Tick Data Backtesting Vs 1 Min OHLC
To my understanding your EA tested with M1 candles is profitable by accident because the M1 open price simulation skips all the intra-candle fluctuations.
Like a few folks on here, one of my EA's is profitable on 1 min OHLC (or lowest time frame, see above photo) and not profitable on Tick Data. I have no intentions of going live with this strategy and know that it wont work, however I am curious. I would like to understand why the 1 Min OHLC results cannot be simulated with tick data, Im aware that it wont work, but I want to fully understand why. Why can you not just code into the EA to only take trades on 1 min candles...?
A little about my EA, the entry is taken on a live candle, mid fluctuation if you will (not closes or opens). Once the trade is taken whilst the candle is still making its ups and downs, the trade is held and could either be stopped out on that same very candle, or if profitable it closes on the close of that very candle it opened on. I have read about some peoples stories here and here and have read about ticks in various places, like here. So ill ask my question again, why can you not simply add into the EA code to only open and close trades on 1 min candles. Like, you still back test using tick data, but say an entry criteria is met, then the EA waits for the 1min close, then takes the trade. I cannot wrap my head around why this would not work, even though I know it wont because so many people have asked the same question, I am just not satisfied with the answers I have read:)
Here is an example from my ea, Below is tick data (left) vs 1 min profitable tests (right) here is just one example where 1 min catches a large move, but tick data doesn't. There were a few tick data caught that 1 min didn't, but not enough to make it profitable.
Any info helps, thanks!
For various reasons.
The OHLC mode returns 4 ticks (OHLC) per candle, the everytick mode returns much more ticks.
Trading once per bar, at bar opening may help you, by searching for "on new bar" on the forum you'll find plenty of code snippet from people doing it.
Don't ignore :
- Trading only on bar opening = 1 operation per minute = less profit
- Every bar opening is a just another tick thus every tick may be considered as a smaller timeframe's bar opening (timeframe 1s or 10s etc... why not? :))
Like a few folks on here, one of my EA's is profitable on 1 min OHLC (or lowest time frame, see above photo) and not profitable on Tick Data. I have no intentions of going live with this strategy and know that it wont work, however I am curious. I would like to understand why the 1 Min OHLC results cannot be simulated with tick data, Im aware that it wont work, but I want to fully understand why. Why can you not just code into the EA to only take trades on 1 min candles...?
A little about my EA, the entry is taken on a live candle, mid fluctuation if you will (not closes or opens). Once the trade is taken whilst the candle is still making its ups and downs, the trade is held and could either be stopped out on that same very candle, or if profitable it closes on the close of that very candle it opened on. I have read about some peoples stories here and here and have read about ticks in various places, like here. So ill ask my question again, why can you not simply add into the EA code to only open and close trades on 1 min candles. Like, you still back test using tick data, but say an entry criteria is met, then the EA waits for the 1min close, then takes the trade. I cannot wrap my head around why this would not work, even though I know it wont because so many people have asked the same question, I am just not satisfied with the answers I have read:)
Here is an example from my ea, Below is tick data (left) vs 1 min profitable tests (right) here is just one example where 1 min catches a large move, but tick data doesn't. There were a few tick data caught that 1 min didn't, but not enough to make it profitable.
Any info helps, thanks!
Just as already explained on this thread, if your strategy tests well against M1 data, but not against Tick data, then simply alter your strategy (and code), to only place and/or close positions on the Open price of the M1 bars. This includes but is not limited to, the S/L and T/P as well, by manually closing on the open price of the M1 instead of leaving the T/P and S/L to be hit by the randomness of the tick price fluctuations.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Like a few folks on here, one of my EA's is profitable on 1 min OHLC (or lowest time frame, see above photo) and not profitable on Tick Data. I have no intentions of going live with this strategy and know that it wont work, however I am curious. I would like to understand why the 1 Min OHLC results cannot be simulated with tick data, Im aware that it wont work, but I want to fully understand why. Why can you not just code into the EA to only take trades on 1 min candles...?
A little about my EA, the entry is taken on a live candle, mid fluctuation if you will (not closes or opens). Once the trade is taken whilst the candle is still making its ups and downs, the trade is held and could either be stopped out on that same very candle, or if profitable it closes on the close of that very candle it opened on. I have read about some peoples stories here and here and have read about ticks in various places, like here. So ill ask my question again, why can you not simply add into the EA code to only open and close trades on 1 min candles. Like, you still back test using tick data, but say an entry criteria is met, then the EA waits for the 1min close, then takes the trade. I cannot wrap my head around why this would not work, even though I know it wont because so many people have asked the same question, I am just not satisfied with the answers I have read:)
Here is an example from my ea, Below is tick data (left) vs 1 min profitable tests (right) here is just one example where 1 min catches a large move, but tick data doesn't. There were a few tick data caught that 1 min didn't, but not enough to make it profitable.
Any info helps, thanks!