real vs "fakeout" SMA crossovers

[yurtrader]

Hi Guys. Here's what always gets me - in the above example there is an SMA crossover, meaning it's a bearish signal and ostensibly a signal to go short.
However, we see an identical scenario having formed earlier that just ended up being a temporary pullback.
Is there a way to tell with better than 50/50 probability what's going to happen next, using all available data/methods?
All I can think of right now is to wait for the SMA 200 to go down, however it went down also in the first example.
thanks in advance

 

[tomorton]

MA cross-overs are poor entry signals, and the result of using them as such is the problem you have illustrated. The point in time at which the MA cross-over takes place has no relation to current or immediately recent price action.

In the long run they can make a modest profit for you, with a very slim margin of error. But that is largely down to buying in uptrends and selling in downtrends, not on buying on Golden Crosses and selling on Dead Crosses. You might just as well spot an uptrend and then toss a coin each day - heads you buy, tails you wait. The results over the long run would be equally good - i.e. equally poor.

Meantime, in an uptrend there is no reliable way to tell if a pull-back will just be a temporary interruption pf the uptrend, or a reversal into a frantic downtrend - except that a pull-back is far more likely.

 

[yurtrader]

thanks tom

MA cross-overs are poor entry signals, and the result of using them as such is the problem you have illustrated. The point in time at which the MA cross-over takes place has no relation to current or immediately recent price action.

In the long run they can make a modest profit for you, with a very slim margin of error. But that is largely down to buying in uptrends and selling in downtrends, not on buying on Golden Crosses and selling on Dead Crosses. You might just as well spot an uptrend and then toss a coin each day - heads you buy, tails you wait. The results over the long run would be equally good - i.e. equally poor.

Meantime, in an uptrend there is no reliable way to tell if a pull-back will just be a temporary interruption pf the uptrend, or a reversal into a frantic downtrend - except that a pull-back is far more likely.

thanks tomorton, could you point me in the right direction - currently I'm searching for some checklist of entry/exit conditions that will better and more consistent results

 

[tomorton]

Hi - I like the simple approach but you might wish to adapt these basic principles - for entries in an uptrend -

• wait for a day with a lower daily high
• set a buy order just above the high and a stop-loss just below the day's low
• adjust your position size according to the entry-SL distance so that only a small percentage of your account capital is at risk
• if the order is triggered, exit at the close of the third consecutive day with a higher close
• if the order is not triggered, reduce the entry price level to the new lower high: remember to re-set the SL and adjust position size accordingly.

 

[yurtrader]

Hi - I like the simple approach but you might wish to adapt these basic principles - for entries in an uptrend -

• wait for a day with a lower daily high
• set a buy order just above the high and a stop-loss just below the day's low
• adjust your position size according to the entry-SL distance so that only a small percentage of your account capital is at risk
• if the order is triggered, exit at the close of the third consecutive day with a higher close
• if the order is not triggered, reduce the entry price level to the new lower high: remember to re-set the SL and adjust position size accordingly.

thanks so much for your reply! I understood all except the very first point, could you provide some visual examples?

• wait for a day with a lower daily high

since from my niewbie's ability to understand all of the above would be lower highs.

 

[tomorton]

Entries -

See the D1 chart for GBP/CHF.
Its a clear uptrend so the plan is to buy when there is an opportunity. But price has been rising with energy and it has made 9 consecutive higher highs (and higher closes, most unusual). Eventually it made a day with a lower high, B: this is Day 2. Day 1 was the last day in the sequence which made a higher high, A.

At the close of Day 2, set a buy order at B with a stop at C.

On Day 3, price did not rise through B so the buy order was not triggered. At the close of Day 3, lower the buy order to Day 3's high at D and re-set your stop to Day 3's low at E. Not forgetting to adjust the size of the position as the entry-stop-loss distance is now different.

The next day, Day 4, price rose through D and triggered the buy. Price rose on Day 4 and the next 3 days.

 

[tomorton]

Obviously, this is the skeleton of a strategy. Every trader will vary their precise entry and exit timing and their stop-loss placing, and whether they use pyramiding or TP's and so on.

 

[Dowser]

View attachment 297402

Hi Guys. Here's what always gets me - in the above example there is an SMA crossover, meaning it's a bearish signal and ostensibly a signal to go short.
However, we see an identical scenario having formed earlier that just ended up being a temporary pullback.
Is there a way to tell with better than 50/50 probability what's going to happen next, using all available data/methods?
All I can think of right now is to wait for the SMA 200 to go down, however it went down also in the first example.
thanks in advance

You could look at candlestick patterns on a higher time frame (i.e. weekly) to determine whether the new trend has legs or not. Couple this with price action around significant or potential areas of support/resistance (eg. weekly highs/lows) and I think you'd become more confident in your prognosis.

 

[Dowser]

Here's my take wrt my previous post: looks like 120 is a Pivot area and for the last 2 weeks has acted as support. The shallow bullish trend line is also providing support. I'd like to see 120 broken this week or I'd be inclined to think this was just a Pull-back.