Peaky on Steroids

[alecsa72]

V 1k is in post 1, with the fix for incorrect lots per dollop of cash calculation that I described in the SPB thread.

Turn off "Match whole word only" if you want to DIY to find the 5 double declarations that need changing to int's. Ignore the 'possible loss of data' warnings when you compile - they are irrelevant.

Steve

small error in "Gertje's portion" of the EA version 1K

This input parameter doesn't exist anywhere in the rest of the code

Code: Select all

extern int     Group3SixthsMinimumPips=450;//Note the change to a minimum. Sixth must be <= this

but this portion probably should be using that input parameter - first line should be using "Group3SixthsMinimumPips" instead

Code: Select all

 Group3SixthsMinimum=Group3SixthsMinimum;
   Group3MarketDistance = Group3MarketDistancePips;
   Group3RecoveryProfit = Group3RecoveryProfitPips;

[SteveHopwood]

Steve

small error in "Gertje's portion" of the EA version 1K

This input parameter doesn't exist anywhere in the rest of the code

Code: Select all

extern int     Group3SixthsMinimumPips=450;//Note the change to a minimum. Sixth must be <= this

but this portion probably should be using that input parameter - first line should be using "Group3SixthsMinimumPips" instead

Code: Select all

 Group3SixthsMinimum=Group3SixthsMinimum;
   Group3MarketDistance = Group3MarketDistancePips;
   Group3RecoveryProfit = Group3RecoveryProfitPips;

Great spot. Thanks very much.

Fixed in 1l in post 1. The DIY could not be simpler. Do a search for "Group3SixthsMinimum = Group3SixthsMinimum;". Change it to:
Group3SixthsMinimum = Group3SixthsMinimumPips;

[Wavegarrick]

Hi Steve,

OK, so we have to refine our trading but we are light years away from where we were in 2011 at the inception of this forum. What next?

I agree whole heartedly. Many minds are pulling together, we are progressing at a rapid pace. I find it hard to keep up, but oh boy, this is good More importantly is that I am seeing new kids in the block with great ideas and it seems that you are finally getting your coding skills through to many.

Many thanks for all you are doing Steve.

Cheers
Leon.

[SteveHopwood]

Hi Steve,

I agree whole heartedly. Many minds are pulling together, we are progressing at a rapid pace. I find it hard to keep up, but oh boy, this is good More importantly is that I am seeing new kids on the block with great ideas and it seems that you are finally getting your coding skills through to many.

Many thanks for all you are doing Steve.

Cheers
Leon.

My pleasure Leon. We are all gaining from this collective effort.

[SteveHopwood]

Referring to Leo's post just above, here is another thought - or collection of thoughts. Not sure which yet as I am writing this as a 'stream of consciousness' type of thingy.

The latest members of various families of EA's are proving profitable when trading live. The latest members of some of the families are coded by members building upon my work (Awesome+ etc). My work in turn built upon the work of all the fantastic traders and coders who have contributed so selflessly over recent years. It may stun Old Stagers to be reminded that SHF is over seven years old. Time flies.

I have been bored recently for the first time in a decade. I have had little to do. I coded a 'private' EA over the weekend. I have fixed the lots-per-dollop thingy. Mind, I only did that because I had nothing else to do.

What passes for my brain has not been churning automated trading ideas for the first time for a decade. Maybe we are finally winning? Perhaps we finally have a huge legacy to leave to our descendants? Fingers crossed, OK?

Maybe we have finally reached

No guarantees because us Old Stagers know what total bastards the markets are, but maybe we are finally starting to see the end of a war we can win.

[Gertje]

Which makes me wonder......
Would recent results be scalable?
Would trading full lots be as smouth as trading 0,01 - 0,1 lots?
Would one have the balls to see 100K DD using Empty4 during NFP?

Time will tell, that’s for sure.

[SteveHopwood]

Which makes me wonder......
Would recent results be scalable?
Would trading full lots be as smouth as trading 0,01 - 0,1 lots?
Would one have the balls to see 100K DD using Empty4 during NFP?

Time will tell, that’s for sure.

No, but nor would it need to be. In early times, we recognise that high risks = high rewards. Fine, so long as we are playing with stakes we can afford to lose. Gareth's $461 account is 25% into DD. It blows up? I don't care, refund it and start again.

