How close tester results to reality?

[Sergei Toroshchin]

How close strategy tester results to reality? Can I expect at least 90% of the results?

[Sergey Golubev]

There is one old Chinese thread here (some quotes from the thread):

Forum on trading, automated trading systems and testing trading strategies

...

Sergey Golubev, 2016.10.29 12:16

...

Some of their comments from the thread:

• "To interpret this chart - take a look at the test history from the start time and end time, and look at this delightful amplitude and time line. "
• "Since 1971? My sister was born at this time, now almost 50 years old. Since foreign exchange transactions don't allow you to make a quick buck, there's not much to talk about."
  
• "I look at these graphs and I feel better and better. "

[Sergey Golubev]

About the reality (concerning MT4 and MT5, and so on):

• read this summary thread: All (not yet) about Strategy Tester, Optimization and Cloud
• Testing trading strategies on real ticks and the explanation is on this post (all the particularities related to the differences of backtesting in MT4 and in MT5 with "every tick based on the real tick").

Look (go to) at the links above - it is well-known information.

[Alain Verleyen]

Sergei Toroshchin:
How close strategy tester results to reality? Can I expect at least 90% of the results?

It depends of the coder and the strategy and not that much about the tester.

If the coder is competent and honest, and if the strategy is testable (doesn't depend of latency, or external events), then yes the tester results can be close to reality.

If the coder is incompetent, are a cheater, then no the results are not reliable.

So there is no general answer, you need to check live if you have any doubt.

And of course, additionally, even if reliable, a backtest is always on past data, there is no guarantee for the future.