This blog post did not answer it's own question because it's conditions were not day trading (over 3 trades in 12 months). That condition selects for people choosing individual stocks hoping for a moonshot, for which people tend to choose riskier stocks rather than stocks actually likely to make them money. So no wonder 80% lost money. On the other hand, notice that the 20% who do make money have a large power distribution curve.
A better way would have been to compare performance versus frequency of trading (I don't expect this to prove one way or another though). The fact is, there are many strategies one could take, and it's how well you execute them that counts.
Personally, I don't find daytrading riskier than holding stocks. By far my biggest losses come from holding the wrong stocks for a long period. It just seems riskier because you have to confront yourself with the possibility of loss each day, rather than hold "long term" and deny that you are wrong.
I now take the Doyle Brunson approach. The poker champion loved to pick up small pots and felt it was critical to do by aggressively playing small hands. That way, these little wins pay for the risk of playing bigger hands over time. I've had the same experience - my daytrading tends to be small money but it stems from work done for holding long term stocks. So why not put it to use?
But, to actually make good money daytrading is still really difficult. Commissions alone can make you have to be 55/45 correct, but there is also the steamroller affect where people tend to hold on to losses and double down further. So it's also about mastering yourself in addition to your market. Otherwise, there is not reason why you can't be better: you are putting in more work than others to make good decisions, and that's how you profit. Trouble is, when are you still outgunned informationally?
I think the individual retail trader is almost always outgunned informationally when it comes to intraday trades. Most short-term price action is driven by order flow and cross-asset correlations, which machines are very good at trading. They are often net trading cost earners due to rebates and capturing bid-offer spreads. That means their win rate doesn't need to be as high, so they can pull the trigger on a trade before you can, just by having lower fees and superior execution. In addition to that, they're faster, more scalable, have more access to liquidity, and are more disciplined than humans could ever be. For the ones who trade off statistical correlations, they have the best data and armies of PhDs working on signals.
There are some event-driven fast trades still done by humans, like after news or responding to economic releases, but hedge funds have highly-educated people modeling the effects of an interest rate change or earnings release as their full-time job. This area is also becoming dominated by bots doing sentiment analysis.
I think the retail trader could have an edge in a few ways. One would be an illegal edge like inside information or market manipulation. Another would be finding illiquid stocks that proprietary traders and hedge funds won't bother with and trading using similar techniques. You could also look for extreme situations that model-based traders can't understand well due to lack of data, like a merger target breaking away. Those trades would be very risky though.
FWIW I'm a professional trader and never day trade my own account or pick individual stocks. My company allows it, but I don't believe I have any edge in doing so. I just buy and hold a portfolio of ETFs.
I believe there sometimes are opportunities you can take, because a "Hacker" knows more about some parts of the world than most traders. My prime example: When Apple released the iPad it quickly became a great hit. For me it was obvious that a lot of cheap copycats would spawn soon. What would you use for a cheap iPad knockoff? An ARM processor for sure. Strategy: Buy ARM and hold for years. First iPad was released April 2010. ARM.F [0] at $3. By the end of 2010 it was near $6. Now near $20.
"My prime example: When Apple released the iPad it quickly became a great hit. For me it was obvious that a lot of cheap copycats would spawn soon. What would you use for a cheap iPad knockoff? An ARM processor for sure. Strategy: Buy ARM and hold for years. F"
Do you think that Wall Street hasn't figured that out?
Apple is one of the most watched stocks in the world. To the original commenters point: the banks have reams of 'technically astute' Analysts making those assessments, and much, much, more.
Wall Street has people all the way up and down Apple's value chain. They pay 'consultants' the world over to eek out any tiny bit of information WRT Apple's supply chain.
Again - I agree with the original commenter: individual traders are severely outgunned when it comes to these things and that's why they lose money.
Yes - theoretically, there are some very narrow areas where a savvy investor might be able to win, and surely, Wall St. has some gaping systematic problems ... but by enlarge, I agree with the sentiment of the analysis as well, even as other have pointed out flaws.
Investing is 90% gambling. The world only has so much GDP growth and there's not enough to go around to make every investor super happy. Most returns that an investor makes are someone else's loss. The losers are usually those with less information, or are lazy about it all (like some 'big dumb funds')
>Do you think that Wall Street hasn't figured that out?
Yes, I think Wall St hasn't figured that out. Sentiment regresses to the mean, and so do analyst expectations.
If Wall St had figured out what the iPad meant there would have been a huge rush towards ARM.
That didn't happen. There's your answer.
It's also worth pointing there's a big difference between tactical short-term trading and strategic long-term trading.
Short-term trading is not a winning option for non-professionals.
Long-term trading can be, because - here's the obvious point - most traders are looking for quick returns.
Unspectacular stocks with robust longer term returns tend to be systematically undervalued - more so if they're in a specialised technical niche not many people understand.
Conversely stocks with a good historical record but signs of a less robust future tend to be overvalued.
There's a huge amount of fashion and trend-following on Wall St. People who research fundamentals in depth - like Warren Buffet - are very much the exception.
