Clemmo's Trading Odyssey

"Success is a lousy teacher. It seduces smart people into thinking they can't lose."
- Bill Gates

I had no right to these pips but I got them anyway. Bad trader! Bad!

Mistakes:

1. Trading right before BOE announcement
2. Not checking higher TF trend
3. Not waiting for sufficient deviation

But I'm hot right now and seem to be unable to do anything wrong. There's something dangerous in my personality (subconscious?) that wants to TEST that luck, actually check if I'm invincible (clearly not) and I don't know why that is. It's my naturally contrary nature, I guess? Usually it's a strength in trading, but when you're winning it's better to cheer for the favorite, not the underdog.

A good read. Best of luck in your journey.

Hello! Thank you for your thread ! I am taken with your analogy of treading water. Analytical trading statements are not nearly as fun as an analogy, especially if it involves treading water in the ocean. My theory is that I at least, as a trader, am unique. I know this is statistically less likely then I would like to think but hey. Whilst you are working on algorithms to predict functionality of a person(trade) treading water, my personal take is that if I am going into the ocean, I'd prefer to take a step ladder ie I look for optimal entry, tight SL, ready escape, quick abort... for me as a manual trader it means I tend to swamp trades, so my trading costs are high, and I'm yet to maximise my profits ( I TP too quick).. but im working on it . As a visual i find your analogy quite appealing, and a random selection of 20 traders embarking on an ocean swim holds plenty of appeal. Thanks for your contributions, I have a strong sense that I will never be an EA trader, but good luck and I shall follow!
Cheers h

The stepladder in an ocean analogy is a good one. As traders we have to try to find the pools of liquidity that won't swamp our ladder, which is (usually) puny compared to the overall ocean (market). Knowing when a trade is 'too deep' and when we can climb out of it, is at least half the battle I think. At the same time we don't want to be scooping teacups out of the ocean because that's inefficient.

DXY falling convincingly since I wrote my post about swaps. Will it continue or is this just a blip? The FOMC has 'officially' thumbed their noses at Trump but if he polls well in the re-election... And he has raised a shocking amount of money for his campaign already. Then again, I wonder if he really has anything to do with the price of tea in China, the Fed may need to cut if employment naturally slows down. This kind of navel-gazing is why I generally disregard the effectiveness of fundamental analysis; all we can hope to do is react quickly and proportionately to what's actually happening.

Damn, Bitcoin, why you gotta make me look bad?

The power of a reverse indicator.

One of my brokers subscribes to a 'trade of the day' type service where a 'Genuine Forex Expert' emails me one trade notion per trading day. The email includes a thesis about why the trade is a good idea, and it gives an entry, target, and stop level for each trade. Their trading instructions are to wait for price to hit the entry, and if it fails to do so to discard any pending orders.

So I've been tracking their performance since I signed up. I simply use Gmail's 'star' system to label the incoming email.
I allow about 5-10% grace so if price gets close to the target without actually reaching it I'll still classify that trade as a win, otherwise I let price hit the stops.

If you can see in this screenshot, whenever their trade hits the target I give it a green checkmark. (9)
Whenever the trade fails to reach the entry level I give it an orange chevron. (25)
Whenever the trade hits the stop-loss before reaching the target I give it a yellow exclamation point. (31)
31/40 = 78% loss rate which is pretty...pretty...bad. Suspiciously bad.

Hopefully you can see why my broker likes this service. They are keeping lots of their client's money for anyone who follows this straight. For me I appreciate a good trade idea even if I have to reverse it first.
Identifying details have been blurred to protect the guilty but PM me if you're curious.

Tips for Chess beginners from IM John Bartholomew accompanied by my trading interpretation

1. Stop blundering (stop making basic errors like trading against the trend, failing to take into account info from all TF, failing to check the calendar)
2. Patience (Patience - don't trade when there are no good trades)
3. Don't overreact when things don't go as expected (Don't revenge trade, don't martingale, don't immediately exit unless that's your trading plan)
4. Make decisions objectively even in losing positions (Experienced traders know how much pain their account can sustain and can trade out of bad positioning, they also stick to the plan even when they're down)
5. Look for the best move at all times (Look for the best trade at all times. Trade only when the market is offering you clarity; Look for the least ambiguous trades; pair weak with strong at the right time)

Thanks to BP user Macilme who suggested I should check out Darwinex and open an account. I had checked them out when they first started but there was something that kept me from sticking around, maybe a restriction on Canucks? It seems like they've removed that now, yay!

So I've done that, and wow, Darwinex is pretty cool if nothing else. Stats, stats and more stats. It will take me a year or two just to wrap my head around how everything works but for now I'm curious about how some of the top-performing Darwins do their magic and I might do an article on what I find out, if anything.

I'm thinking about taking July/August off to focus on my C+/MQL coding training. I need to be able to build my ideas.

For one thing, when I get tired I get careless. As you can see from the way I finished this week I would have been better off studying.

I thought I posted a 1.88 call for GBPNZD but I guess I never did, sorry guys.

