EC Money / JL Challenge

[Skakus]

For what its worth I'm now at 1517 uints.

EC Bets 4498
Win      2300
Loss     2198
Flat Bet +102
z-score 3.4   

[Skakus]

For what its worth I'm now at 1659 uints 

EC Bets 4708
Win      2411
Loss     2297
Flat Bet +114
z-score 3.58 

I'm getting stronger!

[TwoCatSam]

Good work, Skakus!

[Skakus]

Good work, Skakus!

Thanks Sam.


For what its worth I'm now at 1822 uints 

EC Bets 4907
Win      2517
Loss     2390
Flat Bet +127
z-score 3.77 

[Bayes]

 

[superman]

z-score 3.77

Almost there mate, good luck for the other .23

[Tomla021]

That Skakus and his magic chart--go for it

[Wally Gator]

If it shows enough promise as the test matures then I will explain the process. If not, then I will shove it where the sun don't shine!

Is your post at #52 the basics for your success thus far?

[superman]

Just curious Skakus, are playing strict mechanical rules or varying your selection on what's happening?

[Bayes]

I wondered the same thing, he told me not strictly mechanical but some "educated guessing" a la Spike. 

[Skakus]

Is your post at #52 the basics for your success thus far?

Hi Wally G.

One of my favourite childhood cartoons, by the way.

The bet selection is the same, the progression is the same, but I dropped the stages and now just play the 1 level.

The stages explained in post #52 were too constricted, and were costing me too many lost sessions. I wasn’t making any headway because I wasn’t giving the system enough chance to grind out of many short negative fluctuations.

How I’m playing now is probably still not ideal, but it’s much better than before.

I’m now resetting the game on +12 or -108 units, whichever comes first. I would have increased the unit size a while ago because I only need a 320 unit bank for this level of play, but the 50 unit per bet challenge limit has restricted any proportionate stake increase. Not to worry because the bankroll is not so important to me anyway.

One other mod I’ve introduced is to begin each new session with any extra chips won or saved from the previous session. So I don’t always start a new session with 1 unit. If I won 18 units in the previous session I will begin the next session with 6 units. If I’m losing 100 units and the next bet is 16 units I will reduce the bet to 8 units. If I lose the next bet I will end the game at -108 and begin the next session with the 8 units saved from the previous loss.


Cheers.   

[Robeenhuut]

Hi Wally G.

One of my favourite childhood cartoons, by the way.

The bet selection is the same, the progression is the same, but I dropped the stages and now just play the 1 level.

The stages explained in post #52 were too constricted, and were costing me too many lost sessions. I wasn’t making any headway because I wasn’t giving the system enough chance to grind out of many short negative fluctuations.

How I’m playing now is probably still not ideal, but it’s much better than before.

I’m now resetting the game on +12 or -108 units, whichever comes first. I would have increased the unit size a while ago because I only need a 320 unit bank for this level of play, but the 50 unit per bet challenge limit has restricted any proportionate stake increase. Not to worry because the bankroll is not so important to me anyway.

One other mod I’ve introduced is to begin each new session with any extra chips won or saved from the previous session. So I don’t always start a new session with 1 unit. If I won 18 units in the previous session I will begin the next session with 6 units. If I’m losing 100 units and the next bet is 16 units I will reduce the bet to 8 units. If I lose the next bet I will end the game at -108 and begin the next session with the 8 units saved from the previous loss.


Cheers.

I like your staking plan. Keep it up 

[Skakus]

Just curious Skakus, are playing strict mechanical rules or varying your selection on what's happening?

I’m playing strict mechanical rules with a strict money management strategy.

The MM is 100% strict, and the bet selection is 90-95% strict.

The bet selections come from two separate systems. The first system is the dominant selector. The second system is for making bets when the first system is in downtime.

Whenever the two systems select an opposite EC I don’t bet. Whenever this happens and I don’t bet but the dominant system would have won, I will go into an educated guess mode for the next bet/bets, which could mean skipping over one or more selections, and very rarely betting opposite one or more selections.

The educated guessing component is very minimal and probably complete bollocks anyway, it’s just a feel good thing I think, so the mechanical aspects of this system could likely be coded 100% with the results remaining virtually unaffected.

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

[Skakus]

>>> not strictly mechanical but some "educated guessing" a la Spike. 

That's very cheesy.

[Bayes]

That's very cheesy.

No offence meant, I only said it because Spike coined the phrase. In fact, it doesn't really say what he intended it to mean (that he uses no mechanical system). You could say that in a sense many mechanical systems are educated guessing, and the rest aren't "educated" by anything. 

The idea of educated guessing, in the sense he intended it to mean, is incoherent. What Spike seems to be saying is that he uses past spins to inform his bets, but that there are no rules in which to interpret that history, and no triggers are possible (because it's all random). For example, he often said if he came across exactly the same set of spins as in a previous session, he wouldn't bet the same way. But if there are no rules which can tell you which past spins indicate future spins (and what those future spins are likely to be), then it would be impossible to get any advantage, much less the 72% win rate which he claims.