Money is a Team Sport! Help with optimizing a semi-automatic system

Derek Moneyberg that I have been following since 2014 says: Money is a Team Sport.

Elon Musk says 'Robots will replace humans in everything. And when I say everything, I mean E-V-E-R-Y-T-H-I-N-G.

So I would like to share my system with you and if you feel excited by its mechanics then let’s embark on a fun journey to optimize it and make money together.

I have always been reluctant to share my system but the Prisoner’s Dilemma proves that in the long term it is always the best to cooperate as the Expected Value is immensely more positive.

I always get the long-term direction right but I never seem to be able to properly place stop losses and take profits or effectively manage my position mid-way like partially Take Profit, Move SL, trail stop etc and also be emotionless.

Hence, I’m implementing a robot that can be optimized to do the aforementioned in order to extract as much money as possible from the forex markets. AND it comes with detailed backtesting performance!!

So first things first; I try to identify long-mid-short term strong currencies and trade them against long-mid-short term weak currencies. I go about doing that in a few ways as means of Multi-Confirmation and Top-Down analysis!

And to stack more odds to my benefit, I start off using the COT report to copy what the professionals do and create the first currency strength meter for my system.

I go to tradingster.com where the COT report is neatly organized. I first go to Non-Commercials Strength = Longs/Shorts.

For example, this is AUD for this week 18,835/98,664=0.19090042974

The AUD has got a strength of 0.19. Apparently, if the number is less than 1 it translates that the professional traders involved in the Non-Commercial category, short the AUD more than they long it.

Anyway, we repeat the same process for all 8 main currencies for this week and we have the following currency strength readings for COT Non-Commercial:

image

I have a google sheet file where I copy in these values from tradingster and these numbers are calculated automatically to make the process of calculating these strength readings less tedious :wink:

Now, sorting these currencies in descending order we have:

image

That’s all done for the 1st currency strength meter which is based on the COT Non-Commercials. Moving on to the 2nd CSM…

And this time we throw, the Commercials in the mix. Reminding yall that Commercials are Hedgers so they bet against the direction of the trend. Therefore we need to do Short/Longs to get the CS readings…with a small twist… we will use the following formula: (NC Longs+C Shorts)/(C Longs + NC Shorts)=COT Mix CSM.

Again, the super easy-to-build google sheet formula can calculate all these readings with a simple copy-paste in the spreadsheet. And so we have another column of currency strength from strongest to weakest in descending order.

image

Not much difference at all as you can see for today between the 2 metrics. We continue with the pure pro money makers…the Asset Managers section of the COT as the 3rd CRM confirmation/filter, whatever you wanna call it…

Same process as the Non-Commercials. Asset Manager Strenth=Longs/Shorts. For example, today’s GBP for Asset Managers is 36,175/79,449=0.45532354088

Sorted:

Interesting to see that the currencies don’t fluctuate massively. Some criss-cross across the columns which is something I take as consolidation and non-tradable. Finished with the COT…

Moving on…

The infamous contrarian sentiment as our 4th CSM filter…the banks love feeding their positions off retailers stop losses which are found on the opposite (contrarian) side of their sentiment.

I use myfxbook’s community sentiment Forex Sentiment | Myfxbook. I look at a pair and consider the MINORITY sentiment as the long-term direction where the pair is going towards.

A 2:1 ratio of traders trading the EURUSD speculate that it is going to drop. But it is the minority that tends to be correct. The EURUSD will go up, but we will need to confirm across all 28 pairs.

The EUR is strong, the USD is weak. EUR gets a +1 for long and USD gets a -1 for being weak/short. That’s because when we buy a pair we long the base currency and short the quote currency. When we sell a pair we short the base currency and long the quote currency.

We repeat the process across all 28 pairs.

