Support Board
Support Board
Search Board
Control Panel
View My Posts / Threads
Direct MessagesDate/Time: Sat, 18 May 2024 10:41:55 +0000
[User Discussion] - Backtest report reliability
View Count: 763
| [2021-03-31 13:48:00] |
| User731689 - Posts: 7 |
|
Hi there, I have a spreadsheet strategy trading H1 timeframe. The performance is unbelievably good with ES, GC, CL, 6B... I am worrying about whether it's real. The profit per trade is way higher than trading costs, I am talking about hundreds of dollars per trade, thus it's not trading costs or slippage problem. Anyone has the same experiences? How do you check it? Any problem about the reliability of backtesting system that I should know? Date Time Of Last Edit: 2021-03-31 13:56:03
|
| [2021-03-31 16:30:12] |
| bradh - Posts: 859 |
|
I suggest replaying some of the winning trades at speed = 1, to see if they get the same result. Also re-test some losing trades. You might also try doing some forward testing with a live (simulated or paper money) account. This will give more realistic trading conditions. There is a delay from when the market does something until you receive it, and there is another delay for your trading decision to get back to the exchange. Backtesting has its own inherent delays, but it is hard to say if they are realistic. There may be more ideal conditions in backtesting than in forward testing. |
| [2021-03-31 16:43:57] |
| User731689 - Posts: 7 |
|
My entry is limit or stop order and exit is market on bar close. Delay might affect market order and lead to several ticks slippage. I might try to do forward testing or manually backtesting and compare the results. Did you have problem like this before, bradh?
|
| [2021-03-31 16:49:40] |
| bradh - Posts: 859 |
|
Yes, I have seen spreadsheet systems behave better in backtesting than live.
|
| [2021-03-31 16:53:25] |
| User731689 - Posts: 7 |
|
I think backtesting better than live is normal. If the difference is too big, it becomes a serious problem.
|
| [2021-03-31 19:25:04] |
| Sawtooth - Posts: 4000 |
|
Which of the three back testing methods did you use when you see unbelievable results? - AutoTrade Bar Based BackTest - AutoTrade Replay BackTest - Replay Chart The most accurate/efficient backtest is a Replay Chart using 'Accurate Trading System Back Test Mode'. Also, if your autotrade system relies on synchronized multiple charts, you must use this method. https://www.sierrachart.com/index.php?page=doc/ReplayChart.html#ReplayMode Date Time Of Last Edit: 2021-03-31 19:26:06
|
To post a message in this thread, you need to log in with your Sierra Chart account: