Is it hard to capture tiny profits trading if you are not greedy?

[Caduceus]

Is there a different modality to large trading and small trading profit probabilities? Let's say I use a tiny sum - $20,000 - equivalent to 500 shares of Vanguard Emerging Markets ETF. No leverage. I always sell at the market open, and then place an automatic limit buy order for the whole sum for $0.01 less. I make $5 for the day. Trading cost is $0. I can't do it for the next two days because of settlement. Then I repeat.

Is there really a huge likelihood that the market will throughout the course of the entire trading day never dip for $0.01 less?

Or is this a case that I will be likely to make money 99% of the time, but that the 1% of the time the market takes off a tear right from the opening bell, all my tiny gains will immediately evaporate for a one-time, single-day huge loss?

I don't actually want to do this - but I am curious about the theory/research behind it.

[greg24]

The concept feels greedy.

[Thesaints]

Is there a different modality to large trading and small trading profit probabilities? Let's say I use a tiny sum - $20,000 - equivalent to 500 shares of Vanguard Emerging Markets ETF. No leverage. I always sell at the market open, and then place an automatic limit buy order for the whole sum for $0.01 less. I make $5 for the day. Trading cost is $0. I can't do it for the next two days because of settlement. Then I repeat.

Is there really a huge likelihood that the market will throughout the course of the entire trading day never dip for $0.01 less?

Or is this a case that I will be likely to make money 99% of the time, but that the 1% of the time the market takes off a tear right from the opening bell, all my tiny gains will immediately evaporate for a one-time, single-day huge loss?

I don't actually want to do this - but I am curious about the theory/research behind it.

The theory has been written by construction site experts, researching people picking pennies in front of steamrollers.

[Mister A]

This is at least as viable as Richard Pryor's scheme in Superman II.

[Jack FFR1846]

The risk is zero. What could happen? You don't buy for that penny less, so you're sitting there in cash. Then at tax time, you have to pay tax on the total of your gains. I guess if you never do buy back in at that penny discount, you decide the game is done.

[miamivice]

Is there a different modality to large trading and small trading profit probabilities? Let's say I use a tiny sum - $20,000 - equivalent to 500 shares of Vanguard Emerging Markets ETF. No leverage. I always sell at the market open, and then place an automatic limit buy order for the whole sum for $0.01 less. I make $5 for the day. Trading cost is $0. I can't do it for the next two days because of settlement. Then I repeat.

Is there really a huge likelihood that the market will throughout the course of the entire trading day never dip for $0.01 less?

Or is this a case that I will be likely to make money 99% of the time, but that the 1% of the time the market takes off a tear right from the opening bell, all my tiny gains will immediately evaporate for a one-time, single-day huge loss?

I don't actually want to do this - but I am curious about the theory/research behind it.

Yes, there will be times that it doesn't dip by $.01. Actually quite frequent in a bull market.

You also state that you sell at market open. So you might buy sell at $50.00, buy at $49.99, and the next day you sell at $45.00. While you made $5 with your initial transaction, you lose $2500 the next day.

[ohai]

Why do you have to sell and then buy? Why not buy and then sell? Isn't this easier, since you won't need a locate for EEM?

Anyway, OP, you should absolutely do this. There's nothing that could go wrong.

[miamivice]

Is there a different modality to large trading and small trading profit probabilities? Let's say I use a tiny sum - $20,000 - equivalent to 500 shares of Vanguard Emerging Markets ETF. No leverage. I always sell at the market open, and then place an automatic limit buy order for the whole sum for $0.01 less. I make $5 for the day. Trading cost is $0. I can't do it for the next two days because of settlement. Then I repeat.

Is there really a huge likelihood that the market will throughout the course of the entire trading day never dip for $0.01 less?

Or is this a case that I will be likely to make money 99% of the time, but that the 1% of the time the market takes off a tear right from the opening bell, all my tiny gains will immediately evaporate for a one-time, single-day huge loss?

I don't actually want to do this - but I am curious about the theory/research behind it.

Yes, there will be times that it doesn't dip by $.01. Actually quite frequent in a bull market.

You also state that you sell at market open. So you might buy sell at $50.00, buy at $49.99, and the next day you sell at $45.00. While you made $5 with your initial transaction, you lose $2500 the next day.

[Thesaints]

The risk is zero. What could happen?

The market tanks at opening, 9/11 style ?

[BanditKing]

Pretty sure the frequent trading rules of the brokerage would kick in.

There's no risk. You end up with a cash position end of day if it never dipped below open.

What you are functionally doing is increasing the number of shares you hold, not the value of your portfolio (although that value would increase over time, presumably as the overall item increased in value while you were holding it later in the day.

[Caduceus]

Pretty sure the frequent trading rules of the brokerage would kick in.

There's no risk. You end up with a cash position end of day if it never dipped below open.

What you are functionally doing is increasing the number of shares you hold, not the value of your portfolio (although that value would increase over time, presumably as the overall item increased in value while you were holding it later in the day.

