(New member) Have any of you broken down a keno game?Prev TopicNext Topic

• Datapile

  

  United States
  Member #181,332
  April 22, 2017
  68 Posts
  Offline
  Apr 24, 2017, 10:25 pm

  ▲▼

  The diamond is all 61 numbers possible for each slot, using 1 row for each number. The ellipse is formed by all the numbers that have actually been drawn. (It radiate out from the central line.)

  Edmond: There are 72,519 stones in my walls. I've counted them many times.

  Abbe Faria: But have you named them yet?

  The Count of Monte Cristo by Alexandre Dumas
• Datapile

  

  United States
  Member #181,332
  April 22, 2017
  68 Posts
  Offline
  Apr 24, 2017, 10:27 pm

  ▲▼

  You're going to confound my efforts to offer up something interesting\new, aren't you?

  Edmond: There are 72,519 stones in my walls. I've counted them many times.

  Abbe Faria: But have you named them yet?

  The Count of Monte Cristo by Alexandre Dumas
• Novan60

  

  United States
  Member #180,181
  February 23, 2017
  125 Posts
  Offline
  Apr 24, 2017, 11:40 pm

  ▲▼
  In response to Datapile

  Quote: Originally posted by Datapile on Apr 24, 2017

  You're going to confound my efforts to offer up something interesting\new, aren't you?

  Nope, just trying to save you some time and encourage you to start thinking beyond just the math of it. Although if you start going down an avenue I have not already explored I will be more than happy to assist.
• Datapile

  

  United States
  Member #181,332
  April 22, 2017
  68 Posts
  Offline
  Apr 25, 2017, 6:41 pm

  ▲▼

  Ok... I have something for you, but I recommend you graph this to get the ramifications, because it's hard to explain. (I remember your preferences for no visuals.)

  So the Apex rows that I put up earlier, there are multiple correlations in the data. Here is how I made the table at the bottom:

  1. For each drawing, note the apex line that it was on. So you have DrawID, ApexofN1, ApexofN2, etc.
  2. For each record in the new table, take the abs(ApexofN1 - ApexofN2) as ApexDelta1 or AD1, abs(ApexofN2 - ApexofN3) as AD2... abs(ApexofN19 - ApexofN20) as AD19. So now we have the absolute difference between the Apexes for each N to N+1
  3. Tally the values for all the drawings, and you get:

