My experience is that currency futures can be traded for 1 to 5 day swings. I have found that the average trade (over time) relative to the day to day noise to be low, so it takes a lot of patience to get the money out. I mainly trade(d) EUR, JPY and CAD and AUD vs the US For me, the size of EUR for instance is too large to scale well, I would like 50k face value, not $150k, so I don't trade them much anymore. I realize that the answer is Forex (interbank), but haven't don the research on slip.
As much as I dislike categorical statements and sweeping generalizations, I can't help but make one myself and note that "a high success rate" and "very tight stops" (anywhere near 25 pips you mentioned) are mutually exclusive, for the majors on your list. Can't be done, IMO; the intrinsic nature of those markets won't allow it. You can design / achieve one or the other, but not both in combination. Especially with a trend following system, as you specified. If you are looking for a profitable forex strategy, searching for that combination would not be a good use of your time. However, as an aside, such combo may be feasible for some highly correlated, intra-European crosses... not to suggest that the expected profit would be greater. Incidentally, why are you limiting your search to trend following systems only? As to the Tunnel Method, are you referring to the 1-hour or 4-hour version? If the 1-hour version, I agree... It takes a vivid imagination to design a trading system around the crossing of that narrow space, formed by the 144 EMA and 169 EMA on the 1H chart! Most rational observers would find that there's hardly any difference between the two EMAs, compared to volatility of the majors on the hourly and even sub-1H time frames. The 1-hour Tunnel Method should really be called "the Single 156.5 EMA Crossover Method" (as opposed to the better-known dual or triple MA crossover), with all the usual implications of a single MA crossover, whipsaws included. If you are referring to the 4-hour Tunnel Method or the Vegas Wealth Builder (daily), I personally don't trade them, but have looked into them in detail. I believe they both have a lot of merit, if there's a fit with your time frame and trading personality. Definitely worth a close look. However, they also require considerable discretion, experience, patience and TA skill in 1) applying filters, to keep you out of trades, and 2) stop placement.
how about data for back testing? i currently use futures during US + the european session. i am a little scared about testing something on minute data during the remaining period. and what do ou guys think about the competitive disadvantage to EBS data when it comes to, say 5min and longer, data?
btw what sharpe ratios do you guys require? mine is about 2 for intraday a a min. i ask that because i find that more meaningful than hit ratios and stop distance.
Vegas is just this trader I found on the internet, his name is Wei Weu, he's a PhD NYU graduate who worked on Wall Street doing quantitative research for years before he came up with the tunnel method and he now trades a few currency pairs with a couple other guys around the clock using the method. His site where he explains the method is right here: http://weiwu.us/Blog/?p=115 I encourage others to read it and let me know what you think, I'm curious to hear others opinions, this guy is apparently pretty popular and successful and he posts everything about his method on the the internet for free, but I'm finding holes in it all over the place in my backtesting, it looks like the most frustrating thing I've ever seen with all the whipsawing although his ema's do pick up every major trend and its retracements, I guess by definition they do that though. At first I thought it was hogwash but the method definitely does have some validity, you can test it on anything and it would probably work to a certain degree.
I suppose it would show a positive result but the swings would be ridiculous, which is something I'm trying to avoid. I mean the trade fails 3 out of 4 times basically if you're using 1 hour bars and I don't want to increase my time frame.
I couldn't agree more. What's strange is that Vegas makes the claim that when 144 day=169 day=12 day, those are the most profitable and most accurate setups. Too bad when it does happen you're sitting there puking out for a 10 tick loss in every direction over and over on the same day. I'VE TRIED SOME FILTERS, BUT I CAN'T FIND ANYTHING MATHEMATICAL THAT DECENTLY WEEDS OUT ANY OF THE LOSING TRADES. Anyone know of anything??? Only if you've used this method before....
Any filters in mind? I just hate to overoptimize, but I want to use something statistically valid instead of just using my own discrection based on my experience and everything. I know that has some value but I'd like to be as mechanical about it as possible. The 4 hour time frame just doesn't fit my personality that well, and if you look at the number of whipsaws proportionally, I think the 1 hour is actually better.