I am curious to know what stage of ATS development everyone is in. I am in stage 2 and still have a long road ahead. But my forward tests are starting to look very promising. 1. just started, still actively coding and debugging 2. forward-testing and refining strategies, still paper trading, occasional live tests 3. just started live-trading with smaller positions 4. running full algorithm but still not certain about long-run consistency 5. running for at least 1 year, seeing solid consistent profits 6. ran for few months, lost money, stopped trading and redesigning
Yes, this fact did occur to me -- since no single strategy can last forever, in the ideal case, after successful forward tests, one would presumably go live for a year or so, make significant money, then start to see the algorithm falter, and then go back to steps 1 and 2 and either tweak the algorithm, or implement a new strategy that works again. I am actually working on my 5th strategy. I have been going back and forth between 1 and 2, but each time, my forward testing proved that I didn't have a consistent strategy. Fortunately, my latest one seems a lot more consistent so I am hopeful. A few more weeks, and I will try live-testing with small positions. However, I have heard people run the same algorithm for years and make consistent money. That would be like finding the holy grail...