  ApexDelta	AD1	AD2	AD3	AD4	AD5	AD6	AD7	AD8	AD9	AD10	AD11	AD12	AD13	AD14	AD15	AD16	AD17	AD18	AD19
  0	192723	131590	141090	100588	120092	126479	91615	127596	141262	40162	141574	128350	91533	126575	119625	99795	141176	131169	192044
  1	315683	224704	170788	189711	153001	117934	170123	88156	79624	257317	79249	88055	170210	118268	152705	189495	170862	225287	314287
  2	193464	258088	223203	216837	194465	233182	178209	260636	263952	59623	264328	259948	178986	234961	194516	216344	222971	258256	193044
  3	178805	190116	184165	169181	155378	129976	162600	89258	80150	251399	79437	89615	163457	128877	154782	168976	184742	190659	179806
  4	121682	115399	149874	150225	200433	193401	166526	238686	260566	79517	260768	238996	166104	192684	201491	150757	149341	115036	122582
  5	60201	139920	146178	211016	153961	168136	159584	115168	84698	258065	84890	114547	159881	169027	155204	211361	145943	140427	60513
  6	44534	82997	106595	44005	95769	138257	165804	219357	265024	65285	265494	219935	164531	137745	95123	44141	107077	82345	44404
  7	113411	28329	24998	87347	144213	106948	135924	81753	42132	222185	42265	81426	135765	106426	143566	87357	24722	28091	113319
  8	32679	53574	67579	86804	32583	34187	14164	23543	26228	15513	25816	23826	14078	34221	32626	87649	68017	52846	33003
  9	23726	43585	68733	15788	17741	18820	22906	16409	14929	18010	14870	16139	22892	18852	17934	15641	68316	44198	23892
  10	17558	12416	10328	12660	15309	13451	7698	14790	16080	12307	16317	14716	7660	13513	15125	12696	10492	12579	17603
  11	12700	28560	10469	13176	9712	11530	15256	12660	11869	10602	11785	12446	15367	11490	9586	13071	10294	28277	12682
  12	9179	7160	7803	9014	10862	8818	6945	8883	9722	9704	9689	8999	6983	8711	10843	9011	7811	7092	9093
  13	6465	4957	6820	6908	6588	7099	9880	9148	8729	6194	8654	9168	9774	7135	6603	7011	6839	4984	6662
  14	4760	5007	5044	5780	6058	6452	5689	5239	5866	7623	5697	5241	5644	6407	6289	5660	5033	5039	4614
  15	3354	3210	2849	4872	4724	5006	5079	6855	6485	3245	6495	6651	5131	4906	4620	4919	2903	3230	3357
  16	2240	2342	3318	3284	3663	4223	4584	3215	3323	5784	3306	3269	4661	4249	3655	3196	3280	2346	2389
  17	1614	1804	2176	2470	3727	2630	2857	4166	4510	1549	4529	4200	2763	2714	3938	2472	2235	1807	1641
  18	1148	1097	1923	2610	1910	3097	3050	2358	1926	4113	1956	2355	3024	3014	1906	2684	1935	1125	1090
  19	804	1118	1142	1270	1862	1686	2299	2430	3165	812	3045	2489	2217	1593	1876	1277	1151	1193	734
  20	565	784	831	1174	1724	1747	1895	1583	1199	2727	1200	1642	1851	1741	1809	1175	843	771	499
  21	361	480	635	1009	983	1171	1312	1409	1994	958	2044	1424	1346	1170	1007	958	720	489	378
  22	209	350	429	761	926	1059	895	1204	785	1689	752	1150	960	957	934	779	388	372	235
  23	182	265	467	433	828	722	926	783	1219	582	1186	786	919	753	787	454	415	252	157
  24	109	154	277	430	359	661	506	788	534	1041	507	738	587	676	381	440	243	158	113
  25	73	114	190	287	505	405	501	458	671	346	696	444	465	414	465	266	178	111	86
  26	45	98	153	209	256	301	443	508	325	642	343	492	435	292	281	219	116	77	46
  27	31	48	95	185	201	307	246	263	378	226	406	263	232	293	167	191	82	49	26
  28	18	26	74	100	140	143	214	311	200	345	215	292	276	175	142	104	72	26	17
  29	12	24	41	55	116	124	173	160	232	152	222	183	175	146	114	62	52	22	13
  30	7	10	36	48	82	126	108	173	148	199	127	177	110	102	66	50	43	14	4
  31	2	9	18	41	63	60	107	114	97	85	133	106	117	77	49	55	26	8	6
  32	2	4	14	21	35	55	46	64	85	111	113	68	44	53	49	31	14	9	4
  33	1	4	5	19	25	52	62	73	61	55	60	71	38	52	33	19	4	1	3
  34	1	2	6	10	15	33	42	31	46	60	48	35	47	21	8	14	6	1	1
  35	0	1	1	2	15	26	20	44	36	31	34	35	26	14	20	8	3	1	1
  36	0	0	1	7	13	11	16	13	29	32	33	23	19	13	11	3	0	0	0
  37	0	1	0	7	0	11	17	17	21	14	23	18	13	13	2	4	1	1	0
  38	0	1	0	0	2	10	10	21	17	14	17	12	10	7	6	3	2	0	0
  39	0	0	0	1	2	3	7	8	8	15	6	5	3	3	1	0	0	0	0
  40	0	0	0	0	3	2	5	9	4	6	4	4	4	2	2	0	0	0	0
  41	0	0	0	1	2	2	2	6	7	3	4	3	4	2	0	0	0	0	0
  42	0	0	0	2	0	2	1	1	4	0	5	5	2	3	1	0	0	0	0
  43	0	0	0	0	0	0	1	1	4	3	3	0	2	1	0	0	0	0	0
  44	0	0	0	0	1	1	0	0	3	0	1	1	1	0	0	0	0	0	0
  45	0	0	0	0	1	2	0	0	0	1	1	0	1	0	0	0	0	0	0
  46	0	0	0	0	0	0	0	0	1	2	1	0	0	0	0	0	0	0	0
  47	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  48	0	0	0	0	0	0	1	0	0	0	0	0	0	0	0	0	0	0	0
  49	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  50	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  51	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  52	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  53	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  54	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  55	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  56	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  57	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  58	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  59	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  60	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0