PJ's Cambodia World Challenge fund going up in smoke? Don't care. It could cost me big time 12 months from now. I could be coming on here requesting donations. I could cope.

Would I be happy to see a !00K account go into 25% DD? Best not to have that nightmare.

Nope. In the early stages we increase our lot sizes to take account of the increasing account size. So where do we stop doing so and accept lower profits per trade as the down side of of faster overall profits?

I suppose another way of positing this question is: when do we stop being greedy?

All of a sudden, we have a shoal of EA's trading successfully on live trading accounts. A few hours ago, I was contemplating that I had no EA coding to do. This was the first time that I had no coding to do in a decade. I was regarding my utter lack of doing anything useful here.

Suddenly this question occurs to me: assuming our activities really result in wealth, then what do we do with it? Apart from paying the bills.

Hey ho. Let us hope this is a question that really needs answering.

[florin227]

A bit more than 9 weeks and I have amazing results on live trading with Gertje setfiles on fully auto no manual intervention. As I noticed from Gertje's and my tests "Fill the gap inputs" generate most drawdown when you have several trades on same pair going against you, so I now use minimum MDP of 300. I will soon start another live account using last version of POS with 6th as MDP.

[florin227]

SteveHopwood » Tue May 29, 2018 11:30 pm"

Nope. In the early stages we increase our lot sizes to take account of the increasing account size. So where do we stop doing so and accept lower profits per trade as the down side of of faster overall profits?

I suppose another way of positing this question is: when do we stop being greedy?

After blowing up so many accounts in the last 10 years several of them grown from 2-300 USD to 15k all because of greed I discovered patience when I discovered SHF. I still underfund my accounts when I start but I try to follow a simple rule to overcome my natural greed: when the account grows 4 times i double lotsize so as the account grows and that money start to matter I decrease the risk until I reach a decent maximum DD and start using lots/dollopofcash. Of course I do not recommend underfunding ( GREED ).

Suddenly this question occurs to me: assuming our activities really result in wealth, then what do we do with it? Apart from paying the bills.

Hey ho. Let us hope this is a question that really needs answering.

When we're getting close to that wealth we will have to recode all EA's to split order filling into smaller orders or our thousands lots orders will get too much unwanted attention

[renexxxx]

Maybe we have finally reached

No guarantees because us Old Stagers know what total bastards the markets are, but maybe we are finally starting to see the end of a war we can win.

Steve's conclusion that we are entering a new era, where we have actually a few well performing trading bots is also reflected on https://fxtestify.com/. A quick glance and we see several bots in positive territory. A better look and we see very attractive looking equity/balance curves for Awesome SS, Peaky-on-Steroids, Slopey Griddy Bob and Slopey Peaky Bob. When Thomas and myself started this exercise (some three months ago), almost all bots were losing money.

Now that I've got your attention, I would like to, once more, call for your participation. The truth is this: on this forum (and many other forex forums) 95% of the content (code, new ideas, manual strategies etc.) is provided by 5% of the members. The majority of members are gratefully benefiting from what the minor few are contributing. Reasons for this inactivity can be anything from laziness to inexperience or shyness. Now, we've got a role for you: you can contribute to the constantly growing central database of test results, just by running a bot, with a set of parameters configured by yourself (and preferably not already chosen by someone else) and uploading your results to fxtestify. What do you gain by that? You will be given an access token that unlocks detailed statistics of all bots in the database and it also gives you access to all .set files being tested (notably the one that is doing really well). What do we (as a community) gain by that? The law or large numbers will be at work and it will tell us which bot, configured with what parameters is the most successful under the current market conditions. A TradeReport or TradeChart from fxtestify is far more reliable, than an individual TradeReport or TradeChart (you see sometimes on this forum), as it is supported by the results of many testers.

If anything mentioned here strikes a chord with any of you, and you wish to contribute to the central repository of test results, please visit https://fxtestify.com/testers to find out what you need to do to get started. Or you can wait for this weekend's announcement in the Tests Database Announcements thread, as Thomas will present version 3.0 of the Trade2MySQLConnector that will generate an authorization key on the fly, if one is not presented. That will make the learning curve for new comers even simpler.

René and Thomas.