Markov Chains - anyone tested the method?

Reply #59

flottor — Mon, 04 Jun 2018 18:13:19 GMT

look what it turned in2!! poor markov scratched a couple hotair balloon heads!! whod have thunk??

markov.exe is a nice program as his author cristiano.. if everybody would be as honest as thorn is:

'My Markov chains lotto software is not very robust yet; a few bugs and errors may appear. At most, it will generate 1000 combinations each time. You may need to delete the last line from the output file, because for some strange reason it gets truncated and doesn't have the 6 numbers!

The pr... [ More ]

Reply #58

flottor — Mon, 04 Jun 2018 18:01:55 GMT

random ilogic sez

I am still a million times smarter than that idiot thinks he is. I may make one more post and then leave it to y'all to carry on.

thats the worst form of IDIOT: grown old man blackmailing liky a retarded child. Im gona leave..im gonna leave!!!

u wanna others start cryin 'dont leave..dont leave us.. we gonna kick the but of da guy who got in ur face..

u left once more when i and others showed ur soft needed a function 2 check what happened in da past..a honest and g... [ More ]

Reply #57

watch out — Mon, 04 Jun 2018 17:55:45 GMT

No no no RL let it be, the only reason I come to this site is to read your posts.

Reply #56

RL-RANDOMLOGIC — Mon, 04 Jun 2018 17:47:19 GMT

Sorry bout that, I was unaware that calling a idiot an idiot was considered an attack, stupid me.

With that being said I am still a million times smarter than that idiot thinks he is. I may make one

more post and then leave it to y'all to carry on.

RL

P.S. hey idiot, I just generated a 5of5 match after completely removing the set from the programs

database. Explain that.

Reply #55

flottor — Mon, 04 Jun 2018 17:42:26 GMT

enit crazy??? i show facts facts facts.. yup they r disputable..but da egotistics put their pride ahead instead of disputing facts factually not in hatred and name calling..

this crap started with a call for normal testing as normal persons do..it set afire a couple of hotheads who r known 4 such behavior..

i know u adamcustum and ion saliu. i remember good when u used to lick his behind (not long ago? u had sumthin personal with him..u also licked rls behind then had had ur runins..u a... [ More ]

Reply #54

watch out — Mon, 04 Jun 2018 17:20:54 GMT

LOL it's funny when two different people can reach the same conclusion.

Reply #53

adamcustom — Mon, 04 Jun 2018 16:59:10 GMT

If it was not that abominable sloppy spelling I would say it must be Ion Saliu with his typical arrogance and disrespect for people.

Reply #52

flottor — Mon, 04 Jun 2018 16:35:41 GMT

u still take it personally mate.. aint good 4 u.. u keep calling names for nuthin.. i stated facts, facts facts..

1. u did curve fitting.. every1 here with basic maths/stats knowledge can tell u that.. u did curve hitting thats why u was scared 1st when u saw the results (jackpots wins)..

2. if u had dleted the top draws in ur file ur tests wouldve been realistic.. u generate combos based on data 100 or 50 or 10 draws back.. then u check the out put for da next 10 50 or 100 draws that u de... [ More ]

Reply #51

watch out — Mon, 04 Jun 2018 15:58:06 GMT

LOL that man flew over my 75 years head in 7 minutes into the 1st video.

Reply #50

Todd — Mon, 04 Jun 2018 15:30:55 GMT

To the person who reported a post here, I'm going to link to another post I just made on this subject, which I have written about a thousand times. Tired of repeating it. https://www.lotterypost.com/thread/319834/5559217

Reply #49

RL-RANDOMLOGIC — Mon, 04 Jun 2018 15:20:22 GMT

Jimbo, cut out the fake persona. If you don't think I know what a Markov chain is your an even a bigger

Idiot than I first believed. I can control how much data is feed to the program, who's operating under a

fallacy now. In your earlier post you suggested removing so many draws and then running the test. What

a load of crap you spew. I am hitting the abuse button, good by jimbo.

RL

Reply #48

flottor — Mon, 04 Jun 2018 14:29:59 GMT

again no offence mate.. u took it da hard way and started to call names as vent.. i had no intent 2 bruise ur bloated ego.. u aint got no clue what markov chains r.. u simply recycled ur lexi program with od even low high and those static settings as u sez in other threads.. aint markovs by any stretch..

u got those jackposts cos u did curve fitting.. when u check ur outputs in future draws u aint gonna get no jackpots...none any tyme soon....

markov is not a strong predictive tool but wor... [ More ]

Reply #47

RL-RANDOMLOGIC — Mon, 04 Jun 2018 09:06:53 GMT

For those wondering what Markov chains are and how they are used to make predictions here is a link

to a excellent tutorial series. You should gain what you need to understand what I am doing from the

first video in the series, that is, with the exception of making the final choice.