You're assuming the research Wall Street publishes actually has anything to do with their own market positions, they do not. Research that an investment bank publishes is for customers; they work independently of their traders who have their own proprietary methodology and information sources.
"Do you think that Wall Street hasn't figured that out?"
"Wall Street", whoever that is on Wall Street, may have figured it out, but the stock doesn't instantaneously move to the correct value afterwards. There is a time lag where traders who are paying attention can still get in at a good price.
Event driven trades are definitely not done by humans, especially when they are scheduled such a economic releases or some other periodically released number. These are all computer driven. (I have written these for HFT shops for years now.)
Working on the pro side too, I also just hold ETFs mostly. No way an amateur day trader beats me at my job except by luck or finding trades too small for me to really care about.
Out of curiosity - do you get some kind of template version of the earnings report before it is released so you can parse the numbers? Are they just always worded the same way? Or do they broadcast the numbers directly somehow?
Earnings reports are the worst. Most of the economic numbers and other statistics are released in a machine readable format by a few companies that compete on speed. (There are some anomalies that come over the web and they are horrible to deal with.)
Earnings reports though I've tried numerous things from subscribing to services that release numbers electrically to writing parsers based on prior reports. Earnings though you care about things like one time charges, gaap and non-gaap numbers, etc. They are one of the less consistent things to trade.
Thanks for a great answer! Similar threads keep popping up quite often here and I can see why. It probably seems romantic and all that to beat the machines, beat the armies of PhDs and make some real cash.
But for a retail trader, it probably can be summed as 'don't bother, you are strongly against all odds, proceed only if it is fun to do and don't expect to make any money and lose all'.
I made a >100% return in the last 6 months by buying companies that made products that I understand and use frequently: nvidia and Amazon. I think if you focus on particular industries that you know well you can get an edge on 'the market'. I'm not day trading though, I'm just keeping an eye out for opportunities and trying to balance as much as I can. I'm also just playing with a small IRA account that I can afford to lose.
You could have picked almost any company and made a great return in the last 6 months, assuming you did not buy at the all time highs. The market is reaching record highs right now, so it is easy to believe you have an edge as the times are good.
I would not go so far to say that simply buying companies of products you buy and understand is a viable trading strategy that will give you an edge for >100% returns. When the market is in a downturn this could be very different.
"You could have picked almost any company and made a great return in the last 6 months" - exactly. Like crowing about being a great stock picker in '99. So was everyone.
You're making a logical error of assuming that because you made money, your actions caused it, that's not a valid assumption. It's a bull market, that doesn't mean you're a good stock picker.
I agree with what you say, and I actually have grown to appreciate what the bots are doing. They basically do the supply/demand discovery for you and you can trade off of this information. Especially for stocks you follow that haven't moved yet.
Example: back in February crude prices crashed, and the weekly EIA reports were very bearish. But the bots bought the initial drop on report release and crude strongly rallied after each bearish report. Maybe that's the time to buy oil stocks :)
If you ask me, daytrading seems riskier because you're essentially trading within noise. A company could rise or fall a few (and more rarely, a lot of) percentage points within a day. Is it fluctuating based on anything other than the feedback loop and noise? Usually not, I think.
It seems far more unpredictable and lacking in reasoning than something like "Amazon's strategy for the next couple of years involves X, Y, and Z, so I think they will be successful/fail."
> "Amazon's strategy for the next couple of years involves X, Y, and Z, so I think they will be successful/fail."
That's not really the right question. If Amazon's strategy is hugely successful, but everyone else already thought Amazon's strategy was going to be even more hugely successful and had priced that in, then you could be 100% correct about your question and lose money.
> If Amazon's strategy is hugely successful, but everyone else already thought Amazon's strategy was going to be even more hugely successful and had priced that in, then you could be 100% correct about your question and lose money.
Yet, Amazon is about 100% more valuable stock wise in 2016 vs. 2014. The strategy is about the same.
The real question is: "How will stock X perform vs some baseline". I use a 90/10 VTI/BND portfolio as that baseline. If you have the time to understand how to answer that question, you can make a good return as an individual investor. (Either by identifying good investments, or reverting to the baseline)
For example, if as a tech person, you used your insights into the industry and ended up picked any one of Amazon, Apple, Red Hat, Oracle and/or Google as investments 10 years ago, as part of a diversified basket of 4-5 stocks, you would have beat a safe Boglehead portfolio by a significant margin, even factoring in the market implosion in 2008/9. If you picked Apple or Amazon, you hit a home run.
It depends on your needs though. If you're saving for college and your kid is 16, you need a portfolio with a lower volatility. If you're retired and need income, you want to minimize volatility and minimize tax impact. If you're a mid-career professional like me who won't need a dime until 2030, you can take more risks.
"If you picked Apple or Amazon, you hit a home run."
Well, sure you did. If you picked Yahoo in the late 90s, a home run hit you. If you picked Intel or Microsoft in the recent past, nothing happened to you.
A good article on using insights as a tech person is Phil G's piece on shorting Microsoft in the 80s. (Microsoft was much less appealing to a tech person in the 80s than it is now, and a much better investment.) http://philip.greenspun.com/materialism/money
This is not to say that tech people can't make money banking on their instincts, just that I personally am not tremendously confident about my own instincts. I can certainly make a very compellingly sounding bull or bear case regarding any currently famous tech company.
