lot of money.
United States
Member #41,382
June 16, 2006
1,969 Posts
Offline
Quote: Originally posted by sirbrad on Oct 26, 2006
It is the luck of the draw mostly. I have also had numbers repeat when I was hoping none would, and got burnt pretty bad using my system. But when all that happens that is supposed to, systems can really help you narrow it down. But it only takes one number to render your whole system useless, and cost you a lot of money.
Exactly, Brad.
I've said all along you can be off by one on each number, or off by ten on each number, a loss is a loss.
On another website, I predicted last night's game to be either two groups, or three numbers from one group would
hit, and that's exactly what happened - 42-43-48. (not another 'no group' or 'single group' game). I usually post my thoughts on
my 'blogs' here on LP. I don't predict numbers, I do predict types of games, FWIW.
Before anyone says it, I have numbers in groups of 10 - 01-10, 11-20, 21-30, 31-40, 41-50, 51-55. That will be changing soon due to
something somebody else pointed out to me here yesterday, *IF* my back-testing shows it should.
Will it get me closer ? I don't know... I won't change it if it does not.
mid-Ohio United States
Member #9
March 24, 2001
20,272 Posts
Offline
Systems are designed to work if and when the conditions are right and you can't expect those conditions to exist all the time. For example, in PB since the last matrix change there has been 12 occasions when the winning numbers didn't have any numbers from the previous five drawings so if your system excluded numbers in the previous five drawings it could have only worked 10% of the time but your number pool would have been reduced from 55 to 32-38 numbers too. Even winning with a system takes a little bit of luck.
Note: During the other times you could have match four 27 times and three 39 times.
* you don't need to buy every combination, just the winning ones *
NY United States
Member #23,834
October 16, 2005
4,771 Posts
Offline
Quote: Originally posted by guesser on Oct 25, 2006
We look at the same thing in different ways, I play by tendencies of things not to happen 'forever', on the short-term, and you play differently, I'm not ignoring anything. I factor in the splits: you say it doesn't matter, I have no problem at all with your beliefs, my beliefs tell me the same split generally speaking will not hit more than 3 times in a row, and specifically more than 4 in a row, so if I see that happen, I play against it. I may be right, I may be wrong, there is always another day.
NOTHING WRONG with that.
Either way, we both are battling the same odds, we just see them differently, and as far as I can tell, we both have been wrong as far as the jackpot is concerned.
You're completely right that it's very unlikely that a pattern will repeat forever.
I'm also not sure which patterns you're referring to as 5/0 and 3/2, but as PROB 988 says, it's all based on the relative number of possibilities. Since MM has an even number of numbers to choose from, there are exactly the same chances that all numbers will be odd or all will be even. Eliminating half of the numbers from consideration reduces the possible combinations from 3,819,816 to 98,280, which is not quite 2.6%. That means that the combinations made up of all odd or all even numbers account for just over 5% of the possible combinations. The corallary is that the winning number will be all odd or all even only about 5% of the time, or 1 in every 20 drawings. If you eliminate the last 5 winning numbers from considerationyou'll be picking 5 out of 51, so there will be 2,349,060 that you could choose from, which accounts for 61.5% of the possible combinations. That means that if you assume the winning numbers won't include any of the previous winning numbers you will be right the majority of the time, but only by a small margin.
Now tell me what that does to improve your odds of *winning.*
If all you want to do is pick a set of numbers that will have a pattern (or distribution, or whatever) that's *similar* to the numbers that actually win then your strategy will work about as often as probability says it will. Of course to simply do that you don't need to actually pay for lottery tickets. As you say, a loser is loser, so why would you pay for an increased chance that your losing numbers will be similar to the winning numbers? Assuming your goal is to win, your numbers have to actually *match* the winning numbers. As far as the odds go, your "system" is just as good as any other method for picking numbers, whether it's based on something methodical or throwing darts while blindfolded. That's not because you're on to something, it's because each individual combination, regardless of how it was chosen, has an equal chance if being drawn. What's "wrong" with your system is that it gives you false hope if you think it will actually increase your odds.
FWIW, I don't believe that drawings are actually 100.0% random. The balls can't possibly be absoluely 100% identical as far as weight, balance, air resistance, and whatever other factors might influnece their movement in the machine. That means that your system has a real flaw if it eliminates numbers simplybecause they've managed to work their way to the top of the machine in the past. Since the balls don't have consciousness, let alone memory, there is no such thing as an overdue number. By extension there is no such thing as a number that has come up too often. If numbers come up more or less than expected there are two likely explanations. One is that random events are random, and the other is that some balls are more or less likely to be drawn. Unfortunately there's no way to tell which reason accounts for any given result. Still, if you want a slight advantage you'll either eliminate numbers that have been drawn less than expected or you'll choose a majority of your possible numbers from those that have been drawn more than expected. That might reduce your odds against picking all 5 balls from 1 in 3,819,816 to as little as 1 in 3,817, 213.