What I missed was that the hourly candles are starting on half hours (which comes from being new to the contract). It's not that the candle wasn't closed, just that it hadn't been open very long, so the movement wasn't significant. Oh well, we live & learn (& admit our failings).
Ninja still keeps losing connection, sometimes for seconds, sometimes for hours, sad times.
Last night/this morning I tested a strategy based on an observation, it was not a large sample (it was 2 in the morning, cut me some slack), but it definitely warrants more testing. So I thought this would be an appropriate time to broach the subject.
Many people are under the impression that automated/systems (henceforth systems) trading doesn't/can't work. They're wrong. Now what is correct is that 99.99999999% of the retail systems which are sold don't work, and that the vast majority of systems posted for free on the net don't work, but that is true with traders as well. I also think that systems get bad mouthed by trading coaches, because it threatens their business model.
Where I think most people go wrong with systems is that they just don't go about it in a logical way. You can certainly tell that some people just chuck stuff onto the screen, it's amazing how many people use macd, but don't know what it is. Which leads us to another point, systems trading seems to be synonymous with indicators, aforementioned strategy uses no indicators.
I think there are two ways to go about it.
Firstly: Quantify your discretionary trading (if you're successful), into solid rules. You need to define what a trend is, what a high is, what a pin bar is, whatever it is, it needs a definition if you're going to automate it. So many people miss stuff out, it's like that making a pot of tea thing.
Secondly: LOOK at your chart, your T&S, your ladder, whatever, observe, note patterns, figure out how to make profits from these patterns.
A few more general notes.
1-Don't over complicate it. You don't need to, there are many simple things which work.
2-Pay particular attention to session open/close and news releases, the same things happens time and time again.
3-test test test test test test
These are more opinion based, open to debate, and related
4-I am rather convinced there is a inverse correlation between expectancy and frequency.
5-I view systems trading as a way to give me less work (in the extreme long run, ha), and to allow me to trade multiple markets and eliminate liquidity issues as a factor. It's not there to make much more money than I would with manual trading.
6-Not everyone who trades discretionary can make a profitable system, and visa versa, but I would think most who can do one can do the other if they take it seriously.
Still not got a connection, think I probably have a trade open still, who knows. Might not have results for today :-(
P.S. To show what warrants more testing to me, results were (in ticks):
Had a horrible day, testing going poorly, didn't get much sleep, people annoying me. Nearly ended up having a stop twice the size it should be (see pic).
I am thinking of moving to oil, I knew it was volatile, but had a look last night and it's 10 times more volatile. I find it hard to imagine hitting the $3k drawdown or missing the sharpe of 4 for Pulsar, so the extra volatility only makes it easier really (cetaris paribus). I'll have a look over the weekend. I think I may have a nap and try and write some more method stuff later.
I missed the bounce off the low before close yday, ad I wasn't even looking at screens (I thought it was a lot earlier than it was), so only had one setup, 2 trades. Maybe I should note that, I often get shaken out of trades only to get back in, it doesn't bother me, I'd rather take smaller more frequent losses than have wider stops which get hit less. I think that was a big lesson, learning that just because you've lost money on this idea, that doesn't mean everything about it is wrong (just your stop/timing etc, which are of course very important). Again, not articulating this well, apologies.
Short 1372.25 to 1373
Short 1372.5 to 1.367.5
Again just scraped the $200, but wasn't enjoying today, no news, bit messy. Just scraping the $120, but not happy, pretty depressed, so not wanting it to influence trading too much.
Dovie'andi se tovya sagain.
Last edited by Hotch; March 2nd, 2012 at 12:53 PM.
So as stated, looking at crude, as over the 25 days, would have to get 2*daily ATR as opposed to 20.
Simple stuff leading me to think that crude will go back up to $114. Daily bull flag broken, top of which was a significant level. Hourly got a 4 touch trendline for buying opportunities.
The bull flag covered $11. Top of the bull flag was $103, $103+$11 = $114, 2012 high. Make sense?
Well it shouldn't! If you're doing a measured move like that, you need to take it from the bottom. So it'd be from 95-106. You could multiple it by a fib and fudge it, I don't do much so I'm not an expert, what you should take away is.
1-If you haven't learnt already, don't believe everything you read.
2-Random stuff on the chart can appear to work, doesn't mean it's a good idea.
If this was "normal" trading, probably be looking for $108.5, but it's open, and realistically, if I'm doing this Pulsar thing, I only need 6 days like this over 25. Will wait til ISM now. 40 ticks plenty though.