Conservative Daytrading

A friend of mine has taken up a form of daytrading as a hobby, using a discount online broker. First, he relies on a stock analysis service to identify the best stocks, and only trades stocks that received their highest rating. He then uses technical analysis to identify which of the highly rated stocks are likely to break out (and up) from a base pattern. When one of his targeted stocks begins to move up, he buys; he sells when the upward move appears to be ending (or at the end of the day). For nearly two months now, he's been averaging one such trade per day. Over 90% of these trades have been profitable, usually yielding a gain of 1% to 4%. Since he is quick to cut his losses, the magnitude of the losses has been less than the gains.


This approach to daytrading appears to be low risk with rather high reward. Has my friend just been "running good," or might he really be onto something?

He's running good. His sample is only 40 or 50 reps, so that's just noise. Also, if he's going to hold a stock for a day, what difference does it make if it's a "good" stock? He just wants volativity/movement. The costs will eat him up unless he makes a market. The bid-ask spread beats him down, too. And that technical crap...


I might be willing to wager that a dart-throwing monkey would do as well (given same costs) if the targets were ten heavily-traded NASDAQ 100 stocks. Have the monkey throw one dart once a day, buy (or short, it doesn't matter) that stock, and liquidate at the end of the day. I see no reason why the monkey wouldn't have as good a chance as anyone else in daytrading. Good luck.

I admit I haven’t spent a great deal of time looking for one, but I haven’t yet found a pet store that sells a monkey that can flip (36 out of 40) or (45 out of 50) heads using a fair coin.


For sure, he’d go for a pretty penny.

I'm not saying the guy doesn't have real skills. He might be the best. But 40 trials isn't nearly enough. I'm not a statistician, but I don't think we can conclude the guy's found the grail based on so few trials in a market that's been up the last two months. Well, hell, lemme look at it and see what the Wilshire 5000 has done over the last two months: looks to be up about 15% since 3 April, or in about 45 trading days. About two-thirds of the trading days appear to have ended positive.


The "coin" is not fair, see. Flipping heads 36 out of 40 times when heads is 33% more likely to show up anyway is essentially meaningless. If he does it 360 out of 400, it means a little more. 3600 out of 4000 and maybe we're able to measure something.


Good luck with the monkey, but I hear they bite...

Er, did I say the "coin" was 33% more likely to come up heads? I'm wrong. It's -twice- as likely to come up heads, bit I guess ya catch my drift.


I must admit that my aversion to daytrading may come from my suspicion that I'd love it and burn through a bundle. But at least I'd still have the monkey to keep me company and hold out the cup...

My friend does not have "real skill." However, he does have real discipline in following a fairly simple trading formula. He often buys and sells the same couple of stocks for many days, and is studying technical indicators which he hopes will enable him to hold positions for more than one day with little risk when conditions are favorable.


One reason he has limited his trades to only the highest rated stocks is that he has not had tools which enable him to monitor thousands of stocks for movement, so he just targets the highest rated ones that meet his criteria. However, he also believes his EV is increased by betting on these highly rated (and relatively fundamentally sound) stocks. Is it possible that his EV is further enhanced by the fact that many other investors use the same stock ratings, helping to create greater demand for these stocks?


By the way, I slightly understated his batting average. Thus far, over 95% of his trades have been profitable (though some of the profits were very small after commissions and taxes).

Suppose his technical indicators allowed him to trade ( from the long side ) only when the market figured to go up twice as often as down? And, btw, it still would be difficult, by chance, to get 45/50 heads.


What’s really going on, I think, is this:


He’s taking small profits and, so far, has the discipline to take small losses.


But there will come a day when he won’t be able to take a small loss, either because he’ll be unable to ( because the stock stopped trading ) or because he just doesn’t feel like it.


Just yesterday, he’ll say to himself, I waited another hour, and the stock reversed itself and gave me a profit.


That day will wipe out all the small profits.


>I must admit that my aversion to daytrading may come from my suspicion that I'd love it and burn through a


>bundle. But at least I'd still have the monkey to keep me company and hold out the cup...


Sometimes folks fear what they are, or would be, good at.


Only untalented musicians and actors don’t get stage fright.

I disagree about the chance of the 45/50 heads when heads are a 2 to 1 favorite; I'd look it up in an old stats text if I weren't so damn lazy, but I imagine it's within a standard deviation. And technical indicators are a load of crap: why would anyone not buy something when it's cheap and then buy it when it gets expensive?


But I'm pretty sure you're right about blowing the small profits in one bad trade.


Who knows, anyone could be a good trader, but the only way she'd find out is if someone gave her a big bankroll, certain liquidity, a sales staff, and nearly frictionless transactions. Only the big guys can provide that stuff, and only for their employees, who have limits on what they can trade on their own accounts. If you can't afford to be wrong, can't dump your position, and have to pay retail, you get chopped up and you have to sweat your positions every damn day. It sounds very exciting, but not the life for me. I'll stick with index funds and low limit poker...

"But there will come a day when he won’t be able to take a small loss, either because he’ll be unable to ( because the stock stopped trading ) or because he just doesn’t feel like it. That day will wipe out all the small profits."


I don't think he would ever ride a stock down based on a feeling. He does not trust his feelings/knowledge and is very disciplined. As far as trading be stopped in his stock, I think his sticking to the highest rated stocks may provide some protection. If it did happen, it might wipe out most of his "small profits," but many of his trades had large profits. If the whole market crashes, he is more likely to be able to cut his losses than traditional investors like me.

"why would anyone not buy something when it's cheap and then buy it when it gets expensive? "


Perhaps a stock price becomes cheap for a reason, and begins to rise for a reason. I don't think my friend would buy relatively "expensive" stocks (the stock ratings take into account value indicators), but I do know he avoids stocks that have become cheap...at least until their price begins to rise again.

