Chinook's EUR/USD (E/$)Mumblings

Seems like a lot of "micro-chopping" right now - a lot of trading - similar to yesterday after the rise; and tug-of-war on every single tick it seems. Uncertainty in the markets because of no strong sustained trending, I believe.
Remembering it's friday again, I will be very conservative about getting into trades ... actually, I probably won't watch too much of the action from here to not get too tempted again.
Nice week, and the action has been quite ok, so perhaps we'll see some better pressure, trending around these new higher levels. A close over 1.2520 would certainly bode well for the €.

Good luck trading, and remember to take profits when the markets are uncertain.
It's so hot here now, I'm gonna take a dip in the pool.
:cool:
 
I took the day off and slept in really late this morning! This morning's activity was surprising to me. Bonds and EURUSD went in different directions after 8:30 am EST. Right after the # release, they were coupled but then some people used the dip as buying opportunity after 6-7 minutes.

Lately, there is too much internal chatter in my head. There's internal negative feedback regardless of making or losing money in a day. If I lose money, the chatter is like why did you take that trade...If I make money, then the little voice starts saying why didn't you exit with earlier with extra profits. It looks like my ego is getting out of control. I'm planning to just watch the markets for 2-3 days next week. I'll paper trade my system and simplify it further, possibly down to 2-3 potential trades per day. I overtraded for a while taking many impulsive trades overriding my system. I almost forgot that the reason I trade is to make money and not to be right all the time.
 
Quote from chinook:

...
I'll paper trade my system and simplify it further, possibly down to 2-3 potential trades per day. I overtraded for a while taking many impulsive trades overriding my system. I almost forgot that the reason I trade is to make money and not to be right all the time.
More...

Overtrading will stab you in the back always; it's the worst of all sins to me. It either will run your losses like crazy, or make you a sorry "winner" with all the commission.

Try identifying "time-zones" when trading is safe - and learn: what are the signs of transitions into other "time-zones". With "time-zones" I mean market conditions. It's much easier to trade certain strategies when certain conditions are there.

Identify those conditions - and keep identifying them always; when they start - when they fade away. Don't lilt this to just the "good conditions", also identify the "bad conditions". That way you can at least get a much higher probability, and this should filter out what times and conditions are not applicable to your type of trading. Use this as the first discretionary control, a sequence of binary questions - yes/no. Make a checklist when you know some of the good zones, and how to identify them - how long they last - and if they are ending - it's not pausing - it ended, and you will have to find a new starting zone again.

Then use the checklist to allow you to trade. Do this for the paper-trading.
I have this kind of "mental checklist" which prevents me from trading at certain conditions, and compels me to trade at other conditions. Then I can allow me to experiment and take some risk in other conditions, because I feel I can at least control my trades in such a way that they will not hurt my equity.

You could still report your findings on these "right conditions" for your trading style while studying, papertrading - I think it's good to discuss it, although I'm probably not the best one to comment with my style of trading. We're not really that far apart - as you look for the big intra-day moves ... the highs and lows of the day. For this to be successful I think you need very good analytical skills, and knowing to identify the markets and conditions which prevail - i.e understanding what's happening is first and foremost, don't you think ?

Just don't let the chatting take your concentration away - do it when you've already identified an "untradable zone".

Find those triggers which will get you into "homing, target seeking mode" BEFORE you look for an entry, then find the entry and identify target or trailing strategy etc. that you use.

Did any of this make any sense ?
:)


edit: Here's an example of the "zone" which was working well for me for a part of the day...
Right after the first numbers, new highs were set and volatility set in with some ok trading, but light volumes in the books. The "waves of pressure" was sustained for at least 10 points up and down with no particular defenses. The directional changes were easy to identify because of stalling which triggered new waves of directional biased traders, when they started taking profits - that would trigger a directional move backwards again, either with those not taking profits getting back in if they were stopped out, or the other directional biased camp getting in. The losers in all of this was of course those who were looking for bigger profits, which only would happen if someone extended the range a little. Well, those extensions were also good entries as they were quickly stalling and reversing. I had 4-5 times where I would enter on best bid/ask but not get a fill, just to see it run from me .. and each time it went AT LEAST 10 ticks. It made me a "little frustrated", but I maintained my calm and got 3 of them right for 3x 10 ticks today.

Later on things started to get more tightly traded, and rather than trying my luck on traders which might have a lot of uncertainty about the direction, and no trending bias evident, I gathered that choppy conditions would start, where no clear signal would be easy to use as entry. Well, the range was there, and sometimes ranges changed a little, but basically it was the normal doldrums with less turnover - just like normal lunch-hours. Since it was a friday, I would not risk getting into a trade when I just made 30 ticks on 3 trades and try to ruin that. I traded "mentally" a few times, but it was so-so .. therefore I judged that I would probably end up over-trading, and prohibited any further trading by taking a dip in the pool since conditions did not seem to be returning - esp. because of friday.

Phew. :D
 
Thanks Gringinho. This all makes sense.

I agree about defining what you call the time-zones. For me it's the transition from range trading into trending. Mainly, I'm looking for volume correlation with the price. Of course this doesn't work all the time so to prevent me going against the market I use the adaptive MA as short/long filter. However, eco # release days are a different beast and I work in a shorter time-frame after the numbers come out.