In the end, we will end up with something like the following:

1USD6 4CAD3 5EUR2 6GBP1 3CHF4 2JPY5 7AUD0 0NZD7

Number on the left indicates how many times the currency has been a +1, while the number on the right indicates how many times the currency has been a -1

Sorted:

Finished this 4th step but I’d like to mention that it would be way more accurate had I took in account the number of positions opened for each pair. For example, the EURUSD hover menu in pic above shows Positions as Shorts: 40109 and Longs: 19484. Therefore EUR gets 40109 Shorts and 19484 Longs while the USD is reversed, meaning 19484 Shorts and 40109 Longs. Then we repeat this process across all 28 main pairs, then sum up all Long positions for each currency, then sum up all short positions for each currency, then do the division Sum Longs/Sum Shorts for each currency and sort them differently now…in ASCENDING order because we are interested in MINORITY sentiment. However, this is too tedious for me, so waiting to get someone to accept my project proposal in order to code this one up for us from Freelancer or Fiverr.

next…

Number 5 is Dennis’ CSM in his years long-lasting thread, Trading the Trend with Strong Weak Analysis

The current look of our table so far with 5 Currency Strength readings

Another 2 confirmation columns we can add are the Intra Day and Long Term Targets on cTrader provided by Trading Central partner.

Their Long Term Targets agree with the above Top-Down analysis the overwhelming majority of the time.

So, for steps 6 and 7, I will be going about using their targets to create another currency strength reading by assigning +1 and -1 values to each currency according to the direction of these targets. Similar to what we do with contrarian sentiment except for this time we don’t do anything opposite.

Any time frame below the daily shows the intraday target for that pair and any time frame from Daily and above shows the long-term target.

This is it. 7 steps and the final table looks like this:

Now we need to extract tradable pairs. We need to find which currencies are stronger (placed above the other on the spreadsheet row) than other currencies across all 7 columns. This is a 7 step confirmation process. And so we have B for Buy and S for Sell:

5 pairs are likely to go towards the indicated direction. We have followed a process in order to stack as many odds to our side and increase the Expected Value. Now, where do I place TP, SL and what equity% risk?

1 out of these 5 extracted pairs are put in bold letters. The reason is that my Optimised robots have got open positions towards the same direction as these extracted pairs in bold.

I bought these robots from cTrader’s biggest partner regarding developing tools for cTrader called ClickAlgo. 10 robots each for 1 pair on the H1 timeframe:

AUDUSD
CADJPY
CHFJPY
EURGBP
EURJPY
EURUSD
GBPCHF
GBPUSD
USDCHF
USDJPY

When these robots have an open trade towards the direction of the extracted pair, then I activate another optimized robot on the m1. This m1 robot has been optimized during the period while the most recent trade of the H1 robot was active.

For example, the last time the GBPCHF robot had an open trade was on the 13th of January. Therefore I only optimize the EURJPY robot on the m1, only on the 13th of January. It makes sense to use the most recent open trade active period to optimize on, since the market moves in cycles, and the more recent the better!!

So I activate the GBPCHF m1 optimized robot and my SL, TP and risk are all to be managed by the optimized robot. Let the trades begin!!

I also have access to telegram groups that mirror signals from 300+ signal providers on Telegram. The vast majority are scammers but the very few good ones can give us leading signal confirmations. I will probably be adding them into the system later on. Else I will find signals from traditional sources like darwinex, Naga, myfxbook, ZuluTrade etc which come with signals performance detailed reports.

Also looking to rent a supercomputer to sharpen optimize my robots.

Meow that’s it for now, thanks for taking the time to read this post, pls tell me what you think and rest assured that I will change this first post accordingly and even the system as there is aaaaalways room for improvement…

We will help each other uncover their blind spots (psychological scotoma) and ill be very happy and eager to explain any question you might have if you are interested to become somewhat involved in whatever degree you wish with this system :smiley:. Money is a team sport and hopefully, we are about to embark on something unique :smiley:

It is 5:37am here so chat you all later :))

3 Likes

Doing very well. I feel very vibrant, optimistic and excited about this strategy’s performance!

No need to stare at screens. I filter trades the night before and activate the relevant robots to trade so there is no time and emotional investment into this sort of trading :smiley:

Current screenshots are from Spotware’s ctrader since when I started doing this strategy last month or 2, my ICMarkets ctrader couldnt open because of some sort of cache problem. Deleted cache as per the instructions given by official ctrader support on Telegram and now icMarkets ctrader is running again normally. I will stop trading on the above spotware ctrader account and get the robots to trade in the icmarkets ctrader account.

Definitely on a 1:500 leverage so I dont have any ball busting margin alerts all the tim as i will be using large lots that translate to 20% drawdown re-adjusted to what the optimization results return which are usually around 4% so a x5 to get the 20% drawdown.

