Automating my FOREX Strategies
how long it really takes
“quote” Algorithmic trading is a form of trading that uses advanced mathematical models and historical data to make decisions about when to buy and sell stocks. While algorithmic trading can be beneficial in terms of speed and accuracy, it can also be subject to certain drawbacks. One of the main drawbacks is that trading one algo will never work because it does not take into account subjective and qualitative factors that can have an impact on the market. Additionally, algorithms may miss out on trades because the latter does not exhibit any of the signs the algorithm’s been programmed to look for. Furthermore, when the market gets shaky, algorithms can (and do) pull out, which can lead to “flash crashes” and sudden drops in the securities market. “end quote”[1]
To date, no one (to the best of my knowledge) has been able to trade one algo that works in all market conditions. Not even the algo’s running on the New York, Chicago, Tokyo and London exchanges have been able to achieve this.
Traders who are transitioning to automated trading soon learn through simulations that their system is a failing system without all the variables included in their system they realise they need to state, when they finally make that move to automation.
Here’s some of a conversation I had with a friend (an “ex broker” here in Melbourne Australia.
Ben: “How can I trade a manual strategy successfully, and not make money when I build it and simulate it”?
Me: “possibly you’re not taking into consideration when you take profits or cut off losing trades”
Ben: “what do you mean”
Me: “These decisions require the model/algo to be stated or the model wont know to do that. You brain is working so fast with so many variables considered that it instinctively makes these decisions without you even recognising or acknowledging the decision making process, again, because it’s so natural, you don’t realise It’s happening. Sometimes you take profits early, sometimes you let it run in better market conditions, other times you close out loss immediately and other times, you let the loss run! Right”?
Ben: “Yea, I guess so”
Me: “Well, all of those decisions need to be pre stated in the model, you can’t just take 50 points profit or wait till the bar close [1] rises above the 70/80 RSI. It’s not that simple”
I know, I know, it sounds like rubbish right! I can tell you, It happened and it’s not the first time!
My point is, you can measure so much more automating, than you can manual trading. I consider automation a much fairer game when trading FX. But hey…That’s just my opinion and we all trade differently, right? Again, I’ll never state which one is better than the other, all though I strongly believe automation is the only way forward. It’s up to everyone to make up their own mind and I respect all opinions.
All of these decisions require the model algo to be stated in it’s conditions.
Listen, I have no doubt that there are plenty of manual or discretionary traders who make money. What I’m saying is, “lets be honest, if It’s really is a winning strategy, why wouldn’t you automate it! It’s really that simple! Run it on everything you can, on every time frame you can? Because It would be absolutely NUTS NOT TO right? That ladies and gentleman, is called “The Automated Journey”, and to do it properly, takes years!
You’ll soon figure out that it’s simply easier to keep doing what’s making you money now than it is to automate it. Well, that’s what I’m usually told anyway.
The first thing you’ll learn when automating, is the sheer amount of conditions that occur in your system that you take for granted, because of the earlier mentioned, you instinctively just do it, that require automating.
If you do choose this journey, Strap yourself in, learn to code or pay for software allowing you to code, and get ready for years of troubleshooting and hundreds of hours of simulating!
Surround yourself in a community of algo traders who encourage you. Share your thoughts and idea’s with others and realise that a few hundred a month, really isn’t that much to pay for software, or whatever it is, that allows you to automate your strategies!
One model, It’s not going to cut it! Trust me. I have built over 1000 models and hundreds of hours of simulations to get to where I am and If I wasn’t so passionate about automation, I don’t think I would have continued.
I my humble opinion, anyone who tells you automation is easy…I would be sceptical why they’re presenting this challenge to appear so easy.
I know beyond any doubt, you would have to offer me a damn lot of money to use one of my portfolios.
If you think the algo’s mentioned earlier on the large exchanges don’t get turned off in certain or specific market conditions, you are wrong!
I have heard a lot of traders say to me “Any automated system you trade, I can trade manually” and they soon change their mind after they see a snippet of what my automated strategies are. The fact is, you would have to be “John Nash” to keep up with some of the calculations model do while they’re trading.
In conclusion, trading one algo has never worked and will never work.
Get used to portfolio’s trading if you’re going down the automated pathway. It’s the only way forward!
Sources: https://theconversation.com/explainer-the-good-the-bad-and-the-ugly-of-algorithmic-trading-68477
