Index futures automation

Discussion in 'Journals' started by hilmy83, Jul 3, 2022.

Can a fully automated trading strategy work in the long run?

  1. YES!

    56 vote(s)
    66.7%
  2. Hell naw.

    15 vote(s)
    17.9%
  3. I don't know, I got my own trading to worry about.

    13 vote(s)
    15.5%
  1. BKR88

    BKR88

    When do you start the HAF (Hilmy Algo Fund)? :)
    I'll start now with $5K (5K/1K per contract = 5 contracts NQ). I'm comfortable with high ME for daytrading.
    Increase by $5K (5 contracts) each month as long as previous month is profitable.
    You keep 30% of profits based upon high water mark at the end of each month.
    In the long run you'll make far more than $250K based upon your performance shown.
    Send me the LPOA (Limited Power Of Attorney) when you're ready. :)
     
    #401     Jun 16, 2023
  2. hilmy83

    hilmy83

    At 5k, you'd start with 1 micro contract (based on my simulations). Enough to pay your gas every month.

    Honestly I'd rather just give my IP/algo away for a lump sum. Selling signals is basically giving it away for free. Managing money is similar unless I do it through my own accounts.

    I'd sell it for 250k. With 250k account, you'd make it up in 2 years lol.

    Or how bout if I crowdfund it across multiple buyers.
     
    #402     Jun 16, 2023
  3. NoahA

    NoahA

    Imagine how amazing this would look over at C2?

    https://trade.collective2.com/
     
    #403     Jun 17, 2023
  4. BKR88

    BKR88

    If a broker allows 1 ES/NQ contract @ $500 for daytrading, why not trade it that way?
    As long as you don't martingale or let it run without a SL it shouldn't be a problem to trade 1 ES/NQ with $1,000. I believe your algo uses stop-and-reverse which is OK.

    Why do you have $11K trading a small number of micros? You'd be blowing away the markets if you used a bit more leverage. Put 50-75% of that in the bank @ 5% or invest in ETFs and keep trading the same # of contracts.

    If you traded my account with serious leverage, you'd earn multiples more than you're currently making and your $11k could be invested elsewhere. Hedge funds make serious money by trading OPM. :)

    I'm not a tech or algo person so not interest in buying one.
    Nothing wrong with selling it though as you can still trade it yourself & continue selling it to others. 4 buyers = $1M :) To get buyers interested though you need to increase the leverage to show it can make a higher ROI. You've got a great track record just need more $$$$ to show for it.
     
    #404     Jun 17, 2023
  5. Mo06

    Mo06

    If I understand that correctly (mnq...?) that's 2281.28 Nasdaq points ?

    That's not too shabby at all, equates to around 667 SPX points (what I currently trade).

    39 SPX points per week. (17 weeks trading ?)

    Impressive.

    One thought - would prop trading be one way of increasing the trade size ?
     
    Last edited: Jun 17, 2023
    #405     Jun 17, 2023
  6. hilmy83

    hilmy83

    C2 is a crappy platform for futures scalping. I can use it to build a track record i guess, but not for selling signals. Their latency is bad, all orders are routed as market orders on their end, and once you have enough followers, they limit your size cause CME restricts max contracts for the signal provider (mostly due to subscribers doing stupid shit like trading large sizes).

    Suffice to say i know my strategy that $50,000 is the minimum capital required to trade 1 NQ contract. multiple reverses and large ranges will kill a puny 5k account. I build that requirement based on decent margin of safety. This is not a make 20% a month strategy. This is an average 40-50%...60 on a good year.

    Plus the 11k is not my account. i've been posting this throughout indicating that is a model account to compare to the benchmarks using my 20k performance results.

    I rather sell it then manage OPM. If track record is an issue, feel free to let me know which audit firm takes a retail client. It doesn't make sense to pay like 10k+ for an account of 20k

    It's in ticks, so just divide by 4 for points.
     
    #406     Jun 17, 2023
    NoahA likes this.
  7. BKR88

    BKR88

    With the 11K account I imagine you're trading 2 MNQ? (Sorry, I haven't followed this thread much.) :)
    If so, what's the largest intra-day drawdown (in dollars) you experienced since you've last revised the algo, if that's an easy number to get.

    Suggestions: (IF max intra-day drawdown is less than $200 trading 2 MNQ)
    1-Reduce account to $2,000 ($1,000 per MNQ) and continue trading 2 MNQ. Backtesting might show times of greater drawdown but since you're still small why not increase a bit & let it run with 2 contracts. You're not going to lose any more money just increase ROI (& drawdown). IF you run into a large drawdown at some point then maybe create a max daily drawdown to stop the algo if it occurs.

    2-Leave account @ $11,000 but increase position size by 1 MNQ after every green week & reduce by 1 MNQ after every red week. Mechanical position size based upon performance.
     
    #407     Jun 17, 2023
  8. NoahA

    NoahA

    Would you be willing to share a chart of performance by trade number? Here is the reason I ask.

    I find it interesting that you're up every single day. I have looked closely into your stats before and see that your max consecutive losses is 2, and average loss is $208. So clearly, your system keeps trading, since you never have a losing day, and this is perhaps what I find most inspiring. It just doesn't give up!

    I don't know how much "brains" you built into this algo, but when I'm trading, I find that when I have a few losses in a row, I get discouraged and stop, and take the hit. But the smarter thing is to realize that those losses are perhaps showing me that I'm trading in the wrong direction. Those losses are actually a signal that I should be using to give me more insight into what can happen next.

    I know you mention that you have the reverse function, which is I guess where the "brains" of your algo come in. Reversing of course isn't easy because if you lose on a short lets say, and then decide to go long, you're not going long after an extended run perhaps, and that is usually not the best time to enter. So you still need to look for a good entry.

    Anyway, so I wonder if you don't mind sharing the profit by trade number. Could it be that your algo makes a guess at a trade, and if its right, you take your profit and close up shop, but if its wrong, then this is when it keeps trying? So maybe the secret to your magic sauce is simply to keep trying if you fail at first, in a logical way via strict entry rules of course.
     
    #408     Jun 17, 2023
  9. Mo06

    Mo06

    'It's in ticks, so just divide by 4 for points.'

    So I get 2281/4 NDX points, or 570 points, which equals 166.25 SPX points.

    Or 9.78 per week traded. Still not too bad.
     
    #409     Jun 17, 2023
  10. hilmy83

    hilmy83

    Don't want to go too deep into it, but yes, the algo is designed to keep going after loss up to certain point.
    Sample trades you can see what it's doing after a loss (red). Synchronization is mt5's post trade commissions/adjustments.

    upload_2023-6-17_16-56-1.png

    upload_2023-6-17_16-57-2.png
     
    #410     Jun 17, 2023
    shuraver likes this.