Designing an Automated Trading Program

[expiated]

Aaahhh, ok it all makes sense now as to what he is trying to do. NO wonder he mentioned that FX is extremely predictable.

I also trade a tiny OANDA account. I will feel confident that I am trading my system consistently once I have doubled my opening account balance, which I have not done given all the testing I have been doing, but once I finalize all the rules and eliminate all "experimental" trades, I expect to make steady progress toward that goal.

My last 3 OANDA trades, as mentioned in previous posts:

 

[expiated]

MORNING OBSERVATIONS:

USDCHF is evidencing an inclination to climb higher. It will have to maintain its rate above 0.9937 for a sustained period however (perhaps 2 hours?) before I will feel more confident about the pair actually following through.

GBPUSD has turned south, but the longer-term trend has not yet come with it. I am therefore opting not to join this leg of the journey, but will wait instead to see in the immediate situation reverses to come into alignment with the broader context. If that happens, I will enter a long position.

(On the other hand, if the rate crawls below 1.2595 and remains below that level for a couple of hours, I might be convinced to short the pair, depending on what the data is saying at that time.)

EURGBP looks like it might be toying with the idea of heading north. If it climbs above 0.9006, I will probably enter a long position.

 

[expiated]

GBPUSD has turned south, but the longer-term trend has not yet come with it. I am therefore opting not to join this leg of the journey, but will wait instead to see in the immediate situation reverses to come into alignment with the broader context. If that happens, I will enter a long position.

It happened...I have entered a long position. I decided to buy EURGBP as well.

 

[expiated]

I think I'm going to try using the "outside 64 setup" as a pseudo-swing price range strategy with the green moving average as the trigger line and the central channel as the take profit target. If it works, it should lead to much more significant gains per trade with less effort (though it will involve a lot more "waiting around" for payoffs).

In terms of automation, the USDCHF trade leads me to surmise that a logical rule to include in trades involving the green moving average as a trigger line is to exit trades with profit if the candlesticks pull back to the white moving average before take-profit targets are hit. (Or better yet, exit when the candlesticks cross behind the white moving average cluster on the lower-timeframe charts.)

 

[expiated]

It happened...I have entered a long position.

GBPUSD hit my profit target...

But EURGBP is presently dead in the water. USDCHF never followed through and is bearish once again, so I will have to eat that loss. Both of these last two trades illustrate the value (to me) of having an automated system that exits positions (usually with a small profit) as soon as it looks like the forecast is changing rather than being stopped out for losses.

 

[expiated]

I was waiting for the chance to enter a NZDUSD long position, but it appears the pair offered a relatively small window of opportunity to do so.

As currently envisioned, an automated system would have entered the position somewhere between 0.6805 and 0.6809, and then exited somewhere around 0.6814, realizing just a few pips profit. So obviously, the system will need a mechanism for exiting trades following substantial gains before too much is given back.

But now that I think about it, the system would have reentered the position at 0.6815 and would still be in the trade right now. So actually, most of the potential profit would have been realizing anyway. Still, with a “trailing stop” sort of mechanism, the potential profit would not have merely been recuperating, but duplicated.

 

[expiated]

I was waiting for the chance to enter a AUDUSD long position as well, but I should’ve done it 35 minutes ago, or at least 20 minutes ago, instead of writing down my thoughts on NZDUSD.

The structure is such that I might argue I should wait for a pullback, but I’m also working on diagnosing trends that are under the influence of fundamental factors and are subject to forces too strong for pullbacks to manifest themselves.

AUDUSD currently meets those parameters (that criteria), so I went ahead and entered a long position anyway (@ 0.7191) believing that the pair does not intend to be held back or dissuaded in any manner whatsoever.

There were other options such as AUDPY and NZDPY, but they would be fighting against a longer-term trend, whereas AUDUSD was at a point of confluence (decision/pivot point) were a reversal north would be expected.

UPDATE: AUDUSD began to form a red 5-minute candle just below a level I have marked as an area of significant statistical resistance (0.7200) so I felt it prudent to go ahead and exit the trade satisfied with the nine-pips profit I was able to eke out of my belated entry.

 

[expiated]

THOUGHTS...

AUDJPY should be falling instead of climbing, so I will be looking for an opportunity to short the pair over the next few hours. The same is true of NZDJPY.

At 0.6851, NZDUSD has reached a level I have marked as a significant area of statistical resistance. Should it happen to climb as high as 0.6876, I will deem the pair to be under tremendous, almost overwhelming pressure to turn south.

CADJPY can continue south for as long as it likes, even another two or three days, if not longer. However, structurally speaking, I would not be surprised to see this pair turn north at any moment. Consequently, I will be watching the numbers carefully for any indication it is ready to do so, and should the figures warrant, enter a long position at that point.

 

[expiated]

The performance of my Tickmill account was much better than OANDA or NADEX.



The solution to NADEX is easy...I forgot about NADEX when GBPUSD temporarily reversed direction, so I need to remember what I'm doing over there.

Other than that, the only trades I made via Tickmill were ones that did not require a lot of focus and did not need to be made in a timely manner, so I'll need to remain cognizant of this and note if there might not be something about these particular types of trades that leads to their resulting in more successful outcomes (or if this was just a fluke).

 

[expiated]

Several hours ago I made an unsuccessful attempt to buy USDCHF. However, the odds of the pair rising from 0.9928 from a statistical standpoint are much better now than they were at that time, so I am inclined to give it another try. (Though actually, if I had the luxury of monitoring my positions, I could have exited even that trade with a profit.)