How Hard is It to Develop a Viable Algorithmic Trading System? (Part 2)

I left off with the idea that trading strategies in the public domain will be quickly developed and traded by institutions until the Edge runs out.

While I think this sounds good, it is only speculation. As with so many claims about investing and trading, the only way to know for certain would be to develop a large number of trading systems in the public domain and to trade them live for a period of time. Only then would I have the data necessary to collate the returns, to analyze the results, and to use statistical testing in determining whether Edge exists and for how long Edge persists. This could take a lifetime of work and many millions of dollars in trading capital to test.

Not gonna happen! Therefore, I can never truly evaluate veracity of the claim.

In the world of trading and investing, I believe ideas that sound good do have persuasive impact because many traders lack critical thinking skills. For this reason, I think traders perceive the words of apparent authority figures and trading gurus to be meaningful and true when in fact they are little more than good advertising pitches and marketing claims.

As mentioned above, though, I also believe I cannot know this for sure. The only way to evaluate the latter claim would be to interview a large sample of traders to determine what it takes for them to believe “good-sounding ideas” and to determine how much critical evaluation is involved. Just the task of operationally defining much of this (e.g. what it takes to believe, what constitutes a good-sounding idea, a scale for critical evaluation) strikes me as an extremely complex task that may or may not even be possible to carry through.

In case you’re keeping score, we have here a claim that cannot be evaluated for reasons that themselves cannot be evaluated.

I resign for the night before I get too confused.