Entry strategy inspiration for Nasdaq with 1-10 minute TF

Forums ProRealTime English forum General trading discussions Entry strategy inspiration for Nasdaq with 1-10 minute TF

Viewing 6 posts - 1 through 6 (of 6 total)
  • #153132

    Hello,

    First of all, it’s very inspiring to read about all different initiatives and see how helpful people are when sharing knowledge and spending time to support the development of different ideas! Impressive community!

    I’m new to the world of algo trading and I try to read and learn as much as possible here at the forum while at the same time start the work on some own strategies. One thought that I have is to try to create a strategy for Nasdaq (US Tech 100) using multiple entry criteria with the aim of taking few and short trades for each entry condition, striving for high win-rate. At the moment I have started some small work with entries based on multiple time frame RSI in combination with Directional Index and some MAs on longer time frames to only trigger trades in the direction of the trend, in the backtests this triggers few trades but with good win rate, so I’m currently looking for inspiration of additional entry criterias that could be added to the strategy, any ideas of entry criterias that could be worth to trying on 1-10 minute TF at Nasdaq? (win-rate is prioritized higher than gain-potential for the individual trade).

    Thanks in advance!

    #153271

    Hi there FKNO,

    Of course I don’t know all, but I can’t imagine one strategy for all Nasdaq stocks. Btw, you are talking about Auto Trading – that would even be more impossible IMO than thinking this over for manual trading.

    Anyway … if you have good results BUT like a higher win rate instead of fewer with a high gain :

    in the backtests this triggers few trades but with good win rate

    just apply it to more stocks.
    So yes, it is as simple as that. Of course take into account the minimum brokerage fee.

    My stance ? better learn the market(s) more and see how you can profit from that. Example … see attachments;
    I tried to hide the underlying real figures, but while both look the same for graph, the one (ES = S&P 500) earns breakfast, while the other (NQ = Nasdaq) earns a nice car. All relates to Europe where you live in,, and how the ECB by the mouth of Christine Lagarde held its last presentation for 2020. You can see where she finished …

    No Auto Trading can beat this, never mind you may think the USA is not related much. It is a 100 % …

    My 2c (excl. profit 🙂
    Peter

    #153292

    Hi Peter,

    Thank you for sharing both your thoughts and trading examples!

    Fully agree with the comment of no strategy to fit all stocks and realise that I should have been a bit more specific. The intent is to create a automated trading system for Index trading and more specifically for NDX. I have started the build and have somewhat of a base structure developed with inspiration from this forum, but I would now like to expand it with some more triggers to enter the market as the current strategy takes very few (but based on backtest sussefull) trades, so hoping to find some more potential trade opportunities.

    So what I’m looking for is basically examples of indicator combinations that has shown to be successful when auto trading this specific Index 🙂

    //F

    #153468

    Hi again F,

    See the first attachment; this is on a 5 minute bar the opening from yesterday (Friday Dec 11). 0,39%.
    Second attachment is the opening of the 10th.  0.50%.

    Watch out please : because of a maximum of 4 attachments (apparently) not all show. But you can do it yourself.

    Then the 9th. 0.28%.
    And the 8th. 0.22%.
    Lastly for this week the 7th. 0.24%.

    This is 0.326% on average.

    Because this happens by standard each opening *and* because within a bandwidth (I did not examine that, but see the last attachment) this happens several times in the first 10 minutes at least, you should be able to make this profit of around 0.3% several times, each day again. Of course you must be able to catch the happenings properly and you’d need to practice manually with this first.
    What I would do is determine a top and bottom after the first (second ! or even tick) bar comes in (you can base these on the minimum variation which is general, for last week 0.22%), set a short at the top and at the same time a long at the bottom. From there keep on “oscillating”, rigorously inverting positions at each end. N.b.: Doing this manually is virtually impossible because of psychological impact – I would not be able to do it, because a. it goes too fast and foremost b. I would wait to long (more more more !). But a program won’t know about psychics.

    For each day for numerous weeks you can determine how long this oscillation lasts at least, so before that found smallest period of time found ends, you bail out definitively.

    I myself really have many ideas about these USA indexes , but I can not Auto Trade where I like it (IBKR environment). Also see next post.

     

    #153477

    Continuing …

    What I do with all my Automatic Trading, is look at the graphs with my eyes, and make the program do what I would do. So, no indicators of any kind, but of course numerous averages of various kinds.

    FYI : It is not so difficult at all to achieve 100% wins (over a 100 times per day) …
    But the point is :

    Say that you’d put 10K USD into this. Say that one win is 0.2% (I roamed off the missed peaks to stay in safe area), then this would bring USD 20 per trade. Say that this would work 4 times per day (like the last attachment in my previous post shows) then this would be USD 80 each day. Obviously, if you throw 100K at this, it would be 800 per day.
    But nothing will remain after IG’s spread is let loose on it …
    On a not unimportant side note, a trade of 10K on this index will probably cost USD 2 or so (I never tried) with hardly spread to incorporate. So that leaves you with 18 (or 180) per trade.

    And so you see that for IG (the only broker you can use with Auto Trading at this moment) the gains must be significantly higher – and my suggested “algo” won’t work out with IG …

    One more post to come …

    #153480

    Envisioned that it can work after all, you should be able to improve significantly, by means of observing the related future. Thus, the current NQ0321 would represent the NDX and you could have it open as well. That Future will tell you quite well where the NDX will head at the opening, and with some interactive means on your side, you could steer your program to the price point you observe in the Future, will know that if the Future is positive for the day, the NDX will open at a highest point for the first oscillation, and you thus go short on the first trade and at a pre-set price.
    Of course this now requires attention, so I would not really do it like this myself (can’t be Back Tested either).

    Watch out :
    I don’t even know – and maybe others do, but I would attempt to use the Future itself and observe it in your program outside of normal Trading Hours. The results will be quite the same (0.364% on average for the past week), and you will be able to see in your program where the price is at the opening. Thus, it is not about trading the Future outside of Trading Hours, BUT the opening will imply the exact same happenings.

    And now the thing I am waiting for myself (Auto Trading with IBKR) … When I observe Futures outside regular Trading Hours, there are sooo many more opportunities to make nice money. This is at least 0.5% down and up each day and thus implies 1% each day. You see these coming easily in your program (after a mile of experience at observing what actually happens) and with virtually no cost and spread, you now have a second and mostly third gain moment on the same instrument, each day – over and over again. Below you see yesterday’s (Dec 11) example, which is there every day (always going down severely at some stage, and always recovering).

    Mind you please … Here too I would not be able to trade this manually, because there are too many false indicators while you want again more more more (in this case you will be too early at going Short). HOWEVER, I use the very same principle to stay Long (it will recover all right, no matter how must loss you see appear).
    But you must have something like Bloomberg running live at all times, because WHAT IF Trump sets fire on the White House or whatever he could do. This usually implies (very fast)  downward trend and it will not recover soon, which for me means I should bail out the Long, and which would mean for you that you should never invert your Short position (you’d lose severely after that).

    In all cases mind the cruciality of the bar size. See last attachment and to the top right of the mouse pointer. That’s quite fake news for a bar of say 5 seconds. But for a bar of 5 minutes (like this one) it works out.

    What a hobby eh !
    Peter

Viewing 6 posts - 1 through 6 (of 6 total)

Create your free account now and post your request to benefit from the help of the community
Register or Login