Cold numbers coming out

[GillesD]

Recently Maggie raised an interesting question about the odds of cold numbers coming out.

Beaker defines a cold number as one that has skipped 10 or more draws. I agree with that.

But I have to disagree with some of the answers provided by Beaker:

1 - I do not think that the odds are the same for 3 or 4 numbers to come out. The odds will change depending on the number of cold numbers in all the 49 numbers and the odds of getting 3 of those cold numbers will always be greater than getting 4 of them.

2 - Looking at my database, I find only 6 draws where were 5 cold numbers came out; these are draws #476, 694, 772, 1180, 1542 and 1679. Beaker, what are the other 3 so I can check my database?

3 - As far as the longest absence, I have #7 coming out in draw #1572 after an absence of 79 draws. Followed by #28 in draw #333 after skipping 72 draws.

4 - I am not sure about your comment that "at least 1 number that hits will be cold". How about draw #1937 where none of the numbers coming out would have been considered cold. I think that in about 14% of the draws, no cold numbers came out.

5 - Obviously, for the first 9 draws, there were no cold numbers but after that the number of cold numbers averaged nearly 13.5 for each draw. The maximum number of cold numbers was 20 for draw #1241 and the minimum was 7 for draw#933.

By comparing our results, this will allow to validate our database and our calculations.

 

[ComboManiac]

msg to GillesD

Is it really true that a number skipped 72 times
That is very bad ..How many months is that ?
Regards

 

[dwoods99]

Well, 2 draws per week would make it 36 weeks, which could be 36/4 = 9 months... however 3 months is actually 13 weeks, therefore 36 weeks is 13+13+10... so in months, that would be 3+3+2 months plus 2 weeks = 8 months, 2 weeks... Ok, so now I am

 

[ComboManiac]

8 months is very bad

Dwoods99,

That is a nightmare, I think for bc/49 maybe it's 2 months no show but not 8 months.
I never thought that would happen

 

[Maggie]

Combo,
The last time #6 was drawn in 6/49 was May 29th. I believe that would make it 22 draws ago. It'll be interesting to see when that one makes an appearance. I'm thinking it won't be long now.

 

[Beaker]

Recently Maggie raised an interesting question about the odds of cold numbers coming out.

Beaker defines a cold number as one that has skipped 10 or more draws. I agree with that.

But I have to disagree with some of the answers provided by Beaker:

1 - I do not think that the odds are the same for 3 or 4 numbers to come out. The odds will change depending on the number of cold numbers in all the 49 numbers and the odds of getting 3 of those cold numbers will always be greater than getting 4 of them.

2 - Looking at my database, I find only 6 draws where were 5 cold numbers came out; these are draws #476, 694, 772, 1180, 1542 and 1679. Beaker, what are the other 3 so I can check my database?

3 - As far as the longest absence, I have #7 coming out in draw #1572 after an absence of 79 draws. Followed by #28 in draw #333 after skipping 72 draws.

4 - I am not sure about your comment that "at least 1 number that hits will be cold". How about draw #1937 where none of the numbers coming out would have been considered cold. I think that in about 14% of the draws, no cold numbers came out.

5 - Obviously, for the first 9 draws, there were no cold numbers but after that the number of cold numbers averaged nearly 13.5 for each draw. The maximum number of cold numbers was 20 for draw #1241 and the minimum was 7 for draw#933.

By comparing our results, this will allow to validate our database and our calculations.

Thanks GillesD. Let me clarify some points.

1. The odds will change for the number of cold numbers - this is correct - BUT the odds will be the same for any 3 or 4 number combination regardless of how you classify them. This was the point I was trying to make. If there are 6 hot numbers and 6 cold numbers the odds are the same. You are correct in saying that the odds of getting 3 will always be greater than 4.

2. I included the bonus number in my list so our draw numbers will be different. With the bonus included I get the following:
297,404,476,694,785,1275,1314,1436,1449.

3. Again, bonus numbers were included in my calculations. 1523,1567 and 1571 had 7 as the bonus in that stretch between 1492 and 1572.

4. If we include the bonus number, starting with draw 11 and ending with 1936, I get 82.3% of the draws have at least 1 number that has skipped 10 draws or 8 out of 10 draws will have at least 1 cold number. So if you include the bonus your 14% is a little higher - 2 out of 10 draws will have no cold ones. I'll take the 8 and you can chase the 2 thats why I play a cold one every draw.

5. I didn't calculate this but it is good to know. Based on the figure of 13.5 we should be looking at 2 cold ones/draw.

Thanks again Professor. It is always good to ensure that we are as clear as possible when we publish our stats.

 

[GillesD]

Long runs without coming out

Actually, it is not that uncommon that a number will not come out in 52 (6 months) or more draws. It already happened 17 times.

Numbers 03, 07, 08, 10, 12, 23, 28, 29, 38 and 40 all had one occurence of not appearing in 52 or more draws.
But this also happened twice for number 18 and 49 and surprisingly 3 times for number 06.