  Edmond: There are 72,519 stones in my walls. I've counted them many times.

  Abbe Faria: But have you named them yet?

  The Count of Monte Cristo by Alexandre Dumas
• Datapile

  

  United States
  Member #181,332
  April 22, 2017
  68 Posts
  Offline
  Apr 25, 2017, 6:46 pm

  ▲▼

  AD1 is almost identical to AD19, AD2 to AD18, etc.

  If you graph it, those 19 columns turn into what appears as 10 lines.

  The other thing you will see is the several jagged patterns that appear in the graph.

  Edmond: There are 72,519 stones in my walls. I've counted them many times.

  Abbe Faria: But have you named them yet?

  The Count of Monte Cristo by Alexandre Dumas
• Datapile

  

  United States
  Member #181,332
  April 22, 2017
  68 Posts
  Offline
  Apr 25, 2017, 7:05 pm

  ▲▼

  If you track the MaxApex for all the drawings, and count them up, there is a pronounced ripple in the bell curve.

  Query1

  ApexMax	CountOfApexMax
  5	4
  6	135
  7	800
  8	2625
  9	7697
  10	13329
  11	26311
  12	35075
  13	48149
  14	60791
  15	61379
  16	79221
  17	70641
  18	84230
  19	70873
  20	81632
  21	73412
  22	68138
  23	65365
  24	59848
  25	56124
  26	50017
  27	43922
  28	41364
  29	37157
  30	32049
  31	27511
  32	24796
  33	21274
  34	17742
  35	14554
  36	12903
  37	10816
  38	8049
  39	7303
  40	5591
  41	4046
  42	3701
  43	2746
  44	1883
  45	1747
  46	961
  47	865
  48	538
  49	414
  50	221
  51	163
  52	110
  53	62
  54	26
  55	19
  56	13
  57	1
  58	5

  Edmond: There are 72,519 stones in my walls. I've counted them many times.

  Abbe Faria: But have you named them yet?

  The Count of Monte Cristo by Alexandre Dumas
• dr san

  

  bgonçalves
  Brasil
  Member #92,560
  June 9, 2010
  3,781 Posts
  Offline
  Apr 25, 2017, 7:08 pm

  ▲▼

  Hello,DATAPILE my keno lottery date, has 100 numbers and are drawn 20 numbers,
    If you can play up to 50 numbers, it is possible to adapt unprovised in this study,
    So it changes from 80 numbers to 100 numbers, the rest is the same
• Sunglasses

  Belgium
  Member #173,925
  March 26, 2016
  1,396 Posts
  Offline
  Apr 25, 2017, 7:21 pm

  ▲▼

  20/70 pick 2 to pick 10 playable by forms with options for x days and x times.

  There is no break down in odd/even, or low/high. You can play only low, or only high if you want to.

  There is no wheel, just play 20 numbers spread over two pick 10's.

  There is no size for everywhere, payouts differ. If I get 5000:-1 for 7/7, I can play that. If I get 3000 or less, I don't. ... Statistics? How much of that do you want? I stuff you!
• Novan60

  

  United States
  Member #180,181
  February 23, 2017
  125 Posts
  Offline
  Apr 25, 2017, 7:39 pm

  ▲▼

  While identifying and understanding patterns is helpful, I would advise that you focus your energies on developing a method of picking within a given pattern. I spent a lot of time trying to find the optimum pattern so that I could have my computer make all my picks for me. What I found was that while some patterns may have offered better odds than just picking randomly from the big pool of numbers, in the end I still had to figure out the best way of making a 1 in X choice and that I was better at making those choices than my computer.

