Algorithmic Trading is called all sorts of things:
- Algorithmic Trading
- Quantitative Trading
- Algo Trading
- Quantifiable trading
- Trading with an algorithm
- Quantitative analysis
Heck, you can call it macramé underwater if you want. But whatever you want to call it its all the same and today I’m going to show you why you should not trade without it.
Here’s an example of how you can get tangled up in Algo trading
One Saturday night you hook up with your friends to have a few bubblies and play your usual game of Texas Hold’em. You’re having a good time laughing and making jokes. Then out of the clear blue sky one of your buddies, who is a trader, like you, blurts out, “You know you should only go long when the market is oversold. And only go short when overbought.”
All heads start bobbing like the little bobby head you have on the dashboard of your car.
You grab another brewski (that’s Canadian for beer) and continue to play cards. But your mind is no longer on Texas Hold’em. It’s swirling around this unquantified, out of left field statement.
The next morning you wake up with the requisite Sunday morning hangover. Without warning, you notice your brain, once again, begins to rotate around this old trading adage, “Buy when the market is oversold and sell when it’s overbought.”
You’re the kind of guy who doesn’t take what people say at face value. You’re the kind that likes to get into the rationalization, the evidence behind statements like this. You decide to verify it. To quantify it. You want to prove this is true. Or maybe prove it’s not.
Where do you even start?
You’ve come to the right place because this is where I was 15 years ago. I remember quite clearly a webinar back in 2003 where the moderator articulated, with a straight face, you could see the bulls starting to arrive when the 7 minute MA crossed over the 13 minute MA on a 3-minute chart. What?
If that is true then tell me what would the 6 minute MA crossing over the 12 minute MA tells us? What empirical evidence do you have to substantiate this nonsense? What quantifiable evidence do you have to be able to say that with a straight face? Hogwash!”
This is not the time nor the place to start a dissertation into everything that goes into algo development but where we’re going to get down and dirty as to why you would. This is the place where we discuss the reasons why you would want to quantify it.
How many times have you heard:
- “Let your profits run and cut your losses”
- “Never trade without a stop loss order”
- “Never short a market above its 200MA”
- “Buy an oversold market. And sell when overbought”
- “Sell in May and go away”
- “Never short a dull market”
- “Buy the Dip”
Let’s take those each and see the questions you should be asking.
- “Let your profits run and cut your losses” Define profit. $0.01 or $10.00? Define run. Forever? 1 day? 2 days? 3 months? Define losses. $0.01 or $10.00?
- “Never trade without a stop loss order” Please just google “Why stop losses don’t work.” Then read this. Then read this
- “Never short a market above its 200MA” This one makes a little more sense but I need to ask, “What quantifiable evidence so you have to back this up?”
- “Only go long when the market is oversold. And only go short when overbought” This is a personal favorite of mine. What the heck is oversold and overbought. 20 different traders in a room will give you 20 (or more since traders can have multiple personalities) different answers.
- “Sell in May and go away” Sell what? Everything? If so, then what do you do with your money in June. When do you buy again so you can sell again in May?
- “Never short a dull market” Define “dull.”
- “Buy the Dip.” Of all the trader adages, this one drives me to drink more than any. What the heck is a dip? Quantify a dip, then give me evidence you should. How do you know the Dip isn’t a multimonth drawdown?
I think you see the problem.
Quantify before you believe
No one likes to be lied to. I know I don’t. I’m sure you don’t. However, I’m not talking about lying, I’m talking about statements people make they fully believe but can be wrong.
As Howard Bandy says, “The universe of trading systems is divided into two – having confidence and having faith. If you want quantifiable confidence, the kind of confidence that tells you whether or not to hit soft 17 at blackjack, whether or not to hit the blot in your inner table in backgammon, or whether or not buy a new 52 week high – use quantitative analysis. Everything else is faith.”
Use quantitative analysis. Everything else is faith.
Bottom line, dear reader, you’d have to enter the world of Algo Development to test them all.
So what do you do?
The good news is you don’t need to test them all. Just don’t trade anything you have not tested. It doesn’t matter if “Sell in May and go away,” is true or not, if you don’t “Sell in May and go away.” Who cares if its true or not?
What difference does it make if “Never Short a Market above its 200MA” is true or not if you never short above a 200MA? But if you do, then maybe you should verify this statement. You need to find someone to build you an algo so you can test it.
So call Algorithmic Trading macramé underwater if you want, just never trade without it.
We’re the Plan in “Plan your Trade and Trade your Plan” TraderJanie
Check our out the Performance Matrix.