The University of Birmingham is accepting external users which is a beautifully inspirting place to trade. Back in a uni library since March 2020 when i covid wouldnt allow me to use the University of Cyprus library, the epitome where I was HEAVILY immersed into the trading world and business in general, spending on average 14 hours every single effing day in there skipping parties and russian girls ■■■■ to make some fat money that translates to freedom!!!

Embark this journey, dream has been re-ignited!

Last night’s pairs extraction analysis. I only trade the following pairs since this is what the robots are made for:

AUDUSD
CADJPY
CHFJPY
EURGBP
EURJPY
EURUSD
GBPCHF
GBPUSD
USDCHF
USDJPY

Same cBot pairs extracted as yesterday… let’s see which H1 cBots have open trades right now

None…so we dont activate any 1min cbots…let’s wait for tomorrow :slight_smile:

Today’s results…i had a last trade of the day hard knock out since the CADJPY tanked hard…

Interesting strategy. I think it’d be possible to code up an indicator to do most of that work for you without the copying and pasting. I might take a look at that at some point, but time is an issue.

As for the banks and stop hunting, I’m not sure that’s really a thing. Retail trader’s money isn’t going into the market, it’s going to a broker who has natural hedging (shorts vs longs) and the fact that most people lose meaning they don’t put it into the market. So how does stop hunting work?

1 Like

Thanks for the contribution brother. Automation is a magical thing and I will explain why. The brain is designed to generate ideas, not to store them. When an idea is executed then, your brain frees up mental RAM for a new idea to be generated!

As for the stop hunting or whatever, matter of fact is that there are countless contrarian sentiment articles out there that showcase that, as the majority positions rise; the price moves towards the opposite direction.

Most prominent source im aware of with plenty daily updated examples is what dailyfx posts on their website sourced in IG clients position. Here is one example:

I mean contrarian is so obvious, i wanna die

1 Like

I’m not saying contrarian isn’t a thing, and it’s crazy to see that graph where people are betting on reversals so much. I’m saying that stop hunting doesn’t make sense. The theory that big banks are doing it to find your stops and create liquidity doesn’t make sense because your money never makes it to the real market. The only people to benefit from stop hunting are the brokers, and that raises questions to their impartiality, especially when you see weird candles that only exist on one platform.

How is this strategy working overall?

Not enough trades as I have just found myself again back in a proper library accepting external visitors after Boris Johnson dropped omicron restrictions to save face and I just can’t work or focus working from home :crazy_face:

First post updated!!

At the time of posting there are no active H1 robot trades so I dont activate any m1 robots

Im thinking that if I optimize the H1 robots based on the sole criteria of Maximing Winning Trades among many others, then when H1 has an open trade and then activating the m1 robot and opening a trade, then this m1 trade would be riding a highly probable to win H1 trade, hence the m1 trade would be more likely to reach TP too and also get to have more opportunities for new trades to open!!!

Absolutely mind - blowing!

1 Like

It looks complex, but I’d be keen to trial it

4 extracted pairs to trade but the H1 robots do not have any open trades on these pairs:

I don’t think that it’s a good idea to leave these 4 pairs untraded since all stars are aligned and we possess technology that can do so optimally!

The H1 robots don’t trade these currencies from the box. I have asked the developer of ClickAlgo if I could use these robots to optimize them on other pairs and he said yes. So I have started optimizing the GBPAUD on the H1 based on the single criteria to get as many winning trades as possible!

2+ days to finish a simple Genetic |Algorithm style of optimization. Not even exhaustive. Meaning that our optimization wont be optimal lol.

Exhaustive optimization takes super long. Like 30000 hours. So I’m looking to rent a supercomputer to optimize these robots and get a far better edge in the markets!

Very little complex and easy to understand. One does a few divisions to assign strength value to the currency, sorts them then extracts them. Done!

Yo, I have been doing very well on UK Greyhounds the last few days!!

Anyway, here are tomorrow’s extractions:

GBPCHF Buy is in bold because the H1 robot got positions open towards that direction.

So I’m activating the GBPCHF m1 robot. I need to optimize it first during the period their respective H1 robot had their previous recent trade on which was on the 2-3/2/2022

My retarded Indian freelancer has been struggling to understand simple things to code up the contrarian so the automatic calculation is postponed.

Latest extraction. I re-did the TC Intra as an update

GBPCHF optimization finished:

However the GBPCHF H1 robot trade closed in profit so we cant use the m1 robot

I was having xes with a tinder girl last night so didnt post