Ok, to go any further with this, we have to ask your friend a few questions.


Do all trades begin from the Long side, or has he made some profits starting off with a Short Sale?


Does he add to an initial position during the day when it goes his way?


Or when it goes against him?


Or never?

He has only bet the long side and has never added to his position.

Well, given everything you’ve said, it looks like your friend is on to something, but, fair warning, these days it’s difficult to distinguish between the sound and the unsound.

He is almost certainly NOT onto something. Going 45 out of 50 when you are a 2-1 favorite is a lot more than one standard deviation. In fact it is more than three. However firstly he is a lot more than a 2-1 favorite to have a winning trade because of his technique (but he is laying big money odds) and secondly you run into the bias that we only hear about the lucky ones on this forum. I will let others elaborate.

It's over 8600 to 1 that he could not do it by just luck as a 2 to 1 Fav.


I thought my questions eliminated the big money odds that he's laying, but to be sure, we'd have to see the actual numbers of his 50 days.

...if the expected frequency is 2/3rds is 0.013% (about one in 7500). It's roughly a 3.5 sigma event.

My friend has been trading a small number of stocks which have been outperforming the market. Basically, he looks for highly rated, high volume, and volatile stocks which are trending upward. Once commissions and taxes are subtracted from his results, I'm not sure how much more money he will have made compared to if he had bought-and-held these stocks. Though I think his approach may be less risky than buy-and-hold, I'm not sure it's worth the effort. I'm also not sure he didn't exaggerate his batting average a bit.

Regarding the bias you mentioned, I have many acquaintances who've lost big bucks by buying and holding popular stocks. My daytrading friend is the only person I know who has tried it. However, I admit it is possible that I am not aware of acquaintances who tried daytrading and were unsuccessful.

Well, it's true enough that tone deaf students quickly drop out of the Glee Club.


This must be what Sklansky's getting at.

But what do you think about Sklansky's main point?


I'd like to know, because I've often wondered about it.

As much as I hate being off by several orders of magnitude on my standard deviation estimate, I still think my dart-throwing monkey has as good a chance of repeating the guy's results as he does...

I think that the information we've been provided is incomplete and unreliable. It's quite possible that Sklansky's objections are valid. However, it's also possible he's onto something. There have been some papers in recent issues of the _Journal of Finance_ that develop models in which momentum trading can deliver excess returns.

And what happens after those momentum ideas become fairly well known?

David asked: "And what happens after those momentum ideas become fairly well known?"


The basic momentum ideas are fairly well known. You can find several examples in Jack Schwager's _Market Wizards_ series of books, especially chapters on Turtle Traders. The _Journal of Finance_ articles provide a plausible explanation: "investors" who identify changes in fundamental valuation inevitably reveal their (originally) private information as they trade; momentum traders accelerate this process and capture a portion of the excess returns from opportunities revealed by the investors.


There are some additional factors that are less well known or ignored.


First of all, there is a maximum aggregate asset level that can be applied to a given instrument, since the liquidity is supplied by the investors who, in aggregate, have a demand curve that declines with increasing price.


As this maximum level is approached (or exceeded) by momentum traders, it becomes more difficult and more risky to trade in larger size. Entry costs rise (the more shares/contracts you try to fill, the higher the price you pay for each additional share/contract), and exits can become disastrous.


Nimble traders who don't get greedy (i.e., keep trading small) can continue to capture excess returns. In fact, their profit rates may even rise as the aggregate level of momentum trading increases. On the downside, this means that the optimal time scale for momentum trading is unstable over time. And if the momentum process accelerates, the risks rise even for small traders.


So momentum trading is not a simple process, and you cannot expect to compound your momentum trading profits. E.g., if you can make 1% per day with a $50k bankroll, you should not hope to make more than 0.1% per day with a $500k bankroll (but you will be better equipped to ride out bad luck).


If you find a successful momentum trading approach, the one way you can increase your profits is to trade more securities. The more successful day trading firms have done this. Bright Trading is a good example. An individual trader can only successfully trade a handful of instruments. But the Bright brothers have what appears to be an effective momentum trading approach that they have been able to teach to a few hundred traders.

> But the Bright brothers have what appears to be an effective momentum trading approach that they have


>been able to teach to a few hundred traders.


If the rules are so simple that they can be taught to novices, then our forums have at least ten people who could program the rules on a rainy Tuesday afternoon and we could all sit around making 1% a day.

Am I correct in assuming you are a believer in the efficient market hypothesis? I sure am but remain intrigued by the studies coming out regarding a possible fourth factor (momentum) that helps explain securities prices, the other three being market (stock v. bond), size (large v. small) and book to market ("value" v. "growth").


If momentum is a small factor, it doesn't negate the EMH. The EMH would still be correct "where it counts", or for all practical purposes. Frantic trading is costly, worrisome, and makes for a hell of a tax riddle come April.


I just read an article that explained the author's belief that there is a small momentum factor that often gets captured by "tax-managed" mutual funds. These tax-managed funds track indices but will ride winners a bit longer even after the index changes composition. Of course, the funds do this for tax-deferral reasons, not momentum/trading reasons. Capturing any momentum premium is a nice bonus for investors owning these funds, and they show that there is a passive, "buy and hold" way to get the premium.

I agree. There ought to be at least ten people reading these forums who can write a program capable of generating excess returns. Some of these people already have, and the rest should consider it, especially those who have often wondered about it.

I would guess that if this sort of momentum trading became more popular, the price fluctuations of high-volume stocks would increase in frequency and decrease in magnitude until this trading approach was no longer profitable.

Very Good.

how much m,oney do ya need for making dayrading succesful??