Markov chains calculate probabilities and are very accurate but not so much when we need to know what

will happen in the next draw.

https://www.youtube.com/watch?v=Uz3JIp6EvIg

RL

Reply #46

RL-RANDOMLOGIC — Mon, 04 Jun 2018 08:41:18 GMT

Flabber

First, I am not your mate.. whatever that implies. Second, your feelings seem to have been bruised

so may I suggest a little corner time out. Markov chains, ie followers as you call them are very simple

to generate for almost any data type. The steps are a base-10 to 3-digit code. Each digit 0 to 9 is

coded into a specific code, example 212=5. The program first converts the actual line, ie set of

numbers into it's lexicographical index. The index is zero padded so that it... [ More ]

Reply #45

MillionsWanted — Mon, 04 Jun 2018 00:02:00 GMT

Your Markov Chain tool sounds very promising.

For me it's probably more realistic to play 30-50 lines every draw. The Pick-5/6/7 lotteries in my country are drawn once every week.

Reply #44

Kosh — Sun, 03 Jun 2018 23:47:21 GMT

Dear RL,

I have been reading about your RL's Lex-Master software right here in this forum. It is apparent that you have been getting excellent results. Please make this software available for download. If I have to pay to buy this software, that's fine too.

Thanks in advance for considering my proposal.

Sincerely,

Kosh

Reply #43

flottor — Sun, 03 Jun 2018 20:14:23 GMT

dont fool urself mate.. what u do is curve fitting.. u generate combos in a file then check 4 winners in da same file..

more seriously u arent generating markov chains.. u bs sumthin abou lexies.. markov cahains r about followers not lexies.. markov checks all lines in file and looks 4 followers to every number not lexies..

u gotta check liky i said or adacustum.. c what ur 100 lines give u in the next 10 or beter 100 draws..

thorn was 100% honest when he sez his markovexe hit 5/6 once

Reply #42

adamcustom — Sun, 03 Jun 2018 20:04:17 GMT

Craig

I very much look forward to it. You deserve to make it happen more than anybody else in this world.

Adam

Reply #41

RL-RANDOMLOGIC — Sun, 03 Jun 2018 19:47:28 GMT

Adam

I will make a video and PM you a link. Playing 100 lines is more that my budget allows but I

think I have found a way around that. Might blow your mind, I know it looks too good to be

true but I am as serious as a heart attack. It was by chance that I decided to set it up to

generate 100 lines each run. As a single line predictor it was not hitting much of anything and

I was about to move on to something else. It's still possible that some weird problem exist

but one thing... [ More ]

Reply #40

adamcustom — Sun, 03 Jun 2018 19:29:48 GMT

Yes, ok, but anyway what I would do is running it before each new draw for some 10 days or so before using it in real play.

Reply #39

RL-RANDOMLOGIC — Sun, 03 Jun 2018 19:22:59 GMT

The predictor uses a random starting point for each micro prediction of at least 50 to 250 games. This is the

random part of the process. Next it combines then converts the micro predictions to the step values then

converts the steps into the lexicographical index and from the index to the actual line/set.

I spent several hours testing the code for errors this morning and while I am not 100% certain, I am at least

99% sure it's error free. The program never sees the draw it's tryin... [ More ]

Reply #38

adamcustom — Sun, 03 Jun 2018 19:16:30 GMT

No need to delete any draws at all. Forward testing is the thing to do and not back testing. With draws being held everyday one can go one draw predictions at a time and after 10 days you know how it does.

Reply #37

flottor — Sun, 03 Jun 2018 16:02:56 GMT

no offence mate.. u was predicting in the past.. same ting happens with pairs, hot numbers, skips and all that.. u get dozens of jackpots evry jahr..

do ur test like this.. delete the top 100 draws in ur draws file and save as new file.. do ur markovs for the new file.. check the 100 outputs for the original file but the last 100 lines.. u aint gonna see jackpots no more.. if u do da test for the entire draws file u get many jackpost again.. all of em below the 100 line...

how gr8 would be... [ More ]

Reply #36

Gwoof — Sun, 03 Jun 2018 14:46:39 GMT

I was wondering if the Pick 4 Manager mentioned on AssetClass.cc is available for download somewhere.

thanks GW

Reply #35

Gwoof — Sun, 03 Jun 2018 14:40:50 GMT

Hi RL

Are you going to upload the program.

GW

Reply #34

RL-RANDOMLOGIC — Sun, 03 Jun 2018 14:31:20 GMT

Test confirm it's legit.... No code errors.

