Need help solving an elementary MA Cross Problem

Hey guys,

I'm sorry if this has been answered I looked around for a while before I decided to ask it here.

How do you avoid multiple trades when two MA's cross multiple times in one bar? (Because sometimes an MA can cross the slower one several times before it finally breaks through.)

I could make it so that it only trades once per bar but then I would completely miss the trade if the fast MA crosses the slow MA, withdraws, then crosses again later.

I was also thinking that I could tell it to sell or but if the cross stay for a minute or so but a lot can happen in a minute.

There are a lot of MA Cross EA's out there I'm just wondering how you all overcome this.

Thanks. Happy Trading!

What you said doesn't make sense unless you're not waiting for the bar to close on the MA cross. If you are waiting for the bar to close, then there are a couple of things that help but don't solve the whipsaw problem.

First I have to say that you can't avoid whipsaws any more than you can predict the future.

Moving average systems are trend following systems. By their nature, trend following systems have relatively low hit rates (around 40%) but high reward:risk ratios.

There are entry tricks to narrow the hit rate, such as trading the breakout of a bar on an appropriate timeframe in the direction of the cross. If you have to, you can use your favorite squiggly-line indicator as confirmation but I don't find value in them; I'm more of a price action guy.

Keep in mind also that MA crosses are lagging (as are squiggly-line indicators in general). By the time the cross happens it's already too late.

Points well taken. I had pretty good success with an MA Cross EA I made but it lost far too much money. I'm just trying to look at new systems right now to find a good match. It's hard work but I'm making progress.

stop considering the ma's touching to be the signal for both an upward cross and a downward cross. split the concepts into 2 separate boundaries away from the actual cross point.

say you count a fast ma moving upwards and touching a slow ma to signal an upward cross... change that idea around so that the fast ma reaching x points below/ above the slow ma to be the signal, and similarly opposite for a downward cross. instead of the ma's rubbing repeatedly against each other triggering a million signals the ma's actually have to move some distance to trigger an opposing signal.

think of it along the same lines as setting stop orders either side of a range in expectation of a breakout. you don't consider price crossing the half-way point of the range as the trigger as it will just keep hitting over and over, so you set it outside of the range.

To really gain a statistical edge you'll need to start thinking out of the box. These MA cross systems are rehashed in books and articles all the time and have been beaten to death.

Entry strategies like nubcake and I suggested put the odds more in your favor.