Seuntjies DiceBot -Multi-Site, multi-strategy betting bot for dice. With Charts!

[seuntjie]

I will however, consider this feature for v3 (whenever I get the time for that....)

Ok, I hope you do. I'll see about bribing you if I get any decent winnings soon. How much would you want?

And any ETA for v3?

That really depends on the circumstances at that time.


I HOPE to have v3 done by the end of july, but i got a lot of stuff to do and the bot isn't very high on my priority list, so no, i have no ETA

[m5j0r]

@seuntjie:
Could you please explain the point of the bot?

I mean - there is NO way to ever reach an expected value >1. Why bother?

Somebody suggested doing tons of mini martingale with the smallest bet possible. But if you did that, you'd have to do it 100 million times to get some profit out of it which in turn would increase the loss chance again to the same level.
 
The only consistent thing it can do is lose faster than you can humanly.

So why bother?

Seems to be coded nicely tho.

edit: btw just tried simulation for the heck of it: the exported file doesnt equal the shown summary.. I changed "starting balance" and not even this was reflected in the file.

[Light]

@seuntjie:
Could you please explain the point of the bot?

I mean - there is NO way to ever reach an expected value >1. Why bother?

Because there are hardcore gamblers who don't care about the negative expectation and simply want to martingale. For some reason degens have a tendency to prefer to the martinfail 'system' as it gives the illusion of profits in the short term. And anyways you'd be better off posing that question to any dice gambler - why bother when no matter what you do you lose in the long term.

[f3tus]

Somebody suggested doing tons of mini martingale with the smallest bet possible. But if you did that, you'd have to do it 100 million times to get some profit out of it which in turn would increase the loss chance again to the same level.

Not if you play a modified Martingale and stop playing in time. That's why I suggested this: https://bitcointalk.org/index.php?topic=307425.msg6772063#msg6772063

Starting with 0.1 BTC and expecting 0.2 at the end will make you lose. Expecting 0.0X or 0.00X in profit is the way you play. You accumulate the profit over a couple of days (or hours depending how extreme you play) and you've got 0.2 BTC.

If you want a bigger profit faster, but with far less odds in your favour, play 10x or 7x and increase your wager y^x after n losses in a row. You might get lucky, and you might not, you might get real profit, or you might as well lose most of your BTC. Hell, go play 100x.

You've got more chances with Martingale, but it's slower.

If your initial BTC is 0.000xxxxx, it would be better to exchange them to LTC (or even Doge) so you have 0.0xxxxxxx, because this way you can play Martingale with more odds in your favour. With initial 0.000xxxxx and 0.0000000x bets, you'll lose.

[seuntjie]

@seuntjie:
Could you please explain the point of the bot?

I mean - there is NO way to ever reach an expected value >1. Why bother?

Somebody suggested doing tons of mini martingale with the smallest bet possible. But if you did that, you'd have to do it 100 million times to get some profit out of it which in turn would increase the loss chance again to the same level.
 
The only consistent thing it can do is lose faster than you can humanly.

So why bother?

Seems to be coded nicely tho.

edit: btw just tried simulation for the heck of it: the exported file doesnt equal the shown summary.. I changed "starting balance" and not even this was reflected in the file.

Some people (like me a year ago when I started the bot) doesn't understand the math behind gambling and the 1% edge. Others don't care about the edge as light suggested. Some think they have a strategy that defeats the edge, and others are just too lazy to bet themselves. It's quite thrilling to see the bot run sometimes, can be just as exhilarating as betting yourself. Also, it was/is a fun project, I learned some stuff from it.

About the simulations: the bot has problems overwriting a file for some reason, so if you updated it and tried to overwrite an older export file, it will not show the update (it does show a message that says: "failed exporting to <path>"). I know that the amount of bets shown in the export file is always 1, but the starting balance is correct, and the finishing balance at the end of the file is the same as the one in the summery (just scroll all the way down and look at the balance and profit after the last bet)

[f3tus]

Somebody made a clone and uploaded it to Github: https://github.com/EasyDice/EasyDice

The precompiled .exe steals your bitcoins 100%, somebody with a github account please report it.

[elephantas1]

Somebody made a clone and uploaded it to Github: https://github.com/EasyDice/EasyDice

The precompiled .exe steals your bitcoins 100%, somebody with a github account please report it.

Just reported it thank you for feedback i wanted to use it

[gizzig]

Hi  seuntjie

Please a quick explanation of what program to use to compile.

I want to try some changes. thanks

[gizzig]

Hi  seuntjie

Please a quick explanation of what program to use to compile.

I want to try some changes. thanks

maybe Visual Studio 2010 ?

[seuntjie]

Hi  seuntjie

Please a quick explanation of what program to use to compile.

I want to try some changes. thanks

maybe Visual Studio 2010 ?

VS2010 and newer will work

[Light]

Anyone had some success? Is it just as good as betting manually or are the risks higher do you think?
Any other gamblers tried this?

