MACD divergence on 28 November

[fxmktlgc]

Another thought on MACD (or other indicator) divergence: it may be a good exit strategy. It allows you to exit at a relative peak or trough, instead of waiting for price to retrace against you (e.g., MA crossover). In addition, there is a decent probability that price will reverse anyway.

[initialsBB]

Somebody proposed the following strategy: after you see MACD divergence wait for the MACD line to cross the zero line. Then enter a position in that direction.

[OneClickAway]

The best MACD divergences are when prices are making higher highs or lower lows after a long run and MACD histogram is not, especially in combination with a strong support/resistance level. It is a good entrance strategy and a good exit strategy, because this divergence shows that momentum is slowing before the price shows it. So, in effect you have a small window into the immediate future...

BB --- there are three entry points for MACD as an indicator if there is no divergence:

Fast When the histogram starts sloping the opposite way from below or above
Medium When the Signal line crosses the MACD line.
Slow When the Signal line crosses Zero.

Fast is better to get in early but gives more false signals. Slow is a more sure bet, but most of your potential profit may be gone by then...so maybe Medium might be the way to go,,,,

A.

[initialsBB]

Originally Posted by OneClickAway

Fast When the histogram starts sloping the opposite way from below or above
Medium When the Signal line crosses the MACD line.
Slow When the Signal line crosses Zero.


A.

Where did you get this (i think wrong) classification from?
When it comes to MACD, really effective signals are given when MACD line crosses zero line. In cases where there is little hesitation in the beginning of a trend, it can even anticipate what follows.
Have you seen how this translates on the 15M frame? It works good even on 1H chart. On 8 December Cable rose 180 pips, the cross of MACD line with zero line was in the very beginning of the trend, at 1.7357. Was this a slow signal? I don't think so. It caught the move alomost 100%

[OneClickAway]

Yes, definately I agree with you. There are many examples we can find that supports what you are saying. But the trick is to research that MACD zero cross where it is consistantly giving you that early signal that you are looking for and to rely on it for consistant profit over a given time period.

Consistant profits require knowledge of several indicators/chart patterns and knowing how to interpret them in certain situations.

If we only stick to the strong trades....and let the others go, we will make money.

A.

[birdjaguar]

Originally Posted by OneClickAway

The best MACD divergences are when prices are making higher highs or lower lows after a long run and MACD histogram is not, especially in combination with a strong support/resistance level. It is a good entrance strategy and a good exit strategy, because this divergence shows that momentum is slowing before the price shows it. So, in effect you have a small window into the immediate future...

BB --- there are three entry points for MACD as an indicator if there is no divergence:

Fast When the histogram starts sloping the opposite way from below or above
Medium When the Signal line crosses the MACD line.
Slow When the Signal line crosses Zero.

Fast is better to get in early but gives more false signals. Slow is a more sure bet, but most of your potential profit may be gone by then...so maybe Medium might be the way to go,,,,

A.

Sounds like 1 contract on the histogram change, 1 additional contract on the signal cross and a third contract across zero may be a good idea. On the false reversals you only lose one contract. On the full reversals, you gain three.

[OneClickAway]

Birdjaguar

Yes, scale in one contract (or one lot) at a time...sounds reasonable to me.

A.

[SoundFX]

I've got another suggestion. Call me crazy. After you spend a few hundred hours staring at charts and being able to spot divergence immediately, ya do this...


Use 30 minute chart. Either use histogram divergence as well explained in the above post, or full MACD divergence, then enter 1 short lot (if divergence points to a short position. Then, set an OCO order to either take profit 100 pips below the entry point OR sell another contract if price moves against you 100 pips. If THAT order is hit, then set to close out both contracts at the original entry price, thus netting you 100 pips. However, if price moves against you 100 pips from the second short, then sell 2 more, and place TP order at the second position. This is a Martingale spin-off, but not quite as dangerous as a typical Martingale strategy, as divergence is a huge indication of trend reversal, even if the market keeps skidding for a bit in the opposite direction. All in all, you're constantly looking for 100 pip gains. In a properly funded and diversified account, this works quite well.

-S

As a real time example, I'm looking at the GBP. I've sold at 1.7520, with a TP at 7420. But if price climbs to 7620, I'll sell one more and set exits for both lots at 7520, netting 100 pips. We'll see how this one plays out.