EC Money / JL Challenge

[Skakus]

No offence taken. I was playing on the spike/cheese thing.

I can't speak for spike but I think he meant that he could select different bets for identical sets of spins because his main goal is to reach +5 or +6 units per session and the chance of picking the right bet is always 50/50 so don't sweat it if you pick wrong or pick different, as long as you get to +5 or 6.

[Bayes]

No offence taken. I was playing on the spike/cheese thing.

D'oh! I missed that. 

Thing is, he lead people a merry dance for a long time, and EDAM well knows it. 

[superman]

Would you be interested in working on a bot at some time in the future? >>> it won’t be easy.

Nothing you sent me in the past was easy either LOL I still have some front ends to rebuild since my HDD crash but that wont stop me as I already have BVNZ up and running, also have the file feeder redone to so can import any txt file spins, so yes Danny, you know how to get me on email.

[Tomla021]

so your just using 12345678910 and going back to beginning on a bustout of 100 plus? but skakus doesnt use a classic on this he sometimes goes up or down? Im assuming straight up on a win or loss till plus 5

[Skakus]

Thing is, he lead people a merry dance for a long time, and EDAM well knows it. 

Touche' 

[Skakus]

For what its worth I'm now at 1852 uints


EC Bets 5237
Win      2681
Loss     2556
Flat Bet +125
z-score 3.76


Previous update was 9 days ago. I've made 330 bets since then and have only managed to win another 30 units. That's because I"ve had several 108 unit busts in that time.

Over the whole test I would think I've had more than a dozen, and may be even 20 or so busts. I wish my system would never lose like JL's because then I'd have about 3800 units instead of 1800 units. 

My flat bet score has gone down 2 units over the last 330 bets, and the z-score dropped away slightly, but has picked back up to where it was.


I am thinking very hard about doubling all future bets, so I have attached all my results to this point.

If you go to the result,

08/15/2012
LOW      7   W   1300

That is the point where I gave JL the online program and started betting with the encrypted program. There is a 68 unit discrepancy at this point because Bayes sent me the new program with the current balance, but I played on for a little while before I got it. I reached 1300 units by the time I got the new software so I have just been adding 68 units to the total as I go.

Cheers.

[Skakus]

so your just using 12345678910 and going back to beginning on a bustout of 100 plus? but skakus doesn't use a classic on this he sometimes goes up or down? I'm assuming straight up on a win or loss till plus 5

I’m using 1,2,3,4,5,6,7,8,9,10,11,12,13.. etc, up and down until I win 12 units or lose 108 units.

The second bet of every new session is +1 by default, because I follow the wL registry to decide if the bet goes up or down.

If W is leading then W is +1 and L is -1.
If L is leading then L is +1 and W is -1.

I do this until either W or L are leading by 4, then I hold the next bet at the same level.
If W or L are leading by 5, then I hold the next bet at the same level and also dump the wL registry and start a new one.
Then I go +1 by default on the next bet, then go +1 or -1 depending on which is leading the wL registry.

Whenever I win more than 12 units I start the next session with the extra chips.
Whenever I’m close to losing 108 units I will reduce the next bet so that if it loses I don’t go over 108 units, then if it loses will start the next session with any extra chips saved from the previous session. If the reduced bet wins then I will continue on at that reduced level going +1 or -1 as per wL registry.

That’s the whole shebang for the MM.

Example:

1 w1 +1  w leads by 1
2 w2 +1  w leads by 2
3 L1 -1   w leads by 1
2 L2 hold  wL equal so next bet hold
2 L3 +1  L leads by 1
3 L4 +1 L leads by 2
4 L5 +1 L leads by 3
5 L6 hold L leads by 4 so next bet hold
5 L7 hold&dump L leads by 5 so next bet hold and wL registry dumped
5 w1 +1 w leads by 1
6 w2 +1 w leads by 2
7 w3 +1 w leads by 3
8 L1 -1 w leads by 2
7 w4 +1 w leads by 3
8 w5 hold w leads by 4 so next bet hold
8 L2 -1 w leads by 3
7 L3 -1 w leads by 2
6 w6 +1 w leads by 3
7 w7 hold w leads by 4 so next bet hold
7 w8 hold&dump w leads by 5 so next bet hold and wL registry dumped
7 L1 +1 L leads by 1
8 w1 hold wL equal so hold next bet
8 w2 +1 w leads by 1
9 w3 +1 w leads by 2

So on and so forth until I win 12+ units or lose 108 units.     

[Still]

@Skakus
Very good! And thanks for the methodology. 
I think this is worth charting. 
How can i obtain the data so that i can chart it in Excel?
Thanks!

[Skakus]

Hi Still,

Is the attached .txt file no good for charting?

[Still]

Oops I didn't see that.  Yes thats fine.  Chart should be up shortly.

[Still]

This looks really good! I offset the initial 300 unit bankroll to start the chart at zero, and show only PROFIT.  After a few wicked drawdowns, it looks like Skakus found his legs at around bet #2800 and again at #4000.   @Skakus: are these about the times you made adjustments in your approach?

[attachimg=1]

[Skakus]

Hi Still.

Thanks for the chart, yes it does look good.   

Yes, around 2800 I got sick of flat lining so I made the changes. From the data I had accumulated the changes were necessary, and obvious.

Not sure about the 4000 benchmark, but I suspect it could have something to do with the guessing component of my play. Up to then I was infrequently guessing whether or not to bet, and even when to bet opposite the selections. Around the 4000 mark I felt the guessing part was getting too inflated, and I would dearly love this system to work on a purely mechanical basis so I made a conscious effort to minimize the guessing part since then.

It is interesting to see in the chart that the guessing component of my play might have actually been helping more than I thought.

Cheers, and thanks again for your input.   

[superman]

Good job Skakus, was hoping to see the z score move up closer to 4 but as you had the bunch of losses, getting back to where you were is still an achievment, well done buddy

[Skakus]

Thanks superduper!

I've had 2 more losses today since posting so a z-score of 4 is still a ways off! 

[Tomla021]

Thanks for explaining the MM ---Pretty interesting and now I have an idea
much appreciated Skakus