The risk is that I am unable to re-enter the contract at all for those positions because of a sudden massive run-up in price. It's a loss in comparison to simply buying and holding the position.

I guess what I was wondering/asking, out of pure curiosity, is if there's any way to quantify the probability of the event (i.e. the position not dipping at least $0.01 beyond the opening price) for a period of time.

For example, has anyone done research with any given stock, let's say, Apple over the year 2018, and asked the probability of event not occuring over 1 day, one week, one month, one year. "There is an 80% chance on any given day that the price will dip at least once by $0.01 beyond the opening price; a 90% chance this will be true over a one-week period, and a 95% chance over a 1-year period." Something like that.

I'm just curious - that is all.

[psteinx]

This approach has some similarities to a Martingale approach at conventional (Vegas style) gambling. In a Martingale, you bet 1 unit on a game that is close to 50/50 (roulette say), then, if you lose, double the bet (i.e. another bet of 2). If you win that bet, you're net +1. If you lose, you bet 4 units, and so on. Most of the time, you will win at least once in that sequence of 1, 2, 4, 8..., and so, the modal outcome is a +1 return. But every now and then, you will have a run of bad luck, and hit the limits of the max bet that your bankroll, or the casino's policies, will allow, and have a MASSIVELY negative outcome.

Per your plan, you're winning a very small amount, MOST of the time. But some of the time, the market just goes up, and so, you either have to pay much more than a 1 penny premium to buy back in (hours, days, weeks, months, or years later), or stay out of the market permanently (barring perhaps cataclysmic events that virtually kill off the market altogether).

[venkman]

Remember the SEC charges a fee for selling securities. Not huge, but 26 cents on $20k is statistically significant if you're only making $5 a pop.

[Stinky]

There are about 240 trading days in a year. Per OP, this trade could be done every 3 days. So you would make $5 x (240 / 3), or $400 per year, before any trading costs.

If you’re tying up $20k of capital, a $400 profit is a 2% return. Why not put the money in a CD or MM fund and avoid the hassle (and the risk)?

[HenryPorter]

Pretty sure the frequent trading rules of the brokerage would kick in.

I was thinking about this the other day. I think Vanguard's allowing of a plethora of ETFs to be bought and sold with no commission will entice a lot of short term trading. And I think it will be cause Vanguard to put more restrictions of how many round trips you can do a year.

[randomguy]

The risk is that I am unable to re-enter the contract at all for those positions because of a sudden massive run-up in price. It's a loss in comparison to simply buying and holding the position.

I guess what I was wondering/asking, out of pure curiosity, is if there's any way to quantify the probability of the event (i.e. the position not dipping at least $0.01 beyond the opening price) for a period of time.

For example, has anyone done research with any given stock, let's say, Apple over the year 2018, and asked the probability of event not occuring over 1 day, one week, one month, one year. "There is an 80% chance on any given day that the price will dip at least once by $0.01 beyond the opening price; a 90% chance this will be true over a one-week period, and a 95% chance over a 1-year period." Something like that.

I'm just curious - that is all.

You would need to find the actual data but I wouldn't be shocked to learn that it happens a decent percentage of the time. It seems like a common pattern is the stock opens up, goes up even more and then gives back some of the gains. The story you can tell yourself is people act on new info, that info spreads, and then the early actors take their profits. For example Alphabet today according to google opened at ended yesterday at 1116.05. It opened today at 1111.30, went up to 1127.65, and closed at 1119.93. So basically today you lost almost 5 bucks instead of making almost 4. I have a feeling with this scheme you will miss a lot of good days but enjoy all the bad days.

[Thesaints]

Pretty sure the frequent trading rules of the brokerage would kick in.

I was thinking about this the other day. I think Vanguard's allowing of a plethora of ETFs to be bought and sold with no commission will entice a lot of short term trading. And I think it will be cause Vanguard to put more restrictions of how many round trips you can do a year.

I don’t think any serious brokerage has any limit on ETF trading frequency.

[aspiringboglehead]

This approach has some similarities to a Martingale approach at conventional (Vegas style) gambling.

I just wanted to endorse this and was going to say exactly the same thing. I have seen the experience of "day traders" that I know mirror the highs and lows of Martingale betting. (You can make $400 a day for months and then lose $80,000 in a particularly bad two days.)

[quantAndHold]

Out of sheer curiosity I backtested this. If the buy limit wasn't hit, I bought at open the next day. For simplicity sake, I assumed zero slippage, which probably isn't realistic.

March 11, 2005 (the first day I had full data for VWO) - Jan 1, 2019

• VWO buy and hold returned 103.92% during that period
• The algorithm, trading every 3rd day to let the previous sell/buy settle, had 129 days the limit wasn't hit. Return was 63.49%.
• The algorithm, trading every day, had 323 days that the limit wasn't hit. Return was 22.25%

I also tested it with SPY, during the period 1/1/2002 - 1/1/2019.