I think you have more opportunities in a day since you go typically go after 10-15 pip moves. I need to sit tight and wait hours for my entry conditions. In a non-news day, I'll have at most 2-3 entry conditions. This means that I have to accept missing good moves and still stay focused. This is what I've been having trouble with lately. I start taking non-system trades then I don't have specific exit rules for these and then start blaming myself !! I need to improve my self-control-- similar to what you did today and decided not to trade any more.
 
Quote from chinook:

Thanks Gringinho. This all makes sense.
I agree about defining what you call the time-zones. For me it's the transition from range trading into trending. Mainly, I'm looking for volume correlation with the price. Of course this doesn't work all the time so to prevent me going against the market I use the adaptive MA as short/long filter. However, eco # release days are a different beast and I work in a shorter time-frame after the numbers come out.

I think you have more opportunities in a day since you go typically go after 10-15 pip moves. I need to sit tight and wait hours for my entry conditions. In a non-news day, I'll have at most 2-3 entry conditions.
...
More...

Yes, I called it "time-zones" but only because they're time-limited; i.e they end fairly quickly. When the zone ends, it doesn't mean that I end the trade - just that I can't do an entry again. So then I REALLY watch my profits and will be VERY conservative about letting profits wane.

I normally go after 4-6 ticks in the markets of late, but will many times default to 2 ticks if I get too "jumpy",uncertain. Otherwise I think roughly 8 ticks is the best opportunity ... so I will look for conditions that will allow me 4,6,8 ticks. Then I might get lucky with some stronger move, or extension of range etc. - but that's the exception.

Today's numbers made for very good conditions, where at least 10 ticks were quite "easy". Sometimes during numbers I might "scalp" 30+ ticks in one trade .. like NFP-release in start of september. I just ride these simplistic "wave pressures" until they stall. During such conditions I will take profits IMMEDIATELY on first sign of stall, reverse - although I might use a little discretion about where the move might extend if it just stalls a little for profit-taking.

Also, scalping makes for a lot of opportunities, but also a lot of "fake conditions" which start out looking nice, but is just bluffing. E.g big volume letting smaller volume ahead being taken out then runs stops on those. I'm not so good at trading those conditions yet, but am seriously looking into how to get these type of "patterns" into my book of "successful conditions".

This is basically what I was talking about when I wrote in Technical Analysis,Software threads about trying to identify patterns on the tape and orderbook. These conditions are - I believe - possible to identify programmatically; thus letting my get a "green flag" on entries, or telling me when a condition is transitioning into something else.

I just want to have this "recognition automation" of market conditions which I then use for finding entries, direction and target. That will possibly give me a "better edge" like I talked about in the "floor traders bombing out on the screen"-thread.

To describe these conditions I'm experimenting with new and "aggregated" data which can help me just like traditional indicators try and describe other "phenomens of price,momentum".

:)
 
Another research project for you, perhaps you can use neural networks in your programming. I'm not too familiar with programming nets but it sounds like it might be worth to experiment in your case. The order book gives you certain patterns clearly, and you can also extract the most recent trading range and volume and feed it into the net. One problem is though to get the historical bid/ask data in the orderbook.
 
Quote from chinook:

Another research project for you, perhaps you can use neural networks in your programming. I'm not too familiar with programming nets but it sounds like it might be worth to experiment in your case. The order book gives you certain patterns clearly, and you can also extract the most recent trading range and volume and feed it into the net. One problem is though to get the historical bid/ask data in the orderbook.
More...

I use neural-networking, Bayesian networks, Hidden Markov Models (self-organizing maps), Prolog-type rules and a whole lot of other stuff to try and get those patterns identified. It's still dependent on getting some "aggregated, clalculated data" right as the input ... which is what I'm doing for the moment.

If you look at my software-related threads you will see that my applications record all orderbook-changes .. that's essential to me - and also makes the data-files very,very large.

edit: I forgot to mention that these patterns of course change all the time, so what I'm doing, trying to do is having one process identifying what kind of conditions I want to have identified by looking back on movements and a set of rules that need to be triggered. Then I will (eventually) try and get some good learning out of the time-series I calculate (the aggregated data, looking at turnover rates, orderbook movement etc.), along with more basic data and try and get some identifying patterns - "emerging conditions" which is confirmed against the results of backwards looking process - i.e the output of the prediction, identification.

Getting all this straight and working is a "work in progress" for some time, and I'm often busy trading and living too. :)
 
Just watching the crawl higher since I can't really see a good, clear entry. As a position trader I might have gambled before the numbers, or tried the bounce earlier (the 1.2460 support), but would have to get up 5 mins earlier in the morning for that bounce.
:)

Hehe, Thomson mentions Soros-conspiration theory in US press on the weaker USD as an attack on Bush ... lol.

A little low volume still .. will have to wait for some clear range or trend to begin forming before I get in.

A little lower foreign activity ahead of the election, as could be expected I guess. The increased uncertainty in the US makes it tougher to get the right calls for equities etc.
 
I'm in simulation mode today. But I would have entered in a long position after a low volume 15-16 pips pullback to about 1.2510.

What kind of info do they give on the Soros conspiracy theory?
 
You must log in or register to reply here.
Back
Top