Strategy Tester: Should allow option to change spreads - page 2
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Yes like mt4. The current solution is much worse. If I had a choice between this and mt4 setup, I'd choose the mt4 setup.
My experiences have brought me to the world-view that the Average_of_Something says more than the maximum in cases like these.
However, what I'm asking for is an Option, I'm not asking that it gets replaced, tho should I have-to choose 1, I'd choose the later.
Still not convinced. Take a test on 2013, from 2013.01.01 to 2013.12.31.
With MT5, you have to highest spread, true, but for each minute data ! You can have a peak on a minute from time to time for one minute bar, but most of the time the spread is realistic.
With MT4, you can fix a spread, but it will be the same during all your test. That has some advantages if we talk about testing/debugging an EA, but it's a disadvantage if we are talking about evaluating a strategy in the more realistic way (I mean compared to what if would be on a live chart).
In my opinion, spreads are a big problem just for scalpers.
For the rest of traders, spreads are not relevant today if you have a good strategy and a fair broker.
Imo, spreads are still very relevant today because its part of a good strategy.
Most successful scalping strategies I've studied capitalizes upon the spreads.
Most so-called good strategies eventually turns into losing strategy as a result of the spreads catching up with them.
i agree that scalping must have low spread
but for normal strategy that should not be a problem right
what are the case where there are high spread? News and low volume. What else?
Still not convinced. Take a test on 2013, from 2013.01.01 to 2013.12.31.
With MT5, you have to highest spread, true, but for each minute data ! You can have a peak on a minute from time to time for one minute bar, but most of the time the spread is realistic.
With MT4, you can fix a spread, but it will be the same during all your test. That has some advantages if we talk about testing/debugging an EA, but it's a disadvantage if we are talking about evaluating a strategy in the more realistic way (I mean compared to what if would be on a live chart).
Well, a back-test / simulation is never realistic, it's theoretical. The data in mt5 is-not my broker's data and the spreads in mt5 is-not my broker's spreads.
If someone did a simple algorithm like [ if( minute==30 && spread<20 ) Buy_EurChf ]. //Buy once Per_Hour@30th minute when spread < 20 points || 2 pips.
If this strategy places virtually no-trades within the strategy-tester but then become very active on meta-Quotes demo-account ... how realistic is that?
Should I want to check if my strategy can over-come 2-Pip-Spreads, I would set the tester for 2-Pip-Spreads.
Should I want to check if my strategy can over-come 4-Pip-Spreads, I would set the tester for 4-Pip-Spreads. etc.
I really do-not look at that like a debugging ea analysis, rather, I view this as important what-if analysis.
Hope my above example isn't too extreme. This is just to convey the sense of what I believe is happening with the strategy-tester spreads.
i agree that scalping must have low spread
but for normal strategy that should not be a problem right
what are the case where there are high spread? News and low volume. What else?
Spread is a problem for everyone. The only people who this isn't a problem for are those who think they can predict what direction the market is going to move in.
We could spend all year on this topic and no-body is going to show-up here with Proof that they can perform the above.
The mathematics of spreads however is very basic in my Opinion. Meaning 0-Spreads and even a blind man can trade and break-even in the long run (as in not-lose).
Most people trading with take-profit of <20 Pips are really trading with odds which is worse than Casino-Roulette, I would consider 10-to-20 Pips normal.
If someone wants to trade with odds which are better than Casino-Roulette, they need to set targets which are greater than 20-Pips (the more the better).
But this doesn't mean that the spreads become irrelevant, rather it just have less influence in % which keep decreasing but never goes away.
There is a premium for asking for more profit ... and that cost is paid in time. Meaning someone would need to place trades less often. No-guarantee wins either.
Its like saying, do-not take $100 every-day to the Casino-Roulette. Rather take $100 once a Month. <-This does-not make it a better Venture. You just lose less.
Looking for a smaller spreads is therefore like looking for a Casino which offers Single_Zero on the Roulette Wheel as opposed to Multiple-00.
Talking about good strategy is like talking about someone who could predict Roulette-Numbers.
In that case, it wouldn't matter how many 0 or 00 were on the wheel. As-in, it wouldn't matter what the spreads was.
Because this person would just make the correct call (all-the-time) or (a majority of the time).
Well, a back-test / simulation is never realistic, it's theoretical. The data in mt5 is-not my broker's data and the spreads in mt5 is-not my broker's spreads.
If someone did a simple algorithm like [ if( minute==30 && spread<20 ) Buy_EurChf ]. //Buy once Per_Hour@30th minute when spread < 20 points || 2 pips.
If this strategy places virtually no-trades within the strategy-tester but then become very active on meta-Quotes demo-account ... how realistic is that?
Should I want to check if my strategy can over-come 2-Pip-Spreads, I would set the tester for 2-Pip-Spreads.
Should I want to check if my strategy can over-come 4-Pip-Spreads, I would set the tester for 4-Pip-Spreads. etc.
I really do-not look at that like a debugging ea analysis, rather, I view this as important what-if analysis.
Hope my above example isn't too extreme. This is just to convey the sense of what I believe is happening with the strategy-tester spreads.
I see. Maybe you can try this suggestion to Metaquotes, not sure of your success.
I would love to have it. I'm currently developing a scalper and the spread is very different between live and backtesting. I can adjust it to be profitable live but only if I can customize the spread while doing my backtesting so I can see what's profitable with higher spreads. My average live spread is 20pips but backtesting gives average of 8pips. Makes a massive difference!
I understand that everyone that don't produce non-scalp strategies don't need custom spread but for those of us who do it would really help! Thanks for submiting your suggestion to MetaQuotes Ubzen.