Could it be this easy?

[trader1974]

Algo trading works perfect. But the only problem is that you have to create that algo.

Well, this could be changing with chatgpt
This artificial intelligence is very interesting to generate programming code

 

[SunTrader]

As @Sekiyo already pointed out, commish and slippage alone will take a big chunk out of your profit if you're using market orders. Also, when backtesting, there's the issue with overfitting. To avoid this, it's better to forward-test or run the algo in a sim for an extended period. You will still need to take fees into consideration. In NT, you can configure commission and slippage.

View attachment 302758

View attachment 302759

It's also worth watchin this video

Finally, there's an inherent flaw with a lot of backtest engines on the market. What do I mean? Take a look at this video. The video is about TradingView, but the same limitation he talks about applies to other platforms IMO.

Exactly why the only way I would back-test a strat is to use tick-by-tick data.

No other way to tell if trade entries and exits are real-world possible results.

Also out-of-sample testing and forward testing is the only way to bulletproof final end results.

Finally as previous mentioned, a couple of times, slippage and commissions have to be accounted for.

 

[Handle123]

Have you been able to find systems that work? Or have you came to the conclusion that algo trading is not the way to go.

Find a system that works, NEVER. Took me several years to develop then eventually program scalping systems and then day trading systems, I concentrated on very low losing percentages. But still I was not happy with getting 5% and under losses. I kept working and twenty years later am very content with what I have developed.

I have always looked at financial markets different than most, didn't care about the money, only percentages...knowing that if seldom lose, account equity curve is more 90-ish degrees.

I found studying physics catapulted to higher level.

 

[SteveH]

Your brain learns virtually nothing from a bazillion computer generated "backtest of trading ideas". Quite the opposite is true when you pour over 1000's upon 1000's of charts by hand. Plus, when you also see price moving in real-time over 1000's of sessions, all of those jumping pogo sticks called price bars reveal setups the historical charts cannot show.

Letting go of a position is not a computer curve-fitted, one size fits all proposition.

Laziness gets its just reward...none.

 

[Businessman]

Your brain learns virtually nothing from a bazillion computer generated "backtest of trading ideas".

There is a definite learning experience gained from generating a strategy from a curve fitted backtest, naively being super confident and optimistic about it, only to see it crash and burn quickly in real trading.

That is when you first start to realise that even strict rules based systems trading is a lot harder than it looks in hindsight.

 

[ZTrader888]

I recently ran a backtest on NinjaTrader on 1minute charts over the past 12 months.

I'm not even going to go into the details of the script because it was one of the preloaded MA scripts that comes included as a sample.

The results were a profit of $30,950 on 1 ES contract ($13,000 maintenance margin, I believe). This is a return of 138%.

Now the obvious answer of "if it was that easy, everyone would do it," I think applies here so I must be missing something.

But I'm not really talking about the details of this particular algorithm. I've seen countless algorithms that backtest to similar or better results.

Can anyone give me specific reason why this wouldn't work this way if traded live? Or specific reasons that it would?

The NT backtest engine isn't that good. Better to download replay data for every contract and run the strategy in replay. The results will be more in tune with the real market. If you're serious about developing strategies, TradeStation is the only platform that's powerful enough. Their testing engine is far superior to NT.

 

[farmerjohn1324]

The NT backtest engine isn't that good. Better to download replay data for every contract and run the strategy in replay. The results will be more in tune with the real market. If you're serious about developing strategies, TradeStation is the only platform that's powerful enough. Their testing engine is far superior to NT.

Does TradeStation offer the same leverage as NinjaTrader in day trading?

 

[HeSaidSheSaid]

I recently ran a backtest on NinjaTrader on 1minute charts over the past 12 months.

I'm not even going to go into the details of the script because it was one of the preloaded MA scripts that comes included as a sample.

The results were a profit of $30,950 on 1 ES contract ($13,000 maintenance margin, I believe). This is a return of 138%.

Now the obvious answer of "if it was that easy, everyone would do it," I think applies here so I must be missing something.

But I'm not really talking about the details of this particular algorithm. I've seen countless algorithms that backtest to similar or better results.

Can anyone give me specific reason why this wouldn't work this way if traded live? Or specific reasons that it would?

Would you share the entry and exit rule(s) to achieve the 138% return?

 

[vincentnyc]

listen...i hate to tell you this...u maybe a long time investor...but you sound like a noob trader. one time, i did paper trade on futes by buying 6 long contract on es and and hold it overnight...next morning i was up over $30k. but i know if i did this on live trading....it would end real bad.

 

[Overnight]

listen...i hate to tell you this...u maybe a long time investor...but you sound like a noob trader. one time, i did paper trade on futes by buying 6 long contract on es and and hold it overnight...next morning i was up over $30k. but i know if i did this on live trading....it would end real bad.

No, it wouldn't. If you did it with live money, you would make your $30K. It is no different than sim.

The problem lies with the fact that you can do that only so many times before you get a string of losses and you lose all the money in the account.

This is why Rickshaw man has never capitalized on the overnight melt-up he has touted for all these years. It doesn't work. It is why there is no overnight fund. At best over a one-year period you end up at break-even. But it takes a lot of capital to survive the DD.