RL

Reply #33

RL-RANDOMLOGIC — Sun, 03 Jun 2018 08:39:16 GMT

Must be a mistake, here are 3 test for my 5-39 game. In the first 3 runs it managed a 4 of 5, a 5 of 5 and a 3 of 5 in that order.

If this is for real then But in the past when something like this has happened it was due to a mistake in the code.

I am betting it's the latter.

Test #1

Test #2

Test #3

Reply #32

RL-RANDOMLOGIC — Sun, 03 Jun 2018 08:14:54 GMT

Hi all

I ve been playing around with a new markov chain prediction tool and set it up to generate 100 lines each time it's ran.

On the 14th run it managed a 6 of 6. The best the program was able to predict in the previous 13 test was a few 4 of 6

hits. 3 of 6 hits were common but here a 3 of 6 only pays a free play so without the 6of6 hit it would have lost big time

even at 2 plays for $1.00. There is a random element to the predictor and still need to do more testing. I will post mo... [ More ]

Reply #31

jlg69 — Sat, 02 Jun 2018 23:56:32 GMT

Big Big big Thanks to asset Class

Reply #30

flottor — Fri, 01 Jun 2018 20:02:51 GMT

no need 4 u to w8.. there is another markov thread here next page.. its very free markov.exe lotto.. sum1 sez wrong that the prog was by cristiano ronaldo.. its cristiano lopes or thornc..

Reply #29

Kosh — Fri, 01 Jun 2018 19:41:32 GMT

retxx,

I just checked the website. The download links are still not up yet.

The Markov Chain Generator 2.0 is listed as still being in development with a release date of June 2018. So this means that it probably will not be available until later this month.

I guess will all have to still be patient for a while longer.

Regards,

Kosh

Reply #28

retxx — Thu, 31 May 2018 12:58:27 GMT

can someone who still uses it could post the link to download it?thanks

Reply #27

lakerben — Thu, 31 May 2018 02:05:28 GMT

Are all the other programs going to be available next month also

Reply #26

RL-RANDOMLOGIC — Wed, 30 May 2018 08:01:23 GMT

Hi all

I ve been playing around with markov chains for a few years now without much success. If someone

has communication with Stoopendaal can you ask him to post some of what he's doing here on LP?

I am most interested in how the final selection is made for each chain.

RL

Reply #25

MillionsWanted — Wed, 23 May 2018 21:39:09 GMT

That's cool

Reply #24

Kosh — Wed, 23 May 2018 19:10:16 GMT

Hello All,

Never mind the previous post!!!

I just did a little bit of Googliing a few minutes ago and discovered the reason why the link no longer works.

Apparently, Stoopendaal is developing a full freeware version of the Markov Chain Generator. It will be released in June 2018 for download.

By the way, Stoopendaal has put ALL of the software that he has made relating to lottery games, keno games, Pick 3 and Pick 4 that he has developed all in one place in a web... [ More ]

Reply #23

Kosh — Wed, 23 May 2018 18:48:52 GMT

The link to the Markov Chain Generator no longer works. Everytime I click on the link, I get this error message:

Server Error in '/' Application.

__________________________________________________________________________________

The Resource Cannot Be Found

Description: HTTP: 404

The Resource you are looking for (or one of its dependencies) could have been removed, had its name changed, or is temporarily unavailable. Please review the following URL and mak... [ More ]

Reply #22

JeetKuneDoLotto — Tue, 01 May 2018 02:15:36 GMT

Great thread! Good work!

Your system is quite the bump on a log. Did you define your markov property yet genius

Reply #21

MillionsWanted — Mon, 30 Apr 2018 22:34:41 GMT

Another Bump.

Reply #20

MillionsWanted — Sun, 01 Apr 2018 21:45:10 GMT

Thanks. I'll bump the thread.

I haven't had the capacity to study both Markov Chains and NNLP/Percep-Lotron at the same time.

Reply #19

lottolot — Fri, 02 Mar 2018 14:45:21 GMT

Very good, MillionsWanted.

You do not need to prove anything to anyone.

Best regards.

Reply #18

MillionsWanted — Wed, 28 Feb 2018 20:49:51 GMT

Today I didn't hit anything. 4/9 was the best. Playing ten lines, history was 40 draws.

Reply #17

flottor — Tue, 27 Feb 2018 22:47:12 GMT

I hope someone else will test it on PICK-6 and PICK-5 and give us reports.

I hit 6 of 6 and 5 of 5 a few times last week. I forgot to post the numbers I forgot to play.

No offence, but good testing is posting the combos before the draw then post the actual draw and where it can be found.

I tried the online markov generator for picki5. No success in the future.. not even in the past (should be bigger hits in the past). Promoting the link

Reply #16

MillionsWanted — Tue, 27 Feb 2018 20:03:37 GMT

My tests are finished. Today I won one 6/9 and one 5/9, which gives NOK 40,-(6/9)+10,-(5/9) = NOK 50,- win. I was playing 20 lines costing NOK 100,-.