Even more surprising is that out of these 17 occurences, it happened 10 times with numbers 18 or lower. This seems to confirm the slight tendency of high numbers coming out more often in Lotto 6/49.

 

[GillesD]

Beaker's remarks

Yes it is always a good idea to tell tha basic assumptions we make when stating some facts.

Personnally, rarely do I consider the bonus number in my calculations as I consider this an almost useless informatin (unless of course you already have 5 of the winning numbers). But considering the bonus number is another way of working with numbers.

 

[Bones]

Cold Numbers

With exception of 17 32 and 6 being extremely cold, I show 20 34 38 44 and 48 as due. The unusual thing about this is that the last 5#s I mentioned also some of the most frequent drawn over the past year. It just might rain!!!!!

 

[dwoods99]

Re: Re: Cold numbers coming out

... I'll take the 8 and you can chase the 2 thats why I play a cold one every draw.

Did you mean play or drink ?

 

[Beaker]

Re: Re: Re: Cold numbers coming out

Did you mean play or drink ?

Should have said pay for a few cold ones

 

[Dennis Bassboss]

Yes it is always a good idea to tell tha basic assumptions we make when stating some facts.

Personnally, rarely do I consider the bonus number in my calculations as I consider this an almost useless informatin (unless of course you already have 5 of the winning numbers). But considering the bonus number is another way of working with numbers.

I for myself always use both ...with and without bonus...It is not as useless as it might seems because it can change the whole datas from the skipping factors...and on numerous occasions what has appeared as a bonus in the very near preceding draws will show up as regular number in the few next to come.....The bonus number always plays a role in the current trend somehow....

 

[GillesD]

Bonus number

Dennis, when you say "what has appeared as a bonus in the very near preceding draws will show up as regular number in the few next to come" is most likely true. But then, I can say "any number from 1 to 49 will show up as regular number in the few next to come" and this will also be true.

To verify your statement, I checked when the bonus number showed up in the next draws. In 232 cases, it showed in the next draw and in 195 cases in the second next draw. Overall the average was in the next 8.31 draw.

To verify my statement, I replaced the bonus number by a randomly generated number between 1 and 49 and made the same calculations. To ensure a good sampling, I ran this exercice 5 times. The numbers I generated came out on average 237.6 times in the next draw and 206.6 times in the second next draw. The overall average was for all numbers was 8.27. The individual average for the five trials were 7.97, 8.24, 8.11, 8.14 and 8.61. This is consistent with the fact that a number should come out every 8.17 draw (in theory).

Based on this data, I can not see how the the bonus number can be used for predicting future numbers coming out any more than a random number.

 

[Dennis Bassboss]

The answer is simple it is because we know in that case exactly which number were dealing with so we do not have to look for all the 48 others...but there is more to it often the numbers that precedes or follow the bonus number will show up in the very near next draw to come...and there is also even more to it....consecutives, LDs and trend....If I were you I would search the number 32 ...on many occasions not always and I'm conceding that to you Gilles (One player that I have the deepest respect for) but on many occasions the 32 as a bonus is announcing a triple LD or 2 double LDs and I have nailed many of these just by following that logic that sounds not so logic...I respect your opinion deeply and I am very sincere here but I for one will always use the bonus number for analysing purposes...Since I know that you like facts ...lets see for this exercises I will take the 48-49 pair ...one of the numbers forming the pair came up as a bonus on 12 occasions on 8 of these occasions one number forming that pair repeated in the very next draw you can check the following draws
262,304,316,411,614,903,913,1215, for the other 4 draws see what happenned
-draw 839 it jump the following draw but came up on draw 841,
-draw 1046 was one rare one that it did not happenned,
-draw 1765 it jumped 2 and strikes back in draw 1768,
-draw 1838 it jumped 4 and strikes back in draw 1843Based on that if I ever see the 48-49 pair ...you can be sure that I'm going to bet on a repeat in the very next draw bonus or not!

 

[GillesD]

Bonus 32 as a predictor

I checked the draw where 32 came out as a bonus in the draw before with the following results:

32 came out 35 times as a bonus number with the last digit distribution as follow:
1-1-1-1-1-1: 8 times or 22.9%
2-1-1-1-1: 19 times or 54.3%
2-2-1-1: 4 times or 11.4%
3-1-1-1: 3 times or 8.6%
3-2-1: 1 time or 2.9%

The theorical expectation values for each of these last digit distributions are:
1-1-1-1-1-1: 20.6%
2-1-1-1-1: 49.6%
2-2-1-1: 19.8%
3-1-1-1: 6.8%
3-2-1: 1 time or 2.3%

To me, 32 as a bonus did not very well predict 2 double LD (11.4% vs 19.8%). As far as predicting triple LD, it did better (11.4% vs 8.9%). But with a sample of only 35 cases, is it significant or just normal random variation? I think it os just normal random variation but most likely, you will believe the contrary.

I just hope my data is right because calculations were done manually.

 

[Dennis Bassboss]

Maybe you forgot the bonus in these datas!!!

 

[GillesD]

No I did not use the bonus number (but I did not forget it).

