A Relic Trading On Feeling

Discussion in 'Journals' started by QuasWexExort, Jul 14, 2017.

  1. I successfully use "feel" in my trading and I operate a ML system. Unless AI only trading came in, there will always be crumbs for competent human as the AI's knock lumps out of each other. I expect that even in the largest firms they would have heavily capitalized bots who interfere with each other, different development teams competing for evolutionary dominance, swings and roundabouts of liquidity & order entry efficiency. Probably safe to keep calm and carry on for now?

    The current generation who will just focus mostly on ML could be losers as they just won't spent years developing "feel" skills that will continue to present opportunity.
    #11     Jul 15, 2017
    helgen_1 likes this.
  2. Sprout


    Yes. Very good comment.

    The connection one can have with their inner guidance system is everything. Feelings offer the instant feedback of what we have our attention upon.

    Feel Good -> towards everything you want.
    Feel Bad -> the thoughts currently being thought is creating movement away from everything you want.

    The above are both necessary for contrast, which is the birth of expansion for ourselves as creators of our own experience.

    When I shifted to think about feelings and thoughts as vibrations, many insights, transfigurations and transformations began.

    The universe can shift from a paranoia viewpoint to a Pronoia one,... one mood, attitude, thought at a time.
    #12     Jul 15, 2017
    _eug_ and Slartibartfast like this.
  3. comagnum


    Kind of tired reading of all the quant/AI/ML/coding hype everywhere :vomit:

    They are like bugs attracted to an electric bug zapper.

    Well stated! Every noob that believes their computer will put them on same playing field as the big fish with hundred of millions in infrastructure, R&D and regulatory bribes. Many of these self titled algo/quants/ML/AI will learn programming - and that's about all they will get out of trading. There is nothing new about this - it has been done since the 80's by the Princeton geeks with poor results.

    They live in their backtesting/SIM fantasy world, dreaming of being that one geek to crack matrix. I will take a casino or horse race gambler any day over a gaggle of quant/algo/ML/AI geeks - you cant factor in the brawn & guts of trading into some stupid backtest. 20 years ago we used to get high and laughed hysterically at the many inverted ski slop looking backtests we could so easily create - nothing but fools gold.
    Last edited: Jul 15, 2017
    #13     Jul 15, 2017
    hmcp, _eug_, dealmaker and 2 others like this.
  4. vanzandt


    I had to log in and give that one a like. :D
    #14     Jul 15, 2017
    comagnum likes this.
  5. Yes the currency futures.

    I found my approach when I stopped doing these 2 things:

    1. Indicators (formula based).When I took those junk off was when the real learning started. No maths derivative of price will help you predict price. Watched how longer time frame levels affected shorter time frame prices. I am blind watching a intraday chart if I do not have them marked.

    2. Entering on break outs. Most chart patterns triggers an entry on break outs (eg. buying on break out trading above neckline of bullish H&S/long when swing high is broken). Did that and lost most of the time. On the occasional wins the R:R is mostly trash, very rarely do you capture a sustained trend that makes it worth it. When I did the reverse things worked out much better. You win more and make more on wins. I never know for sure if a break out is going to continue - so it is better to buy closer to the swing low rather than buying at the swing high, in the case it is not a valid break out oftentimes you can capture something from near the swing low entry to the high of the swing. You can't do that if you buy highs. Your stop is also tighter for better R:R.

    I always hold a low conviction of where price is going - but I participate in it buying as near the low as I can so:
    - if it turns out to only be a rally up to upper range consolidation I will still make profit
    - if it's a true break out above the highs I will make a lot
    - if it's a small retrace up for a lower move down I will lose a little bit or nothing if it retraced enough for stop to be moved up
    - if it's not the low and keeps going lower I will lose a bit. I'm buying what I think is near the lows, stop can be tight.

    Everything reversed for shorts.
    #15     Jul 15, 2017
    sss12, dealmaker, themickey and 2 others like this.
  6. Jdesey


    I totally agree with everything that's being said. You have to log hundreds and hundreds of hours watching price action to get the muscle memory and quick decision making skills. In Alders book he talks about the human influence in the markets. Yes it was written many years ago but he has updated it for modern times. I still believe the markets are made up of human beings and that's never going to change.

    I do not believe any amount of fancy computers can replace the skills you gain by watching charts for many many hours during live market action.

    I'll compare fully automatic trading to the idea that we will have robot cars driving us around. I just don't see that being widely adopted.
    Last edited: Jul 15, 2017
    #16     Jul 15, 2017
    comagnum likes this.
  7. Sounds like mean reversion trading. Works for me, too. Unlike you, I do it the modern way, fully automated.
    #17     Jul 15, 2017
  8. Sprout


    True and skills develop at an exponential rate when one begins to manually annotate & log the market's sequential Order Of Events.
    #18     Jul 15, 2017
  9. Jdesey


    I think you agree with me. I just saw a recent article about how algo such are doing poorly.
    #19     Jul 15, 2017
  10. vanv0029


    People are inherently better at complex pattern recognition than computers (why AI is hype). Even for Chess playing that is seemingly pure mathematical searching the best young players are better than computer Chess programs.

    The Financial Times Chess columist Leonard Barton wrote this in the Dec. 30, 2016 issue:

    "The US champion and world No. 2 [Fabiano Caruana] unleashed a brilliant opening novelty, which incidentally showed the limitations of the most powerful computers."
    #20     Jul 16, 2017
    comagnum likes this.