A great example from recent history is LinkedIn's stock tanking hugely in one day and then Microsoft bringing it back to the original level in one day some time later. And I'll be damned if Microsoft ever makes that money back. But I totally should have bought the dip.
> If you picked Yahoo in the late 90s, a home run hit you.
To abuse the baseball analogy, even exceptional hitters don't get on base > 60% of the time. If you picked Red Hat or Oracle, you didn't hit a home run, but you still beat the VTI.
That's why you diversify. One safe way to do it is to buy the market. Another way is to buy a basket of diverse stocks. That basket is a higher risk but also has a higher potential of reward.
I don't understand why your comment is prefaced with "Yet". Did you misread my comment?
It seems like we're agreed that investing should not be done by reasoning like "Amazon's strategy for the next couple of years involves X, Y, and Z, so I think they will be successful/fail."
> For example, if as a tech person, you used your insights into the industry [...]
... then I would be radically under-diversified. If tech is doing well, my biggest asset (my career) is performing well. I'm most likely to need to draw on my stocks in an emergency precisely when they'll be doing poorest.
What other people think is only relevant insomuch as it determines the price. You need to ask "Is Amazon's strategy for the next years going to work well enough that I should buy at the price the stock is currently offered at?".
That's mostly true with growth stocks. If you buy Amazon trading at 300x earnings with the expectation that earnings will grow 50% per year over the next five years, then it's tough to say whether the price will be higher or lower if they meet your expectations.
Value trades rely a bit less on others' expectations. If you were to buy a company trading at 12x earnings with the expectation that earnings would grow 15% per year over the next five years, then it's pretty likely that if your expectations come to fruition then you'll make money on the trade.
And even that isn't necessarily true - the cost of illiquidity and the risk of holding Amazon's stock for a few years could mean that even though you underestimated that everyone else thought that Amazon would be hugely successful, you could still make money.
I think the opposite of every you said. Short term factors are far more predictable, with less noise, but this is only true if you trade defined situations and have good understanding of counterparties. Noise really isn't much of an issue except for illiquid stocks. The rest have highly correlated movements and random noise averages out anyways.
Most of the people I know who've been profitable day-trading have earned less than they would have in an index fund. They consider themselves successul because they're up 17% since 2014. Of course the market is up 22%, so they've actually lost money relatively to everyone else.
It's also easy to make a net return when the market is up. The trick is not losing your shirt when the market tanks.
If your minimum bet is $1000, commissions are less than 1%.
As long as you don't over leverage, the expected return should be positive relative to s&p if you sell options. But you got to hold them to collect some premium.
No I'm not, and I don't see how that is relevant. You claimed that it was hard to find stocks that moved 10% in a day. I show that this is not the case.
I started actively trading around 5 years ago, and only picked longs another 5 years before that. It's difficult to give an exact % because I'm not a day trader per se. I trade around a position I know well, but do day trades when I spot an opportunity. Sometimes you know something will happen, but only on a short time frame.
For example, I went into oil stocks earlier this year, and I knew that the market was panicking over the low oil price. So some days the market would bounce back right when the oil pits close at 2:30 (since the oil correlation can't get much worse then). Take a chance to ride the way back up, why not? But you can't do this all the time, which is why I gave you this esoteric example.
Well, if you still want a percentage, I'd say 10-20% might be a reasonable achievement.
Taxes are different. Taxes won't make a winning position a losing one, but commission very easily can. That is because taxes are on profit. Small profit => small tax and large profit => large tax, but that will not turn a winning position into a losing one.
In real dollars (after adjusting for inflation) taxes will most easily turn a winning position into a losing one. This has always been the problem with taxing capital, often you are just taxing inflationary gains.
As another commentor mentioned: you're not comparing against a 0% baseline, but against the relevant index fund. Or, at the very least, against normal savings accounts interest. So taxes can indeed turn a winning position into a losing one.
A better way would have been to compare performance versus frequency of trading (I don't expect this to prove one way or another though). The fact is, there are many strategies one could take, and it's how well you execute them that counts.
Personally, I don't find daytrading riskier than holding stocks. By far my biggest losses come from holding the wrong stocks for a long period. It just seems riskier because you have to confront yourself with the possibility of loss each day, rather than hold "long term" and deny that you are wrong.
I now take the Doyle Brunson approach. The poker champion loved to pick up small pots and felt it was critical to do by aggressively playing small hands. That way, these little wins pay for the risk of playing bigger hands over time. I've had the same experience - my daytrading tends to be small money but it stems from work done for holding long term stocks. So why not put it to use?
But, to actually make good money daytrading is still really difficult. Commissions alone can make you have to be 55/45 correct, but there is also the steamroller affect where people tend to hold on to losses and double down further. So it's also about mastering yourself in addition to your market. Otherwise, there is not reason why you can't be better: you are putting in more work than others to make good decisions, and that's how you profit. Trouble is, when are you still outgunned informationally?