anyone tried to reverse engineer strategies on collective2?

Discussion in 'Automated Trading' started by trend2009, Oct 31, 2019.

  1. trend2009,

    I have a very important question for you.

    Why not spend the time developing your trading skills through mentorship/coaching, purchasing quality trading courses online, reading quality books, staring at charts, sim trading, asking questions on the forum? In another words, seeking knowledge and skill set.
     
    #11     Nov 1, 2019
  2. It is a good analogy because the quality of trading strategies on C2 (and similar sites) is likely to be very poor. Reasons include:

    • easily gamed
    • poor choice of metrics for leaderboard, incentivising ridiculous levels of leverage
    • likely to be overfitted
    • likely to be produced by people who don't know what they are doing
    • unrealistic results without proper accounting for interest payments, commissions, slippage and so on
    GAT
     
    #12     Nov 1, 2019
    jl1575 and SteveH like this.
  3. bbpp

    bbpp

    Clueless.
     
    #13     Nov 1, 2019
    sysdevel99 likes this.

  4. the big difference is that strategy on c2 has statistics on live trading records. can you find sharpe ratio, profit factor, dd on strategies printed on a book or lectured in a trading-mentor class? if you do not have such records, how do you have idea if the strategy is good or bad?

    even though most of strategies on c2 are garbage, but there are some gems out there if you look into deeper and harder.
     
    #14     Nov 1, 2019
    SimpleMeLike likes this.
  5. Since you didn't like my first response here's another. Generally it's a waste of time to reverse engineer trading strategies, regardless of their source. At best you are going to be able to work out what risk factors they are exposed to. But if they have any genuine alpha then you are unlikely to work out where it's coming from.

    GAT
     
    #15     Nov 1, 2019
  6. Pekelo

    Pekelo

    Have you driven one? It was actually NOT a terrible car, but you wouldn't know it because you watched the spin off from the Grand Tour guy.

    " The keys to its success were said to be its competitive price, reliability, simple DIY-friendly mechanics and simple functionality."

    Terrible cars are usually not made for 4 decades.
     
    #16     Nov 2, 2019
  7. You state your personal opinion like it's a fact - I've a big ego too but I wouldn't venture to make such broad statements.

    As a former system developer on C2 (when profit sharing/% split was more lucrative) I don't think the next 2 statements are true at all.
    Yes you can game C2 by selling a bunch of leveraged volatility products (from vix/uvxy, former xiv ..) and hoping to avoid a blow up while accumulating subscribers
    - doubling down on trades
    but that's about it - those systems are very obvious to detect imo.
    How else can you game it - when you submit the order through the API you buy on the ask and you sell on the bid and you immediately see if the system trades illiquid names

    I'm still subscribing to another system (diff developer) and my fills are usually better or equal than what C2 records - I often get filled @ midprice for less liquid names versus C2 will always have your fill @ market - similar for stops, just cause you place a stop there doesn't mean C2 let's you exit at that price. I would say the results that I'm seeing are very realistic and if you trade with IB the commissions are spot on (incl. basics like Dividend accounting).

    I would agree that reverse engineering isn't the best use of time though

     
    Last edited: Nov 2, 2019
    #17     Nov 2, 2019
    guru likes this.

  8. which system on c2 are you subscribing?
     
    #18     Nov 2, 2019
  9. A discretionary system called the spirit of Nicolas Darvas
    a discretionary system called the spirit of Nicolas Darvas
     
    #19     Nov 3, 2019
  10. R1234

    R1234

    I checked out that Darvas system on C2. What about it appeals to you? Its basically been in a drawdown for more than a year now and up just 2% so far this year.
    And it looks like there are 24 people autotrading it right now. Just curious as to what C2 subscribers base their system picks on.
     
    #20     Nov 29, 2019
    sysdevel99 likes this.