• SPY returned 198.65% during that period.
• Trading every 3rd day, the limit wasn't hit 70 times. Return tracked buy and hold almost exactly.
• Trading every day between 1/1/2002 and 1/1/2019, the limit wasn't hit 205 times. Return was 156.51%.

There weren't any days when the bottom fell out of the algorithm. The problem was that there would be times when the ETF would go straight up for days or weeks on end, and the algorithm would miss much of that, because it was sitting in cash waiting for the limit order to execute.

[Caduceus]

Out of sheer curiosity I backtested this. If the buy limit wasn't hit, I bought at open the next day. For simplicity sake, I assumed zero slippage, which probably isn't realistic.

March 11, 2005 (the first day I had full data for VWO) - Jan 1, 2019

• VWO buy and hold returned 103.92% during that period
• The algorithm, trading every 3rd day to let the previous sell/buy settle, had 129 days the limit wasn't hit. Return was 63.49%.
• The algorithm, trading every day, had 323 days that the limit wasn't hit. Return was 22.25%

I also tested it with SPY, during the period 1/1/2002 - 1/1/2019.

• SPY returned 198.65% during that period.
• Trading every 3rd day, the limit wasn't hit 70 times. Return tracked buy and hold almost exactly.
• Trading every day between 1/1/2002 and 1/1/2019, the limit wasn't hit 205 times. Return was 156.51%.

There weren't any days when the bottom fell out of the algorithm. The problem was that there would be times when the ETF would go straight up for days or weeks on end, and the algorithm would miss much of that, because it was sitting in cash waiting for the limit order to execute.

Thanks so much for taking the time to run the test. When you said you were would "buy at open" the next day if the previous buy limit never executed, but then when would the stock be sold? If I'm understanding you correctly, the backtest you ran would simply buy at whatever the next day's opening price was. So Monday you sell stock at $42.01 and put a buy limit order for $42; unfortunately Monday the stock only goes up, so the order never executes; on Tuesday the stock opens at $44.01, and you buy at $44.01, making a loss of $2. Then when do you sell?

In my original example I was thinking of never buying in again unless the position hit the buy limit (so, in the example used, $42). Buying at open the next day would seem to open up the position over time to much higher negative returns than your numbers seem to suggest.

[carol-brennan]

Is there a different modality to large trading and small trading profit probabilities? Let's say I use a tiny sum - $20,000 - equivalent to 500 shares of Vanguard Emerging Markets ETF. No leverage. I always sell at the market open, and then place an automatic limit buy order for the whole sum for $0.01 less. I make $5 for the day. Trading cost is $0. I can't do it for the next two days because of settlement. Then I repeat.

Is there really a huge likelihood that the market will throughout the course of the entire trading day never dip for $0.01 less?

Or is this a case that I will be likely to make money 99% of the time, but that the 1% of the time the market takes off a tear right from the opening bell, all my tiny gains will immediately evaporate for a one-time, single-day huge loss?

I don't actually want to do this - but I am curious about the theory/research behind it.

Sounds good in theory, but yes, it won't work in practice. Trading for tiny gains also means you have to keep a tight rein on losses, and that will not always work out.

What you'll end up with are losses PLUS a huge accounting headache at tax time.

[Wiggums]

This first time you get stuck holding the stock, the game would be over.

[Caduceus]

I thought I would post an update with my real money experiment. Unfortunately, things have not worked out, so I am posting it here to make the embarassment doubly huge and public so that I remember the lesson.

I sold 1,000 shares of a stock and kept the proceeds in cash, telling myself that I would buy back at $0.10 below my sale price. It has been two weeks, and the stock has soared so much (+$1.50) that had I simply remained in the position, I would now be $1,500 richer.

Of course, this is not the end of the experiment. I could in all likelihood lose even more if the stock keeps going up, or diminish my losses if the stock goes back down. This is a position I intend to hold, and I only sold a small portion of my existing holdings. But the answer to my question posed in the title is ... "Yes, it can be surprisingly hard to capture tiny profits even if you don't think you are greedy"

To put it simply: I, Caduceus, gave up $1,500 in real money for the prospect of making $100. So stupid.

[livesoft]

It was a good lesson to learn.

[oldcomputerguy]

Happens to the best of us.

About five years ago, before I discovered Bogleheads, I had the brilliant idea to stick some money into Corning stock, reasoning that they are fiber optic cable manufacturers, and what with the push for "fiber to the home", there'd be great demand. Several months later, my shares were worth $0.60 on the dollar. I sold in disgust (fortunately it wasn't much, a few $100 - play money) and moved on.

To add insult to injury ... in the last two or three months, ditch-digging crews have been laying conduit underground along the road side out my way. I asked and was informed that it was for fiber optic cable. Later, I found out that it was being pulled for Verizon; I assume it's part of their 5G roll-out.

When the cable had been pulled into the pipe, I had a chance to look at the jacket. Guess who manufactured the cable?

[Caduceus]

It was a good lesson to learn.

[onthecusp]

It was a good lesson to learn.

Yes indeed. It is a lesson I learned a few years ago with the same plan.