  One of the reasons why I converted the number sets to bits was that it allowed me to make simple Yes/No choices.  The problem was that I had to make 41 of them and the best I was able to consistently concurrently get right was 18. I think the highest I managed to get right in total was in the low 30's. It is always that last inch that is the killer.

  As I pointed out earlier I was eventually able to get to a point where I had to choose between 1 and 6 ten times by looking for change patterns in sets of bits. It sounds pretty easy and on average there were 1800 possible winning combinations that I would try to match over a 7 drawing period, but it is not. So far the best I've been able to do is 8 out of 10 which ironically yielded no money because the two I got wrong were in the early positions which is one of the downsides of using that method.

  While I can identify trends in the winning sets, from drawing to drawing it is very difficult to predict which numbers will appear. I just like it because I have a fixed number of digits to choose from each time as opposed to just picking 10 numbers from a large pool of numbers.

  I have a couple of other pattern methods that yield roughly the same results. Those patterns basically all break down to the 1 in 4 odds requiring more potential drawings with each level of detail that limits my choices. So now instead of looking for more patterns I spend my time trying to figure out the best way to reliably pick amongst the patterns I already know.
• Novan60

  

  United States
  Member #180,181
  February 23, 2017
  125 Posts
  Offline
  Apr 25, 2017, 8:04 pm

  ▲▼

  https://www.lotterypost.com/thread/311275

  https://www.lotterypost.com/thread/311108

  These are other threads that you should check out if you have not already. In short what I did was to take all the number and rearrange them in a specific order 30 times.How the numbers are sorted matters because I went from averaging requiring 28 lines down to 23 by time I was finished exploring that method.

  I would then split the numbers lines in 2 and then decide how many numbers from each side the winning number would be made up of. As I went through each level the number of possible combinations would be reduced until I was left with the winning numbers.

  It was a favorite method of mine as it used volume to my advantage, but it still suffered from the same problem as all of the other methods I tried, there were still about 8-10 instance where I had to pick a number between 1 and 8.
• Datapile

  

  United States
  Member #181,332
  April 22, 2017
  68 Posts
  Offline
  Apr 25, 2017, 9:58 pm

  ▲▼

  While I'm working out your posts, I have something different for you. Kim Rossmo is a Mathematician, and former Police Officer. He made a mathematical model to catch serial predators (rapists\killers). It doesn't work on all of them, but it does work.

  Scenario: An intelligent predator lives somewhere (large city), and targets people at random.

  Rossmo figured out that the majority while hunting would follow a few simple patterns:

  1. Wherever they lived, they wouldn't hunt near their home to avoid someone recognizing them, or attracting too much scrutiny.
  2. They wouldn't hunt so far away from their home, as to be in unfamiliar areas.
  3. They would contain their impulses to a middle ground, shaped like a doughnut, with their home roughly in the center.

  How does this pertain to the lottery? They hunted at random, with no other limiters than the 3 above.

  Rossmo figured out that he could turn this back on them. With multiple attack sites, he could assume a circle around the site, and know that the perp didn't live in the circle. He would then make a larger circle farther away, and be fairly cirtain that the perp lived somewhere between the 2 circles.

  By extending the radii of the circles from each attack site, their 'doughnuts' eventually overlap, and this would indicate where the aggressor most likely lived.

   

  The same concept can be applied to a 2 dimensional number line for random number combinations. (1 is the ID lowest possible com, TotalPossibleComID is the highest possible combination.) I could never get it to be practical, but it almost always worked.

  1. For every drawing, find the ID for the number sequence on the number line. The next sequence drawn will not exist in (LastDrawSequenceID +\- Inner Buffer)- Too close
  2. The next drawing would be less than (LastDrawSequenceID +/- Outer Buffer)- Too far.
  3. The drawing from X nights ago... It has a unique Inner and Outer Buffer that can be applied to its SequenceID to generate 2 null zones, and 2 'target' zones for X number of drawings ahead.