Anyway, 1.883 was reached and now GN is at a a crossroads. It could certainly drop further to get close to the Dec. low (as I mentioned) but it could also decide to retrace for a week or five. Although these points don't look any different on a chart, they are actually very dangerous to trade.

Let's see if the building permits news coming out today will give us any clue as to what happens over the next few days, but I don't think orange news is very meaningful by itself.

Here is a (partial) strategy that would have been profitable for at least 4/5 5/8 weeks between Feb. and April 2019 (that I've tested so far) and I'll bring it up to date when I have time. Now updated.
https://docs.google.com/spreadsheets...it?usp=sharing
All you have to do is trade against the pundits who make predictions on forex websites like FXStreet. Look at the forecast poll on Friday. Place 13 trades in each of the instruments listed, preferably on Friday but Monday would probably still work, and exit your trades at close the following Friday.

Of course there are still many details to work out, such as stop placement, but I think it could work without adding much more complexity.

I don't know how so many people complain about being unprofitable in Forex for many years. It seems if you work hard, pay attention, and ignore the gurus there are many paths to success.

Update: now that I've caught up with the data that are available to me the results are starting to converge towards the median and the performance of the 'experts' is beginning to look merely random rather than maliciously contrarian. So, make of this what you will, I'm going to take the most recent poll as a starting point for selecting instruments.

https://www.forexfactory.com/showthr...4#post12375594

Good old wimpy gold. You can always depend on it to back down when the going gets tough.

https://docs.google.com/spreadsheets...it?usp=sharing

The forecast poll spreadsheet is now up to date. So far they have failed to get a simple majority of directional calls in 9 out of 13 weeks, in other words a 31% success rate. That definitely seems worse than random to me, so I might keep the sheet updated if anyone else is still interested, just let me know, otherwise it will stay up out of historical interest.

What is interesting is how much better the monthly forecasts are than the weekly, better than 50% even though there is theoretically more time for unexpected developments and price deviation. The quarterly forecasts definitely suffer from those pitfalls, but the monthly ones, oddly, have better accuracy than the weekly.
Now this could all be randomness and noise, since there isn't enough data yet, but I wonder...

One theory is that forecasters overestimate price moves in the short term - expecting price to have moved significantly one way or another when in fact, within a week, it's a simple thing for price to retrace back to where it started or worse. Might be an argument for

1. trading on a monthly horizon
2. not making forecasts at all

The next provider to evaluate is ScorpionFX who make some grandiose claims on their site, but who knows?

Idea:
(FX) Robot Fight Club.

There are lots of forums dedicated to 'learning' forex trading and a few dedicated to learning MQL coding. There are none that I know about dedicated to EA testing. EAs, when they work, are marvelous, but even better than having an EA, is having a tested EA. In fact without testing they are deadly dangerous.

So the idea is there's a (secret) forum (the first rule of Robot Fight Club is don't share) and we chip in our ideas for an EA that needs to be built then we vote on the most promising system and get it made by a pro developer (which we also pick through a vote) and pay the dev through the forum (monthly) subscription fees. Every paying member gets a copy of the finished product. Submitting tests and settings would earn you credits and rep. Pay more and get the source code, get naming rights, etc.

The main thing is, unlike a forum where you donate so that the site ops can get drunk on Fridays you would actually get something USEFUL for your $ that will in turn would allow you to support the site even more and speed up the development of new robots, creating a virtuous cycle.

Quit whenever you want. Don't pay your dues and you don't get the new bot. Simple, no harm no foul. Rejoin later, invite your friends. However the club membership would be capped to keep things manageable and to keep the systems from getting leaked or too popular.
The subscription fee does a bunch of other useful things too, like:

1. Keeping out the riffraff and trolls
2. Keeping the system(s) somewhat secret
3. Keeping members invested in the process and coming back to report new findings instead of wandering off to find another FX forum.

What do you think? The fee would be a function of the number of interested traders and the cost of the developer. All proceeds would go to paying for development or testing, apart from the nominal costs of hosting the site.

As promised.
https://docs.google.com/spreadsheets...ELk/edit#gid=0
https://docs.google.com/spreadsheets...#gid=534792429
https://docs.google.com/spreadsheets...gid=1286854520

Win rates appear to be random and although the weekly percentages are healthy they appear to be cherry picked, especially glaring are the missing GBP trades from the second week, esp. since the free signal was a spectacular GBPUSD long loser.

So, not interesting to me, moving on.

I seem to have stumbled on a profitable strategy that uses the free MT4 indis of 'OBV' (on balance volume) and envelopes.
Gold trendlines are profitable entry and exit points, magenta lines are losses.

Interested?

If I knew about EAs I'd chip in. Money management and simple PA seems to be all I can manage lol. Good luck.

Like most things in life, if you want something done right you probably have to do it yourself, so although it will take me 20 years, if the world and forex trading are still around, I'll eventually get all my systems 'automatized'.

But just in case anyone's interested in doing it first, here are the rules. All I ask is you send me the demo/live trading results!
https://docs.google.com/document/d/1...vFMhZuARU/edit