Can you give me the theorical probabilities for the following distributions of last digit (including the bonus number):

1-1-1-1-1-1-1, 2-1-1-1-1-1, 2-2-1-1-1, 2-2-2-1, 3-1-1-1-1, 3-2-1-1, 3-3-1

You must have this data because to say that a number can be used as an announcer for future values, it must provide results with a significantly better percentage than the theorical values.

 

[Dennis Bassboss]

After the 32 come as a bonus this is what happenned like it or not!!!
In red triple Lds
In green two double Lds (sometimes with triples Lds)
In blue double Lds


Draw––––numbers––––bonus
0166–07-11-14-18-34-45–-44
0172–11-23-43-46-47-49–-01
0176–07-09-21-28-29-35–-10
0202–17-21-22-25-39-47–-44
0215–01-12-15-19-22-30–-36
0272–04-11-25-32-45-46–-10
0307–08-11-16-20-27-44–-36
0350–03-12-18-22-31-36–-37
0386–14-18-23-25-28-30–-36
0394–09-16-17-24-35-41–-10
0440–06-12-18-24-32-35–-22
0461–02-18-24-25-27-30–-46
0472–02-07-10-14-23-42–-17
0489–02-04-06-11-42-44–-26
0655–02-05-11-27-31-43–-30
0659–03-22-23-31-33-47–-28
0664–01-03-30-36-42-43–-23
0844–01-12-20-25-29-43–-37
0866–14-21-25-28-36-37–-39
0916–22-28-29-30-36-47–-16
0976–12-17-33-34-40-48–-36
1067–09-16-22-27-38-48–-35
1100–03-08-11-12-21-48–-43
1108–01-02-34-35-45-47–-49
1231–02-06-18-28-32-44–-20
1247–01-10-16-24-39-48–-30
1289–03-09-29-31-44-49–-26
1386–02-08-16-31-34-36–-13
1736–09-11-19-22-24-42–-12
1810–02-19-25-30-41-45–-21
1813–03-10-17-22-24-40–-46
1817–02-03-13-17-23-32–-39
1860–04-15-21-31-41-49–-43
1878–07-09-22-35-40-47–-36
1884–07-14-20-30-35-36–-10
Now I'm counting 9 triple Lds and 6 two Double LDs (not counting the ones with the triples)here...ouch out of only 35 draws remember....IS THERE something wrong with my datas???? Maybe my eyes are playing tricks on me but I'm dealing with facts here and only facts!

If I were you I would search the number 32 ...on many occasions not always and I'm conceding that to you Gilles (One player that I have the deepest respect for) but on many occasions the 32 as a bonus is announcing a triple LD or 2 double LDs and I have nailed many of these just by following that logic that sounds not so logic...I respect your opinion deeply and I am very sincere here but I for one will always use the bonus number for analysing purposes...!

As for your theorical value almost half of the time is more than enough for me...I do not care about theory I deal only with facts and this one is overwhelming just like that 48-49 thing that you quickly evaded...In fact I was only saying a true thing...I know that you would have like to find an error in what I was saying but better luck next time!

 

[GillesD]

32 as a predictor

You are right, Dennis, lets stick to facts and forget about theorical values.

With 32 as a predictor, I find almost the same data as you; in 15 cases, I find that the next draw has at least 2 double LD or better (triple LD, etc.). This amounts to 42.9%. You have 16 but my values for draw #1108 are 1-2-34-35-45-47-49, while your values are 2-3-13-27-30-49-7 which happen to be the results of draw #1109. But who cares, 42.9% is still a nice percentage for 2 double LD or better as you say.

I would then suggest than you run the same analysis on all 1937 results up to date. I did this and found out that in 50.1% of the cases, it has at least 2 double LD or better (even a few quadruple LD).

So I would say that in half the draws you can expect 2 double LD or better. With nearly 43%, 32 shows a good performance but it is still less than what we would expect on a random basis. To me it is the same as saying that you got excellent results in a course with a 70% average without telling that the overall average for all participants was 80%.

 

[Dennis Bassboss]

Yes I admit that for 1108 I had an error I just mistype a draw...so what???? I also corrected you before remember you had a mistake in your database last year if you do not I surely do....if this is the little game you want to play...but I'm not interested in that kind of a game. And you are still evading that 48-49 thing....Are you trying to tell me that every numbers that comes as a bonus (all of them are) are announcing lots of triples Lds or two Double Lds in the very next draw???
Hum this is unfamiliar for you to make assumptions without facts maybe I'm dealing with another alias here..???
Prove me wrong and I'll take it ...but it is impossible to do that because what I only initially state as an example is true originally conceding to you that it is not always the case but very often read my initial statement again to avoid further misconception or misunderstanding of what I have said ....and I did not talk about other numbers.....as for the rest I will still always take good notes of the bonus in my stats...Like it or not...I won't respond to you anymore as I have to study the Super 7 for myself for tonite....And I have waisted enough time on that here....And if you had only took the time and look more carefully you would have noticed that many times it is in fact announcing Lds 00-01-02-03.....