For the next days I will play ten lines of KENO9 with new Markov predictions every day. I will use a 40 draw history. I will report if I win something. I hope someone else will test it on PICK-6 and PICK-5 and give us reports.

Reply #15

lottolot — Tue, 27 Feb 2018 18:32:24 GMT

Congrats

Reply #14

MillionsWanted — Tue, 27 Feb 2018 17:55:16 GMT

Interesting. Was it 4/6? How many draws was the history

Reply #13

carrot — Tue, 27 Feb 2018 17:20:52 GMT

Had a line wich hit 4/5, generated 5 lines only, am playing uk649s at the bookies

Reply #12

flottor — Tue, 27 Feb 2018 17:04:50 GMT

Google markov lotto programs. there are many programs but don t expect very good results. Like the neural networks the markovs don t hit big prizes. They are somehow better random generators that hit lower prizes more often than simple rngs..

Reply #11

MillionsWanted — Tue, 27 Feb 2018 01:01:12 GMT

Some success, perhaps. A few 4/8s and 5/8s are enough to cover perhaps 50-60% of the bet and make it cheaper to play every day and possibly win a larger prize sooner or later. I am trying to decide how much history is needed. Perhaps 35-40 draws are enough. And should I create new lines every day or play with the same lines for 4-5 days in a row?

I have been looking for a good predictor for KENO and the Markov Chains generator got possibilities.

Reply #10

lottolot — Tue, 27 Feb 2018 00:02:10 GMT

Hi MillionsWanted:

I m very happy because you are having a lot of sucess with Markov Chains Generator Lottery Games 2.0

Still I haven t tried it.

Perhaps,one day of these ,I ll give it a chance!.

Best regards and a lot of luck for you.

Reply #9

MillionsWanted — Mon, 26 Feb 2018 20:11:50 GMT

Also today I had a win with the Markov Chains generator, one 5/8 prize playing ten lines with a 50 draw history.

I am increasing the history to 60 draws for the KENO draw tomorrow and will play 20 lines of KENO9.

Reply #8

MillionsWanted — Sun, 25 Feb 2018 20:06:21 GMT

Another day with wins with the Markov Chain generator. I played ten lines of KENO8 with a 36 draw history. I won 2x5/8 and 1x4/8, a total of NOK 35,-. My bet was NOK 50,- for ten lines.

It was an improvement, but not enough of course, so for the next draw I am increasing the history to 50 draws.

Reply #7

MillionsWanted — Sat, 24 Feb 2018 19:51:36 GMT

This time I played 10 lines of KENO8 with a 25 draw history and won one 5/8 prize and three 4/8 prizes. That's NOK 50,- expenses and NOK 30,- prizes ((5/8) NOK 15,- and (4/8) NOK 3x5,- prizes).

I will extend the history to 36 draws and see if that helps even further on the next draw.

Reply #6

Lottotree12 — Sat, 24 Feb 2018 13:29:43 GMT

Markov Chains only works if you can mathematically check the combinations with addition and 2 others math tools I care not to discuss.

Reply #5

MillionsWanted — Sat, 24 Feb 2018 10:42:34 GMT

After having pasted in the lottery history, you pick A. Generate Markov Chains .

Next click on D. Generate Markov Chains , and then Numbers of picks to Generate , Numbers in pick and click on Only generate valid combinations (pick-5, pick-6 etc. games) and Sort Numbers .

Then click on Generate Picks .

Reply #4

Nikkicute — Sat, 24 Feb 2018 05:50:17 GMT

I don't get how to use this

Reply #3

MillionsWanted — Fri, 23 Feb 2018 21:51:15 GMT

Yesterday was a good day, considering I was only playing five lines. Won NOK 15,- and played for NOK 25,-.

Today I was using a 24 draw history playing five lines of KENO8 and won one 5/8 and one 4/8. A total of NOK 20,-.

Reply #2

MillionsWanted — Thu, 22 Feb 2018 19:44:23 GMT

Did another try today. This time five lines of KENO8. One line came up with 5/8. History this time was 23 draws.

Reply #1

MillionsWanted — Mon, 19 Feb 2018 20:05:50 GMT

Not a good result at first try. A couple of lines with 4/10. I will try to run some backtests. It will be cheaper.

Markov Chains - anyone tested the method?

MillionsWanted — Mon, 19 Feb 2018 18:41:21 GMT

Here's an online Markov Chain generator, made by Stoopendaal: Markov Chains Generator Lottery Games 2.0

I have given it a try on KENO 20/70 today, generating ten KENO10 lines with a 20 draw history.