Got a rough idea how it is done. I go into a Forex account( I have FXCM)
Open the chart and set up my indicators and some roles to follow etc.
Move one candle at a until a desired entry point record time/date and pip
value at that point. I am a tiny bit vague about how to record the spread.
Or anything else I need to know. I don’t want to really jump in without enough
guidance.
I use excel.
Columns I use are:
Date
Time
long/short
Day
Win/lose
Win/lose %
Balance after trade
I use formula to allow me to change risk % in one cell, which then alters every other cell to show correct balance and win/loss %.
I also break down each day to show %'s, so I know which days are potentially trap days (mon or fri).
Without posting a pic of my sheet it’s a little vague I know, but I have some other bits and bobs that are purely for my strat, and I’m not ready to share those bits yet 
Cord
Thanks very much.
Would you now any software that eliminates this work for me,
without going automatic, as it could speed up things a touch.
Writing an ea for a platform such as MT4 will let you do it automatically.
I would suggest though that doing it manually is far more beneficial. You’ll get to know your system better, and besides, if your system makes you millions what’s a little leg work? 
Cord
I do decided to learn the manual system a way in become intimate with forex trading, however it would be nice to not write down those detail, kind of a semi-automatic.
You dont record the spread, because the spread changes for every pair, and depends also fo the volatility. So you have to back test all your trades without the spread. And once you have the results, you include the spread in every trade you take. For a example after all the back test you find that your optimum Stop loss fin yous system or EUR/USD is 22 pips, then at the moment you place a trade the real optimum Sto Loss is 22+spread at the moment. If the spread at the momento you place the trade is 2, then your stop loss is 24.
hope that helps…
You can fake out a spread for a bit more accuracy in your backtests. What I do is assign a pip value (can be fractions of a pip) as a parameter to my backtester which will be the distance to move from the price in the historical data to use for bid/ask. I guess you can’t really do this when backtesting in metatrader with an EA, though.
For the quickest route: candle by candle. Usually, I would note all of my entries with a vertical line on the candle and go back and note all of my exits with a different colored vertical line on the candle. Then, you can go about recording it in Excel.
It’s time-consuming as hell, which is why I would suggest you pick up an easy programming language to make it do it for you (Python is what I learned).
Perhaps this give more time on a looking and thinking about a chart, so I would assume that even with semi automation one might miss a learning opportunity by rushing too quick, no?
When you are talking about using a programming language, I assume you are talking about full automation. Again, I don’t want to do this at this stage.
It depends. When trying to backtest, you shouldn’t be muddling your head with other vague epiphanies. If you want to study the market, you shouldn’t be studying it while trading. Backtesting IS trading, albeit on paper. Also, it just makes the work longer and harder to focus on.
In my case, I like to backtest ideas individually. If it works, I try to determine if it’s a fundamental aspect of the market I’m looking at or if it’s just due to random chance (because even statistically significant results are false alpha% of the time). Ultimately, you’ll come up with a bunch of individual ideas that provide some slight positive expectancies on their own and you need to fit these ideas together to make a cohesive system. Some of the ideas will complement each other and others will be polar opposites, thereby making it impossible to fit them into the same system.