There is no difference in this betting for you and you betting yourself. All it does is automate the process of martingale - which will eventually bust you depending on how big your bet size is compared to your bankroll. I have no doubt some people will have had 'success' in the short term, but as with all gambling this is entirely luck so you may fail on your first attempt.

[gizzig]

Hi  seuntjie

Please a quick explanation of what program to use to compile.

I want to try some changes. thanks

maybe Visual Studio 2010 ?

VS2010 and newer will work

Hi
I opened the file "DiceBot.sln" with VS2010, compiled and I get these errors:


------ Build started: Project: DiceBot, Configuration: Release x86 ------
E:\Just-Dice ROBOT\seuntjie Dice Bot Source\DiceBot\cDiceBot.cs(9,7): error CS0246: The type or namespace name 'Skybound' could not be found (are you missing a using directive or an assembly reference?)
E:\Just-Dice ROBOT\seuntjie Dice Bot Source\DiceBot\cDiceBot.cs(10,7): error CS0246: The type or namespace name 'Skybound' could not be found (are you missing a using directive or an assembly reference?)
E:\Just-Dice ROBOT\seuntjie Dice Bot Source\DiceBot\cDiceBot.cs(11,7): error CS0246: The type or namespace name 'Skybound' could not be found (are you missing a using directive or an assembly reference?)
E:\Just-Dice ROBOT\seuntjie Dice Bot Source\DiceBot\Browser.cs(5,7): error CS0246: The type or namespace name 'Skybound' could not be found (are you missing a using directive or an assembly reference?)
E:\Just-Dice ROBOT\seuntjie Dice Bot Source\DiceBot\Browser.cs(6,7): error CS0246: The type or namespace name 'Skybound' could not be found (are you missing a using directive or an assembly reference?)
E:\Just-Dice ROBOT\seuntjie Dice Bot Source\DiceBot\Browser.cs(10,21): error CS0246: The type or namespace name 'GeckoWebBrowser' could not be found (are you missing a using directive or an assembly reference?)

Compile complete -- 6 errors, 0 warnings
C:\Windows\Microsoft.NET\Framework\v4.0.30319\Microsoft.Common.targets(1605,5): warning MSB3245: Could not resolve this reference. Could not locate the assembly "Skybound.Gecko". Check to make sure the assembly exists on disk. If this reference is required by your code, you may get compilation errors.
========== Build: 0 succeeded or up-to-date, 1 failed, 0 skipped ==========


Looks like something is missing, please help

Thanks

[seuntjie]

add a reference to the skybound dll file in the bin folder

[xdigital]

I wrote 50 lines of code in javascript and almost doing the same thing.

[seuntjie]

Simple martingale is really not more than 20 lines. The big body of code comes from all the modifications this bot can make to martingale and all the extra stuff it has. If you can show me 50 lines of javascript that can do everything this bot does, i will be impressed.

[Azlan]

This wont work on a Mac will it?

Any suggestions?

[seuntjie]

No it won't.

Suggestions: Get oracle virtualbox (or any other visualization software) and make a windows VM.
port/get someone to port the bot to mono.

[chaosPT]

Did dooglus allow this on his site ?

Is it save to use have any one having problem with it ?

[seuntjie]

Did dooglus allow this on his site ?

Is it save to use have any one having problem with it ?

Dooglus allows bot on his site, but he does not endorse them. My bot gets restricted by the site delays, so it does not really cause any problems for the site.


There were a few incidents where the bot withdrew or tried to withdraw to my deposit address because I forgot to clear the default settings after testing (this is fixed now). In both cases where the bot actually withdrew something, I returned the coins to their owners as soon as I could. Aside from that, I can't really think of anything.

[PacoMartin]

I wrote my own martingale bot (and since I created it, I know exactly how it works with no surprises).  On JD it got through 6900 rolls and didn't quite get to doubling up the bankroll before it hit 12 straight losses in a row and quit (to preserve the initial deposit).  No money lost overall and it was interesting watching it play, but be sure your starting bet and bankroll can withstand a very long run of losses.

12 losses in a row = 1/(.505^12) = 1 in 3634 rolls you should go bust.
14 losses in a row = 1 in 14,253
16 losses in a row = 1 in 55,890
20 losses in a row = 1 in 859,354

If you started with a 0.0001 bet and lost 20 times in a row you would end up betting 104 bitcoins to win back your original bet of 0.0001.  Betting US$13.5k to win 13 cents overall.  Hmm.  You can't martingale bet after 22 losses as the max profit won't allow it.

This is why investors love martingale.

This formula is NOT correct. Try a simple case with coin tosses
2 losses in a row = 1/(.500^2) = 1 in 4 rolls
If you simulate this problem on a spreadsheet an try it 100 or 200 times, you will see that the average number of plays before you go bust is 6.

Simulation (download file)
https://docs.google.com/spreadsheet/ccc?key=0Aor7l039-pqHdDR1SXNONV9KQW1Id3F6cGQ0bUdPblE&usp=drive_web#gid=0

Paper
https://just-dice.com/Just%20Dice%20Streak%20Bet.pdf