Position Size Vs Commission Size: A Quick Visual

A few posts ago I wrote (here) about how your position size/commission size ratio can greatly effect your returns, and how I made the mistake of using a system designed for a much larger position size. Basically, the smaller your position, the bigger the commission amount has on your profitability, which means you can’t make money on tiny wins.

I thought I’d drive this concept home with some stunning graphs and charts.

Below you can see the results of a simple swing trading system that buys the S&P 500-based ETF (SPY). The first two graphs are unrealistic in that they don’t include commission fees. They’re pretty graphs, yes? Money keeps going up. I like when that happens, even if it’s pretend.

And note how the graphs are congruent, in that they’re the same shape with just different start points and endpoints. I’ve used two sets of position sizes: the first has an account of $100,000 and position sizes of $10,000, and the second one has an account of $10,000 and a position size of $1,000. None of these tests allow compounding.

no com,10k,100k
100,000 account, 10,000 position size, no commissions.
no com,1k,10k
10,000 account, 1,000 position size, no commissions.

And now we wake up from that wonderful little dream and pay some commissions to that nasty stock brokerage. How dare they make money on our money!

Here is the same trading system, using a $4.95 commission for each buy and sell. The graph is a little more ragged than the fantasy graph, with substantially less profit, but it still looks pretty good.

100,000 account, 10,000 position size, $4.95 commissions.
100,000 account, 10,000 position size, $4.95 commissions.

Alas, the same can’t be said for the smaller account. In fact, one could say the trading system “fell off a cliff”. A more technical way to describe this system is “broke”. Which also describes your account.

10,000 account, 1,000 position size, $4.95 commissions.
10,000 account, 1,000 position size, $4.95 commissions.

“Holy banana-pants, Batman!” you probably just yelled. (That’s what the kids say these days, isn’t it?)

So why does this swing-trade system suddenly suck? (Which is another technical term, used to describe a system exhibiting continuous adverse excursions). It sucks because it has a profit target of 0.2%. Yes you read that correctly. It has a wonderfully high hit ratio when large positions are used, and makes money the old fashioned way (slowly, a bit at a time). But when a small position size is used, those commissions often turn a win into a loss.

Let’s see, $1000 * .002 is $2, minus commissions both ways at $4.95 * 2 = $9.90, so that makes my profit…uh…negative $7.70 for a ‘winning trade’. Let the swearing begin.

Whereas the larger account would have made $20, and then spent $9.90 in commissions, leaving a small (but real) profit of $10.10.

In fact, this system only tips over into profitable territory (and we’re talking $11 profit over the entire 15 years of trading) at the $1300 mark for position size. I ran the position size as an optimization variable, and here’s a chart showing the results:

min pft

Commissions have a very real effect on the ability of a trading system to make money. And the smaller the position size, the bigger average return you need per trade to make it profitable. Which can most easily be achieved by using systems that hold positions for longer.

Does this mean as I get richer, my trades get shorter?

For those of you with $10k to invest per position, I’ll bet you’re chomping at the bit to get your hands on this trading system, right? Well here’s a link….

Pseudo-Seasonal Stock Trading

As I write this, a long position I have in TASR is up over 6% just half way through the day. I’m trading it based on my Band-to-Band system (details here).

Taser International makes, well, “tasers”. Turns out that’s a brand name for what is generically known as an ‘electroshock gun’. And why is their stock up? Because two police officers were shot in Ferguson, Missouri. It seems that often when some significant news event involves either the shooting of a police officer or an “unarmed black male”, TASR (and other weapons-related companies) spikes. It’s “pseudo-seasonal” in other words.

Now, setting aside the somewhat questionable ethics of this for the moment, could the average retail trader trade something like this? I don’t mean can you “trade the news”…of course you can, but you’ll probably lose as the move is made before you get in, and the profit lost before you get out.

No, what I mean is: can you trade the “anti-news”?

I got into the TASR position because of technical considerations, not because of any news event. Or rather, the news event that triggered my technical signal was of a disappointing earnings report, not a ‘real’ news event. I got in at the right time, BEFORE the real news happened.

So…could you time the position entries so that they occur in between news events, so that you’re already long when the news event happens?

For example: what if you evaluated the frequency of news stories involving the words “shooting” and “police officer” or “black male”? What if you determined that these stories made the national news with a certain frequency, and with a certain minimum spacing between the events? Could you then wait X days after the last news event to enter a position, primed for the next one? Or calculate the average price drop over time after each news event, and put your buy order in after the price dropped the required amount?

I realize this particular news topic is less than savory, and I’m NOT suggesting you trade it. But it’s current and it’s what got me thinking about this oddball idea. It’s sort of like seasonal trading, but different. It’s more of an anticipatory news-event trading system.

I don't have any photos of tasers or weapons, but I do have a photo of an apple. Beautiful model holding it is an extra bonus.
I don’t have any photos of tasers or weapons, but I do have a photo of an apple. That a beautiful model is holding it: extra bonus!

What if for example you knew approximately when Apple (AAPL) was going to make major product announcements – which is not too difficult to figure out – and then shorted Google (GOOG) a few weeks before. Since they’re competitive in many areas, good news for Apple could be bad news for Google and vice versa.

I haven’t checked to see if there’s a correlation or not, but you get the idea. Google was down 2.7% for the five day period after the Apple Watch was announced (September 9, 2014). But that’s just a single test.

Whenever you hear about a significant news event, it might be beneficial to consider a) is the news event cyclical or “pseudo-seasonal” in some way, b) how could I have gotten into the position before the event occurred, and c) what companies or industries are effected positively or negatively by the news?

What are some other news events you can think of that could be traded in an anticipatory or “anti-news” manner? I’d love to hear them, so leave your idea in the comments. Or contact me privately if you don’t want your brilliant idea to be public knowledge. 🙂