I’ve been in the trading space for a while and never came across quantitative trading. Recently, I found a very interesting (free-cracked) course about the importance of knowing your system and backtesting it thoroughly.
What people see as backtesting is going back in Tradingview a few months and if your trading strategy worked, they go for paper trading or live trading.
I backtested poorly one strategy and got good results. When I went to trade live, I found a lot of situations I didn’t even consider, and doubts about how to proceed in those scenarios. I know this happens to a lot of traders, they ask questions to others like “why do you think my system is not working”.
In this quantitative course, they promote backtesting at least 1000 manual events with the same rules (as mechanical as possible) and then you get a lot of information about your system. It doesn’t guarantee future results but it considers all market conditions until now. Also they talk a lot about backtesting different combinations of TP, management, SL and risk, so you can find the best combination for your profile (more benefits or less drawdown)
I just feel like nobody wants to talk about this. I haven’t meet anyone who consider backtesting as an important tool to find out if you actually have an edge or just had an exceptional year (and backtesting with good quality data)
I’ve seen many people talking about how they had systems that worked one year and that year was the one they backtested it.
But this quantitative approach almost always comes with technical indicators (RSI, MACD, etc etc). I trade ICT concepts and can’t find a single person who wants or at least tries to use a quantitative, statistical approach to their trading systems based on smart money or ICT.
People say mindset and psychology is the holy grail, but how can you trust your strategy? How can you say: I will stick to this because this works?
Hi. I personally start testing my strategy in a demo account and after a period of one month I check how much my PNL was; if the PNL satisfies me, I try the same strategy in a live account.
But about this part, let me tell you one thing to clarify what it really means. when I first started trading real account, my strategy was good and I could get more than 60% winning ratio, I didnt have a big account and based on my risk management , I had to set my SL on -0.5USD and my TP on 1 USD. After several months I checked my account, i had made around 50 trades, only 5 of them hit SL and the rest hit TP, but my account didnt grow so much, because my risk management didnt let me enter positions with big Volume. So, I thought, ok, I can predict with high accuracy, so I dont have to follow risk management. I started trading with - 5$ SL and 10$ TP. I can clearly remember, I didnt hit even 1 TP after that! not even 1! and soon my account was 0 balance!
What I mean is that psychology and mindset is not only about when you lose! you have to control your pride as well as your anger and sadness.
I agree with you. But for me, knowing that I spent so much time backtesting my system and creating equity curves, makes me follow my rules strictly.
I’ve been so disciplined last month that trading is now boring for me. I’ve been sticking to my rules and ignoring my emotions and my gut feeling. I feel like a monk of trading…but now that I got that part dominated, I feel fear of having a strategy that won’t work in the long term.
Yes, my system had 58% winning rate in 2022 with a great return. But …is that relevant? How can I know it actually worked in the past if I don’t backtesting hugely.
Believe it or not this is the proof that you have been trading logically. trading is a job its not a hobby, it has to be boring lol.
Steak to a logical risk and capital management system and even you have the worst trading strategy, you will have time to make a comeback, but even with the best trading strategy and a faulty risk management system, your account will sooner or latter hit 0 Balance.
That’s exactly the thing. I use price action and smart money concepts by I try to remove all the discretionary aspect of it, trying to make it as systematic as possible. We define our model strictly and only trade in the exact same conditions.
For example: Define 1D trend. If bullish, find OB and FVG below 50% of the last bullish impulse. Here trend, OB and FVG and the impulse must be well strictly defined. Then wait for a MSS in LTF. Need to be strictly defined. Yes, you miss trades that don’t follow the exact characteristics of your system…but you also remove a lot of discretion.
Personally I use 15m liquidity sweep +1m ms and FVG. Very mechanical. I can’t trade in a discretionary way.
I use Ichimoku cloud Daily trend as a starting point as a back tester that has been proven for decades. For a five day trading week I’ve tweaked the settings to 8-22-44. It tells me all I need to know about whether a trade is suitable.
However, the trend must be in accordance with the Daily, 4hr and 1 hr chart trend like 3 ducks in a row for a positive probability of making a successful trade…
From the very long-term back-testing I’ve seen on Youtube it seems that the common feature of logically devleoped TA-based stratregies is that their performance cycles up and down over many months. They might work really well on a single pair for 6 months or 9 months or 15 months, then their performance on that pair starts to drop. But the performance of another pair is cycling upwards. And the performance of the first pair will eventually recover. And so on.
It seems that the rule is good strategies don’t stop working.
But they are abandoned when their performance decays, and traders post something that says the EUR/USD XYZ strategy has stopped working. And they explain thisby referring to HFT, the strategy became too well known, the brokers are trading against the strategy, the banks have got more aggressive, etc. I think they’re not right.
