Beau
Okay, so we've talked about what these algos are, the strategies they use, the platforms… but now, the part that feels… I don't know, the most intimidating? Actually building one. Like, where do you even start?
Transcript
Beau
Okay, so we've talked about what these algos are, the strategies they use, the platforms… but now, the part that feels… I don't know, the most intimidating? Actually building one. Like, where do you even start?
Jo
It's like standing in front of a pile of LEGOs with no instructions, right? You know you can build a spaceship, but you're just looking at a thousand little pieces.
Beau
Exactly! And my spaceship is probably going to be a lopsided cube that falls apart immediately.
Jo
Well, the first step isn't to grab a piece. It’s to decide what you're building. In trading, that means defining your objective. Are you trying to catch big, slow-moving trends over months? Or are you trying to scalp tiny profits a hundred times a day?
Beau
Right, okay. So you need a goal. Let's say... let's say I want to do that trend-following thing we talked about. Buy a stock when it's clearly on an upward swing and sell when it starts to falter.
Jo
Perfect. That's the objective. Now, the parameters. That sounds technical, but it’s just the specific rules. How do you *define* an 'upward swing'?
Beau
Uh... the line on the chart goes up?
Jo
The computer needs a little more than that. So you might use something we mentioned before, like a moving average. Your rule could be: 'Buy when the stock's price crosses *above* its 50-day moving average.' That's a specific, testable rule.
Beau
Okay, so it's all 'if-then' statements. IF this happens, THEN do that.