   

  It took my db the better part of a week to gather the inner and outer buffers to 30,000 draws backward, but I did it. I never found a way to make all the calculations in less than 24 hours. But when I used the solution to check backward, it worked out almost every time.

  Edmond: There are 72,519 stones in my walls. I've counted them many times.

  Abbe Faria: But have you named them yet?

  The Count of Monte Cristo by Alexandre Dumas
• Novan60

  

  United States
  Member #180,181
  February 23, 2017
  125 Posts
  Offline
  Apr 25, 2017, 11:39 pm

  ▲▼

  You missed my point and the example that you point to is a perfect example of what I am trying to say. They were not able to figure out exactly what house the perp lived in, just the general neighborhood.

  It is the same way with the numbers, you can get in the right neighborhood, you just aren't going to get the house number, only in this instance we are not talking about a couple of thousand homes, we are talking about millions of subset combinations. 

  Also as I have said before if I play enough numbers or look a number of drawings in the future I can "predict" with great accuracy what numbers will appear within that time frame. What I would not be able to tell you is exactly which drawing it would show up in.

  Lastly you have to be careful when working backwards. I can historically look at a set of conditions from several thousand drawings and come up with calculations that tell me that some combination of 25 numbers will always produce a winner. The problem is that rule will not hold true for 100% of future drawings. Those conditions may never hold true again or the percentage could drop to as low as 10% of the time.
• Datapile

  

  United States
  Member #181,332
  April 22, 2017
  68 Posts
  Offline
  Apr 26, 2017, 8:11 pm

  ▲▼

  Novan, I think I need to explain my intentions a little better before we continue with our chats.

  1. I came here to listen to different perspectives on the goals we share, but I didn't want to come to the party empty handed. So I've shared the more interesting things that I have found, on the off chance anyone hasn't seen them before. It is my hope that some of you find it more interesting than '7 is the most common number in every lottery so you should always play that'.
  2. In looking at your SQL code and the threads you sent me, its clear you are painting a picture with data using oils and fine brushes. I, on the other hand, am using finger paints and my nose. This is very interesting to me, as anyone that I've ever tried to talk to about it in person (besides the dog) are completely lost after the first few sentences. Between your and Jadelottery's posts, its clear to me that I'm in over my head here.
  3. Lastly, and most importantly, I see your points and I actually agree with you completely- 20 out of 20 is impossible. But at 47, I like the idea of chasing the impossible- It's much less trouble\expensive than having a mid-life crisis involving some combination of fancy cars, mistresses, and booze.

  Edmond: There are 72,519 stones in my walls. I've counted them many times.

  Abbe Faria: But have you named them yet?

  The Count of Monte Cristo by Alexandre Dumas
• Novan60

  

  United States
  Member #180,181
  February 23, 2017
  125 Posts
  Offline
  Apr 26, 2017, 9:29 pm

  ▲▼
  In response to Datapile

  Quote: Originally posted by Datapile on Apr 26, 2017

  Novan, I think I need to explain my intentions a little better before we continue with our chats.

  1. I came here to listen to different perspectives on the goals we share, but I didn't want to come to the party empty handed. So I've shared the more interesting things that I have found, on the off chance anyone hasn't seen them before. It is my hope that some of you find it more interesting than '7 is the most common number in every lottery so you should always play that'.
  2. In looking at your SQL code and the threads you sent me, its clear you are painting a picture with data using oils and fine brushes. I, on the other hand, am using finger paints and my nose. This is very interesting to me, as anyone that I've ever tried to talk to about it in person (besides the dog) are completely lost after the first few sentences. Between your and Jadelottery's posts, its clear to me that I'm in over my head here.
  3. Lastly, and most importantly, I see your points and I actually agree with you completely- 20 out of 20 is impossible. But at 47, I like the idea of chasing the impossible- It's much less trouble\expensive than having a mid-life crisis involving some combination of fancy cars, mistresses, and booze.

