Hello, I have a question to the trader who used to trade with self made EA that is coded 100% realisticly and was tested. What was your expierience trading EA in real time, were the results similar to the tested ones? I am not talking about scalping or speed trading, just normal trading with specific strategy. Is there anything you would recommend to someone who want to start trading on real market with self made EA?
Not at all. Cause backtesting is completely irrelevant to forward looking realtime tracking. The market do always change so you need to adapt. Best is discretionary trading as there is no flexibility in systematic rules. You cannot expect very good fine systematic strategies that make you rich in no time. There are some few very mediocre systematic strategies working around Sharpe 1 to 2. So you would need around 20 years or so to get any rich from it starting with $10k or so. And then there can happen so many things in those decades that is very hard to trust and rely them for such a long time too. Better you forget systematic trading and start to trade as discretionary as you can. And I know what I am talking about I spent over 10 years on systematic research and built "strategies" with Sharpe 14+ (based on the past). It does not mean that you can only have failure. But mostly when you can make a decent amount of money in short time you have a lot of luck and the current market works with the given systematic strategy.I turned from a fully systematic to a discretionary trader.
finding the edge is almost impossible. programming the edge is even harder. overcoming the hardware, software, data, connectivity are things that await you if you get this far. it's a daunting task, but not if you are committed to tackling and perfecting each phase with tenacity and diligence. once you have done all this then it becomes much easier i think. lesson is there are no short cut's, if you go that route you will end up on this forum for decades talking shit and becoming bitter... at anyone you perceive as doing better than you.
Thanks for reply, I have a next question that refer to your answear. Why sharpe ratio is such important metric? It is calculated based on not critical factors and personally I dont understand why is that? Sharpe Ratio=Rp−Rf / σp Where: Rp is the average return of the portfolio (or trading strategy) over a given period. Rf is the risk-free rate (the return on a theoretically risk-free asset, often approximated by the return on government bonds, e.g., U.S. Treasury Bills). σp is the standard deviation of the portfolio's returns (this measures the volatility or risk).
Yes, I agree that finding an edge is almost impossible, but now I don't believe that from a technical, realistic point of view my strategies work quite well. I would like to ask how flexible you think I can be when choosing specific values as criteria that I will follow in my strategies. What do you think as a man with more experience, how picky can I be when changing values (criteria) in my strategy to achieve ideal results.
Investor here...Not a trader. Wouldn't market markers work against any tested system. I believe it would NEVER be realistic since you are not only working against MM, but your own broker and the HFT distant advantage they have/maintain...
Why trading and being profitable is working against your broker? As long as broker is not MM your profits are other's losses.
An AI answer... Yes, a broker can also act as a market maker, meaning they can provide liquidity and pricing for securities by trading for their own account, becoming the counterparty to customer trades. Here's a more detailed explanation: Broker: A broker acts as a middleman, facilitating the buying and selling of securities for investors, usually on an exchange. Market Maker: A market maker provides liquidity and pricing for securities by trading for their own account, becoming the counterparty to customer trades. Broker as Market Maker: Some brokerage firms are also market makers, meaning they trade securities using their own capital at prices they set themselves. In essence, a market maker broker creates a price feed and liquidity for their clients to trade, rather than simply passing on the price and liquidity from the underlying market . Example: A broker could act as a market maker for a specific ETF, providing liquidity and pricing for that ETF in the secondary market.
So let's say that you have about 100 profitable settings when you run an optimization test. You shouldn't take the very best settings. You should come down on this scale to about the 75 per cent cluster of settings and use that. The idea is that if you use the optimal settings, it won't hold up over time But if the model still makes money at 75% of its optimal settings, then it has a chance to probably endure and hold up over the long term.
Took me 20 years to understand edge and yet was profitable for last 12 years, that was 2008. Once you figure out what "your edge" is, they are popping all around you. I never use the top/bottom 10% of backtesting. Then I check out Internals also each trade, if there is huge gaps whether huge profits/losses, what do I have to do to not take them. Am looking for consistency.