› Forums › General Discussions › What is the minimum edge necessary to beat spread?
 This topic has 5 replies, 5 voices, and was last updated 9 years ago by hankee.

AuthorPosts

January 23, 2015 at 7:01 am #4377
Suppose I have a system with a 1:1 RR that gives me, say, 10 pips 60% of the times I make a trade. I intuitively know that with a 3 pips spread, this system is probably going to be profitable. But is it? And what if for example I had a win ratio of 55%? Would it be profitable? And how much?
I suppose there is some mathematical formula to calculate the “expectancy” of the system, but I don’t even know if that’s the correct term…
Thanks This topic was modified 9 years, 5 months ago by hankee.
January 23, 2015 at 9:25 pm #4402Hi Hankee!
If you look at for example 1:1 and if your take profit is 10pips and stop loss is 10pips (including spread), then you will end up with a net 0% gain. So take it from there. You can easily calculate this in a spreadsheet.
What you need to figure out is the profit factor. That is how much you are gaining for each loss. So if your loss is $200 after 100 trades, a profit factor of 1 means you gained $200, a neutral trade setup. But with a profit factor of 2 that means if you loss $200 then you would have gained $400, doubled it, 100% gain. This is how that looks like:
Risk Reward Success % Profit Factor 1 1 50% 1 1 2 50% 2 1 3 50% 3 1 4 50% 4 1 5 50% 5 1 1 50% 1 1 1 60% 1.5 1 1 70% 2.333333333 1 1 80% 4 1 1 90% 9 Attached is the spreadsheet for you to play with
To answer your question, it depends.. depends on R:R and the system’s success %
Attachments:
You must be logged in to view attached files.Focus, Patience, Determination & Order in chaos
January 24, 2015 at 1:53 pm #4411Thanks Saver0, this is very helpful
January 25, 2015 at 8:19 pm #4442Nice tool Saver, thank you!
May 17, 2015 at 11:08 am #6712Hi Hankee, If you’re still monitoring the thread.
I thought maybe this EA might be of interest. I found it with the code all messed up, but with very limited coding ability i managed to get it working something like i think was originally intended.
Basically it opens a pair opposite positions simultaneously so will need broker who accepts same pair hedging. The take profit takes account of the spread for both positions, then adds the tp input value. The loss is set at 2x spread plus your input.
When both positions open and the price moves in one direction one position will go into loss and close out at loss, then hope the winning position hits tp value of 2x spread plus a few pips profit.
When both positions close it open two new positions and repeat.
I haven’t tested on live feed yet as markets closed. In back test it works nicely, most times it take small profit from the dual trades but has enough trades where first loss is hit then price reverse and hit the other loss so it losing slowly.
I think this EA is a good candidate for martingale type mm as we have fixed tp and loss values and it rarely lose more than 3 or 4 in a row (on interpolated tick that is).
Also it tends in back test to win/lose in series of trade type. So if martingale was implemented perhaps filter the martingale to add the increase only to the last winning type. ie if nominal lots is 0.1 and last win was buy the buy in the next pair trades at 0.2 while the sell remains at 0.1.
It would be interesting to see how it behaves. This reply was modified 9 years, 1 month ago by hushmoney.
Attachments:
You must be logged in to view attached files.June 30, 2015 at 10:33 am #7145Today I was thinking about this same issue, and I remembered this thread and i thought “Well, let’s go check it “.
So now i understand what I didn’t at the time: to calculate how profitable a system is you have to include the spread before calculating the risk/reward, not after.
I also always wondered what the “profit factor” was and now I understand, thanks ;)
Hi Hankee, If you’re still monitoring the thread. I thought maybe this EA might be of interest. I found it with the code all messed up, but with very limited coding ability i managed to get it working something like i think was originally intended. Basically it opens a pair opposite positions simultaneously so will need broker who accepts same pair hedging. The take profit takes account of the spread for both positions, then adds the tp input value. The loss is set at 2x spread plus your input. When both positions open and the price moves in one direction one position will go into loss and close out at loss, then hope the winning position hits tp value of 2x spread plus a few pips profit. When both positions close it open two new positions and repeat. I haven’t tested on live feed yet as markets closed. In back test it works nicely, most times it take small profit from the dual trades but has enough trades where first loss is hit then price reverse and hit the other loss so it losing slowly. I think this EA is a good candidate for martingale type mm as we have fixed tp and loss values and it rarely lose more than 3 or 4 in a row (on interpolated tick that is). Also it tends in back test to win/lose in series of trade type. So if martingale was implemented perhaps filter the martingale to add the increase only to the last winning type. ie if nominal lots is 0.1 and last win was buy the buy in the next pair trades at 0.2 while the sell remains at 0.1. It would be interesting to see how it behaves.
Hedging is not really rational from a mathematical point of view. You could replicate any strategy without actually hedging, and sometimes you would pay half the spread (with the strategy we are discussing I think it would be the same). For example, if the spread was 3 pips and the SL was spread+2 pips and TP was spread+10 pips, then with your original strategy a winning trade would look like this:
Put the two orders at P
The price goes up to P+5 and you lose 5 pips from the sell order
The price keeps going up and hits TP of buy order (P + 13), you win 5 pips (10 from buy order – 5 from sell)
You could get exactly the same results with this non hedging strategy:
Start looking for trades at a random price P
When price goes up and hits P+5 you buy (at this point you still need it to go up 3 pips for break even, but at this point in the original strategy, we had 2 winning pips from the buy, but had losed 5 pips from the sell, so we still needed it to go 3 pips for break even)
The price keeps going up to P+13 and you close the trade that you had put at P+5, winning 5 pips. Exactly the same results without hedging.
Anyway, if your strategy had any edge it would be because of trends. If price had gone up for X pips, then it would be likely to continue going in that direction for Y ammount of pips more. That is, in the hypothetical case that this strategy works. I don’t think it does, I don’t think any strategy that simple has any chance of working.
And if you don’t have a real edge there’s not any money management method that would be profitable in the long run. Martingale will sooner or later blow up an account, mathematically speaking the riskadjusted return is zero or negative. The only money management method that actually works and does not simply delay losses (like martingale) is sizing positions according to the probability of the trade being profitable, according to the Kelly Criterion (but it only works if you already have an edge).
Cheers!

AuthorPosts
 You must be logged in to reply to this topic.