  My gut says that you are much better at math than me so in that you have an advantage. I've just learned what math I needed to as I went along and have spent countless hours brute forcing my way through an equation that could have taken someone else minutes to figure out. I like to say that when I first started I was basically doing the computer equivalent of counting on my fingers. In some cases I may very well still be.

  My motivations for posting were very similar to yours as normally I am a fairly private person. And like you I don't like the idea of just picking a number because it happened to appear in the last X drawings.

  Ironically enough we are also in the same age range although I'll be 50 this year.

  Where we are different is that I have banged my head on this long enough to realize I'm not going to discover some magical cipher that will allow me to predict what numbers are going to appear in the next drawing. I've moved onto trying a method of picking that works for me that offers a realistic chance of picking a winning number in the next 28 drawings as 20 dollars a day.

  To me that means making as few controlled choices as possible. As I mentioned before I was a big fan of the by bit method and had worked it out to making a 1 in 6 choice 10 times. The problem with that method is that every choice had to be exactly right and that even getting 9 out of 10 right could potentially yield no money. The other downside of that method is that by pick 5 there was usually only one possible winning combination.

  The way that I got it down to 1 in 6 per position was to pick an ideal set of bits form the previous drawing and look for combinations where two bits changed.

  Here is a screen shot of how that worked.

  

  The yellow represents the bits that were changed.

  

  Right now I'm bouncing back and forth between a method where I divide the 80 numbers into 10 sections of 8 and another one where I look at common drawings.

  The problem with the segment method is that winning numbers only occur once every 8 drawings or so. I overcome this by re-sequencing the numbers 20 times which produces winning numbers 99.99% of the time with about 1,100 potential winning combinations per drawing. It sounds simple but there are over 1 billion potential combinations.  Divide that by 1000 and it gets knocked down to 1 in a million.

  On the flip side if I just played 20 bucks my odds would be 1 in 450K.  But so far with some tweaking I've been able to get the odds down to about 1 in 25K before hitting a road block.

  I've yet to build a pick tool for this method because like the bit method by about pick 6 or 7 I am locked into trying to hit a very specific pattern although unlike the bit method the is usually more than one path. The problem is that too many of the segment only have 1 possible digit that can be a winner while another section might have 3 or 4.

  The common drawing method has potential but right now it involves making too many picks per number. While most fo the time 1 is the correct pick there are about 8-9 instances where it is not and the value can go as high as 15. It is very similar in performance to the pick by split method that I referred you to.

  In short what it does is pick numbers by what drawings they share in common. These patterns are fairly repetitive and most numbers can be isolated by identifying 4-5 common drawings. The problem is that there are occasionally numbers that require picking 6 or 9 drawings in order to isolate them.

  What I like about that method is that there is no single path to picking a winner. Each pick has multiple potential winner paths. It is just too many picks to have to make per digit to make it viable. I did go as far as to build a pick screen for this method because I wanted to see what the flow was like. Here is a sample of that screen.

  
• Datapile

  

  United States
  Member #181,332
  April 22, 2017
  68 Posts
  Offline
  Apr 26, 2017, 10:04 pm

  ▲▼

  It's been years since I played frequently. When I do, I buy 20 tickets.

  I built my first prototype when I was downsized at the age of 31. It was all duct tape and staples but it ran without crashing. One night I played with $20 of birthday money I had been saving. 19 of the tickets won for a total of $750. (1 for 8,3 for 7,15 for 6). If every person on the planet bought 1,000 packs of 20 tickets, 1 would be like mine was. Tens of Thousands would jackpot.

  The next night, I was set- played 100 tickets with the states money... The coarse odds are like 1 in 10 wins something. Every single ticket was worth $0. I continued tinkering with it, and found that it would do one of three things- Return $25, Pick all 5s on 20 tickets (6 are needed to win)... or much to the amusement of the store clerk, it would pick 0 on 15-19 of the tickets, returning 1 dollar each.

  Since then, I explore different things, and will play once I have a real working model.

  Edmond: There are 72,519 stones in my walls. I've counted them many times.

  Abbe Faria: But have you named them yet?

  The Count of Monte Cristo by Alexandre Dumas