Scalping_My Way with ACV

In other words, whatever the algorithm is, backtesting would be a requirement. Making the statement that no backtesting is required simply says the guy hopes you'll fall for it.

Perhaps he doesn't use conventional backtest engines on platforms like Ninja or TradeStation. But I doubt it. I'd make a bet in Vegas that he did substantial backtesting with every update of the algorithm. Only a fool would run a strategy live or waste time in sim without substantial backtesting while modifying and correcting the code.

Developing and writing code requires continual backtesting to ensure viability of the idea as well as the programming. That's how a scientist looks at it.
Nonsense.
If you want to back test, fine.
But I write algos and live trade them without backtesting.
What's more important is no glitches in the coding.
 
The big tipoff here is this:

MNQ
450 Tick Range Bars
Account Size = $750
Margin is set to $550 per Contract (way more than needed)
Purposely set to trade a Maximum of 4 Contracts Per Trade
NO - Backtesting Required
Trading Time is from 6:30 am to 10:30 am - Pacific Time Zone
10 Days


This is just outright ridiculous...

There's something I don't understand here as well...

If the margin is set to $550 per contract, and you have $750 in account, then after entering the first contract, you'd have to be up $350 on the first contract in order to enter the second one.

$550x2 is $1100. So you enter one contract, and you now have $200 margin left. Ergo, you cannot enter second contract until your open position reaches $350. In the MNQ, that one contract would have to be 175 points in the green before you could layer on a second position. I do not recall seeing that in your charts.
 
Nonsense.
If you want to back test, fine.
But I write algos and live trade them without backtesting.
What's more important is no glitches in the coding.

That's blatantly ridiculous. If you're writing an algo, you have to test it. What you're telling me is that you write a code, deploy it on a live account and then hope for the best. Sorry, it don't fly.
 
There's something I don't understand here as well...

If the margin is set to $550 per contract, and you have $750 in account, then after entering the first contract, you'd have to be up $350 on the first contract in order to enter the second one.

$550x2 is $1100. So you enter one contract, and you now have $200 margin left. Ergo, you cannot enter second contract until your open position reaches $350. In the MNQ, that one contract would have to be 175 points in the green before you could layer on a second position. I do not recall seeing that in your charts.

And that's if you don't encounter volatility intrabar. The whole scenario is ridiculous.
 
That's blatantly ridiculous. If you're writing an algo, you have to test it. What you're telling me is that you write a code, deploy it on a live account and then hope for the best. Sorry, it don't fly.
Shows how much you don't know!
For some, possibly yourself, backtesting is trying out different time frames on indicators.
"Let me try 22ma, nope, lets try 24, nope, oh maybe RSI 10, nope, dang...."
 
Nonsense.
If you want to back test, fine.
But I write algos and live trade them without backtesting.
What's more important is no glitches in the coding.

And maybe you and the OP should look up the definition of "strategy". It's a PLAN of action. You're creating something for use in the future. That means you need to know what happened in the past. The lack of analytical thinking here is astounding.
 
Mate I've been trading longer than you've been alive.

And that means you have tried over and over to make something work. And that "something" was an observation you made in the market, something that's redundant and repeatable. If you saw the same "something" hundreds of times, that's backtesting regardless even though you just looked at it.
 
Back
Top