Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
In trying to answer your question, I have come up with easily a dozen different theories, explanations and rationalizations, most of which I have abandoned or forgotten over the years.
Order flow analysis helped me the most to find peace with your question. It took a while to accept what I was learning and seeing every day and let go of the old baggage.
From an order flow perspective of market structure and price action, I have learned a couple of things:
1. In an intraday trend , one side of the market is in control, i.e., buy side or sell side, this reality is just so clearly obvious on any type of market delta or order flow type of chart. During these times of vertical price movement, all of the volume is on one side of the market. I have no clue who or what these entities are but they control the trend bias and direction when one is underway.
2. Trade with the side in control at the moment.
3. Never assume the side in control will loose control until they are good and ready to let go. In other words, trends tend to run a lot longer than expected or 'predicted' by most indicators.
From my knowledge in regards to the futures market, you're trading against institutions hedging their positions, market makers, algorithms that tether or trade away from other correlated or partially correlated markets, and a small sliver of retail traders. Correct me if I'm wrong, But contrary to the trend on this site, I believe most retail traders trade stocks instead of futures. The thing to gather here is that markets are correlated and if one moves too far up or down and the others don't, You'll have a swarm of algorithms or other traders pick up on that and determine if that is justified or not and then trade on that information. If there is no supporting evidence that the future should have moved away, they'll go ahead and do some sort of arbitrage or take a single position to bring that individual instrument back to equilibrium. A lot of times one instrument will trend in a given direction away from another semi correlated instrument and break a previous low or high and then move back towards equilibrium. This means that regardless of who enters a trade, or how large that trade is, there can be a near infinite potential for a whole host of other entities trading against you if you are trying to move one instrument.
R.I.P. Joseph Bach (Itchymoku), 1987-2018.
Please visit this thread for more information.
Computers don't have families or mortgages. IMO, successful discretionary day trading is a myth. It's too illusory and subject to countless biases, especially if you're gullible. You can make money for 6 months on a break even or slightly losing strategy before your profits start to erode. If you want to approach trading seriously, learn programming and statistics.
To answer your question, approximate volume breakdown of the ES: 35% HFT, 10% market makers, 12% fundamental/consistent buyer, 12% fundamental/consistent seller, 30% index arb/misc speculators, 1% noise (few lots per day).
You forget that computers lack one very important aspect, common sense. I started out in the automated division, but whenever I looked at charts with automated executions on them I often thought: "that's a really stupid trade, I could do better".
And compared to all automated approaches I have developed (except extremely curve fitted ones) my discretionary approach seems to be a rather good idea.
Furthermore the human brain is an incredibly powerful pattern recognition machine. In many cases other than mass calculation (and some others) it is possible to outperform or outsmart computers. Why should that not be true for trading?
I think that it should be readily apparent that successful discretionary day trading is no myth at all. Many institutions do it, and quite a few members of futures.io (formerly BMT) claim to do it too. Why should they lie more about their profitability than the automated guys on here?
Vvhg
Then you improve or debug it. Maybe it's a stretch, but I believe any trading system can be programmed.
You take observations from the human brain and code them, but you don't trust the human brain on evaluating the performance of its ideas. Nor do you trust the performance of it running on sim, unless you have a very large sample size, i.e a setup you get once every two days would probably need to be run for a 1-2 years on sim. Who's going to do that? Even if you were inclined to do so, as a human, you may adapt the strategy, or worse, overfit the strategy to some local time frame, so you're not even testing the same strategy.
Who on futures.io (formerly BMT) is making these claims? How long have they been profitable doing discretionary day trading?
Are you serious?
Poll 1. "What is your Trading Situation?": "I am a full-time trader 32.16%" (210 votes)
Poll 2. "Describe your trading method": "Fully discretionary, no automation 66.42%" (89 votes)
poll 3. "How long have you traded?" : "Less than 1-year 8.59%" "1-5 years 52.53%" "6-10 years 14.65%" "Over 10 years 24%"
poll 4. "Which is more profitable for a normal retail trader?" : "Discretionary Trading 64.58% (62 votes)"
Statistics show that the majority of people on this forum are discretionary traders...
wow
R.I.P. Joseph Bach (Itchymoku), 1987-2018.
Please visit this thread for more information.
No matter how much you improve or debug your system, it still will lack any common sense. You still need to monitor the system and start and stop it manually according to circumstances. Guess what, that's discretion.
And I passionately disagree that every trading system can be programmed, you will not code the invaluable intuition of a master trader into any strategy. What you percieve as the weakness of discretionary trading can and should be its strength.
Discretionary or not, you have to do your homework and in trading that normally is an aweful lot. This includes sample size. I don't know why you wouldn't want to trust the human brain on evaluating the performance of its ideas. As long as you have a thorough and and solid methodology there should be no problem. You in fact do the same when evaluating automated systems. You use your brain to tell you if it is a good equity curve, you run Monte Carlo simulations and eyeball equity curves weigh them against many performance metrics until you make a discretionary decision which system is the best and if it is good enough to take it to the next level.
I can not see any disadvantage of a discretionary trader adapting over time. All the serious automated guys do non stop maintainance aka adaption. There is no set and forget holy grail strategy.
I believe the last two questions have been answered sufficiently.
1. The human mind is subject to cognitive biases. A program is not.
2. The minimum resolution of human reaction time is in the order of 10^-1 seconds. The minimum resolution of reaction time for a program is in the order of 10^-9 seconds.
3. Most trading problems are in polynomial, asymptotic time complexity class, where programs solve problems faster than humans.
A good algorithm should not defy your intuition. A program can land billion-dollar spacecraft in treacherous conditions hundreds of millions of miles away, where vision is not available because it takes 10~11 minutes for a light-propagated signal to return to Earth and another 10~11 minutes for a signal to be transmitted back.
1. True.
2. Sounds about right.
3. You could also classify them as extremely complex and fuzzy (spatial) pattern recognition problems, where humans might be slower on average too, but have more potential for accuracy I believe.
That there are tasks where programs will always outperform the human brain need not to be argued. I just think that there are tasks in the trading universe where the human brain has the advantage (although the very small timeframes belong exclusively to the programs (see 2.)).
Funny thing that with the spacecraft. I hate lag too