Trading strategy from design to production

Discussion in 'Options' started by Aquarians, Dec 18, 2020.

  1. Two important issues I want to address and the problem is they are sort of mutually exclusive:
    1) Privacy.
    2) Scaling.

    The first one, privacy. Here's a quote from a Bloomberg article: "[Some trader] has mastered the ageless rule of hedge funds: Asked about some of his successful trading ideas, he declined to elaborate. Amateurs can crowdsource. Pros keep it to themselves." ( https://www.bloomberg.com/news/articles/2020-12-16/quant-trading-platform-quantopian-closes-down )

    Assume you are producing a trading strategy that works using the tools and setup discussed previously, it's neither easy nor deterministic - you may put a lot of effort and come out empty. But happy case, you got something.

    A thing I haven't mentioned in the "strategy pipeline" is peer review. Once you got something, it's much more reliable and offers more trust if you get another (few) human(s) to check and validate your setup. Not necessary but highly recommended.

    And then the second one, scaling. Eventually this means throwing a lot of capital at your strategy. When designing strategies you want to target those that can handle a lot of capital, by the way. We're talking at least millions if not billions. And there's two ways of getting that lot of capital: all in one go or a lot of smaller pieces.

    The ideal case most circulated around in trader discussions is "use your own capital" but really, I dismiss this as irrelevant by large. Most people who can successfully design trading strategies don't have much capital of their own at all, let alone really scaling: don't be afraid to think billions, that is. It's like in retail sales, Apple didn't grow as big selling one phone to Steve Jobs :) You have to scale up to make money. So eventually you will have to either involving one big professional fund with a lot of capital or a lot of smaller retail or semi-professional traders.

    Which is where #1 kicks again. How do you "reconcile the goat with the cabbage"? If you "let the bird out of the box", you've essentially said good bye to privacy. Another party can clone your idea and scale up by themselves, leaving you out. Recursively if their approach to scaling involves "spilling the beans" to other parties on how the strategy is built, those other parties can AND WILL steal the idea and scale up themselves. And so, sooner or later you get so much capital using the "secret" strategy that it eventually stops working.

    There's no easy answer / solution to simultaneously solving #1 and #2. Definitely better to limit damage by disclosing / selling to one or a few large-capital institutional investors rather than many retail traders. But let's be honest and face the fact that institutions are also made of people. So sell to one large bank, you got the same problem with keeping the strategy private as selling to 100 retail traders. If not a worser problem, I dunno.
     
    #11     Dec 19, 2020
  2. MarkBrown

    MarkBrown

    you will fail because you think you have it all covered? whatever static pos you come up with as a system will blow up right when the markets change and often they do.

    there are so many details that have to be addressed, you are talking about a 10 year project if you did everything right still you might have missed what's right under your nose. there is always the lotto luck and you might be the one. but seldom do those who win the lotto replicate again and again what they did to win the lotto, as in successful modeling requires.

    not to discourage you at all but "the edge" dictates your fate...
     
    #12     Dec 19, 2020
    Van_der_Voort_4 likes this.
  3. Well I'll never know if I don't try.
     
    #13     Dec 19, 2020
    MarkBrown likes this.
  4. MarkBrown

    MarkBrown

    exactly why i love the micro contracts they are great for testing theory. get some skin in the game nothing like it.
     
    #14     Dec 19, 2020