On another post I am developing a project to test a backtesting application. I am soon going to be running some backtesting on real data and wish the results to be as representative of real world as possible. To this end, can anyone reply to this post by filling in any or all of the fields requested below? Instead of taking just one use case I will try to incorporate a “typical spread” and a “typical swap long/short”. When I report on test results I will state what settings have been used.
Information requested:
Spreads (in PIPs) of following pairs. London session, NY session, Asian drift session
eg 1.1, 7.0
EURUSD London NY Asian
GBPUSD London NY Asian
USDJPY London NY Asian
AUDJPY London NY Asian
Rollover, or swap costs:
please state + or - and fractions of a PIP, eg +1.2, or -0.3
Hi @Alg626,
Thanks for the input. It seems there are not many members who know what their fees are.
I have since referred to my own notes (about NNFX method) and found the following statement from Episode 78 podcast - the last of the 2019 series of podcasts on the NNFX method. Short answer – ignore currency swap rates. Swaps are relative to how much you have in the markets and affect long term traders (Forex investors) far more than traders. Affects rollover of contracts and at most would be -1.75% to + 1.75%. Never allow long or short swap rates to influence your decision to go long or short.
So for the purpose of settings in the currency pairs for the FT5 testing thread I am going to use no commission, a two way spread of 2.0 pips and no swap fee (for the EURUSD) assuming entry and exit is during London and New York trading hours (not Asian drift territory).
I also found an interesting article on rollover costs at this link, that relates to interpretation of Islamic law and why there is an increasing number of brokers who do not charge a rollover fee.
Unfortunately I’m sure most people don’t care or even know how their swap fee is calculated or how it affects their trading cost through the year/s. And I do agree they certainly would affect a long term holder more so than me. I rarely get a trade that makes it to two swap fees most of mine are done within 12-36 hours.
Of course I did have a swap fee on an exotic last week that was -$170.00. Bout hit the floor. Highest one I’ve seen to this date. It was an exotic but even for an exotic that was very high. And that was on a 1.0 Lot. I think it was my EURPLN trade.
Hi, and thanks for the feedback. Yesterday I added a post to the thread that set out the boundary conditions for my intended tests of the FT5 platform. The timeframe is “D1”, and I have stated I will use a $0.00 swap fee for both long and short. However, I do intend to run some tests that show the impact of a non-zero swap fee by running the same test with and without swap. At this stage, if I do not get specific about what I am and am not going to test we will end up with thousands of possibilities. Early days, and a lot to think about. Collaborative input is always welcome
Spreads vary from broker to broker.
But there is one pattern, floating spreads are usually lower than fixed spreads, but fixed ones give more security in the market.
