Is paper trading a waste of time or does it yield benefits?

[Scataphagos]

Thanks for all the answers.

Some argue against paper trading because paper trading is super slow compared to back-testing. Back-testing is much faster because testing period can be over several years and results appear within minutes.

So, why paper-trade when back-testing is far more efficient? What are some advantages of paper trading over back-testing?

Suggest (1) do your back-testing, THEN (2) paper trade what seems to work. If your paper trading works, then trade real money... but small at first. When you're confident you can handle the "rough spots"*, then you can play bigger.

*Wouldn't get too enamoured if you're doing well. The test will come when you feel like the market is "kicking your ass"... and you manage to get through it without much damage.

 

[SimpleMeLike]

Suggest (1) do your back-testing, THEN (2) paper trade what seems to work. If your paper trading works, then trade real money... but small at first. When you're confident you can handle the "rough spots"*, then you can play bigger.

Very well explained Scataphagos

 

[prc117f]

It depends on what you trade. You could miss out on the effects of market microstructure depending on what you trade (ie how liquid and size the market you are trading)

 

[JSOP]

Thanks for all the answers.

Some argue against paper trading because paper trading is super slow compared to back-testing. Back-testing is much faster because testing period can be over several years and results appear within minutes.

So, why paper-trade when back-testing is far more efficient? What are some advantages of paper trading over back-testing?

Agreed with @nooby_mcnoob, paper-trading helps you discover problems that you won't find in backtesting and also it allows you a further test of your trading system on a price feed that's closer to real trading environment. For example, in backtesting you might be limited to only be able to test your trades on whatever the price feed that's in the database of the backtesting software provider that might contain error or restricted to certain price types only. Paper trading allows you to test your strategy with real-time prices and more realistic prices that you would encounter in real trading and gives you another chance to test your trading system to see how it fares.

 

[Sprout]

It depends on how you set it up.

In general, papertrading is more in the forward testing domain if you are in fact doing it during RTH. The difference between papertrading and sim before being live is that papertrading slows down time and can be set-up to log various variables that one is attempting to understand the correlation and lack of correlation as they relate to price movement. What has less utility is going through a day historically for when there is no Hard Right Edge, your peripheral perception will subconsciously influence your analysis.

It is a tool in the toolbox and is brought out recursively to work through more differentiation, in other words to assist in debugging something that was working and no longer or applying a strategy to different market conditions and context. It helps to improve performance no what where someone is in their development.

 

[SunTrader]

Almost a hundred years ago a Wall Street classic was published, Reminiscences if a Stock Operator. Here is what was written on this subject:

"I have heard of people who amuse themselves conducting imaginary operations in the stock market to prove with imaginary dollars how right they are. Sometimes these ghost gamblers make millions. It is very easy to be a plunger that way. It is like the old story of the man who was going to fight a duel the next day.

His second asked him, "Are you a good shot?"

Well said the duelist, "I can snap the stem of a wineglass at twenty paces," and he looked modest.

"That's all very well, said the unimpressed second. "But can you snap the stem of a wineglass while the wineglass is pointing a loaded pistol straight at your heart?"

+ + +

The point about practicing trading execution software and such is valid though.

 

[JSOP]

Almost a hundred years ago a Wall Street classic was published, Reminiscences if a Stock Operator. Here is what was written on this subject:

"I have heard of people who amuse themselves conducting imaginary operations in the stock market to prove with imaginary dollars how right they are. Sometimes these ghost gamblers make millions. It is very easy to be a plunger that way. It is like the old story of the man who was going to fight a duel the next day.

His second asked him, "Are you a good shot?"

Well said the duelist, "I can snap the stem of a wineglass at twenty paces," and he looked modest.

"That's all very well, said the unimpressed second. "But can you snap the stem of a wineglass while the wineglass is pointing a loaded pistol straight at your heart?"

+ + +

The point about practicing trading execution software and such is valid though.

He should've met that second earlier. That second would've told him to practice snapping stems of a wine glass at twenty paces with a loaded pistol pointed straight at his heart and practice it 20,000 times over a six month period to achieve a 100% success rate before trying the duel.

 

[Sprout]

Well said the duelist, "I can snap the stem of a wineglass at twenty paces," and he looked modest.

"That's all very well, said the unimpressed second. "But can you snap the stem of a wineglass while the wineglass is pointing a loaded pistol straight at your heart?"

Lol, that's analog is like the samurai class never training prior to combat with live swords, a boxer/mma fighter not training before getting into the ring, or a soldier not going through war games with live rounds prior to battle.

There's a point where is makes sense to go live, to do it prematurely is foolish and the motivation to recommend such a thing is suspect.

 

[Laissez Faire]

It goes without saying that you should simulator test your methodology before going live, unless you're already a millionaire and can afford to lose money.

But you shouldn't try to learn trading by simulator trading like it seems a lot of people do on here.

Rough plan:

1. Observe the market and patterns that occur. This might take you a while. Always test things you've been told. Never trust ANYONE. Trading is a lone business. You can't rely on anyone else but yourself. If you want security and guidance, get a job.

2. Create methodology based on point 1.

3. Test it, both with historical data and casual simulator testing.

4. Test it seriously with simulator trading like you'd trade real money (think flight simulator) using realistic parameters (stops and target).

5. Go in circles for a long, long, long time repeating steps 1, 2, 3 and 4.

6. Trade real money when serious simulator trading proves satisfactory.

Or, get impatient, start gambling and blow up.

PS: The ones who talk about psychological issues most likely don't have an actual methodology AND/OR they're trading with too high leverage. Also, they don't know how the markets move, so they will freak out when they start trading live prices and experience adverse market movement and don't have a plan.

PPS: This is a long, long journey. No shortcuts.

 

[Overnight]

...
PS: The ones who talk about psychological issues most likely don't have an actual methodology AND/OR they're trading with too high leverage...

I disagree. "Too high leverage" is an absolute term, but leverage is in itself a very very personal decision.

To you, would trading one contract on ES with $50,000 cash on hand be considered high leverage? Or would $5,000 cash?

Every person is different.

There is only one technical leverage possibility that makes no sense, and that would be to enter a position at the max the broker would allow.

For example, if your broker allows $500 to enter a single contract on ES, then if you have $50,000 cash in your account, entering a 100 contract position is just plain wrong on every technical level.

That is some subjective territory there, for sure.