Still Here & Still Trading After a Wild Year

It's been a while, hasn't it? Sorry about that. Most of the silence has been simply because there wasn't a lot to report, since I stopped trading completely for a stretch of several months. During the financial chaos that was the end of 2008, a lot of my trading signals turned out to be worthless, and those promising results back in September 2008 turned into an ugly string of losses. Finally I just had to pull the plug on trading and sit on the sidelines while I reassessed my whole system. When the market hits the fan, sometimes you just need to watch from a safe place and preserve your capital.

I really should've posted about it here, I know, but at the time I wasn't really in the mood. And I also didn't have much time either, since I simultaneously had to redouble my efforts at my real, paying work [Internet stuff] to ensure I stayed financially afloat...which I have, fortunately. If nothing else, a bad streak in forex can remind you just how important your day job is!

But I also came through the storm that ended 2008 with some valuable data to run through my trading system. Maybe the most valuable data ever. It's not often you're able to forward-test your trading signals through one of the worst economic meltdowns in history. If you have a system that made good trade calls in those tumultuous conditions, and was making good trade calls before those conditions set in...well, you have a great trading system. So if you get nothing else from this post, get this - the forex market data from the past year or so is incredibly valuable.

What the data highlighted for me was that a dangerously large number of my signals made foolish assumptions about market conditions, and fell apart completely when those conditions changed. Another way of putting it is that these signals tracked correlations and assumed they were causations - but they weren't. Here's the most dramatic example: I had a signal that assumed that when Bollinger Bands were X pips apart, the EUR/USD pair would always go up. This looked logical because for a long time the EUR/USD trend was upward, and at the times it was going up, the BB bands were indeed more than X pips apart. Well, it turned out the opposite was true to...in a downward EUR/USD trend, the BB bands were also at least X pips apart...and now my trading signal was making all the wrong calls.

So my focus for the past month or so has been using the painful yet valuable results of this forward testing to identify and eliminate these false assumptions from my trading system. And just a few weeks ago I was confident enough in the new, improved system to take it back into the market with real money. So far results are good...it's made back about 400 pips. But whether that success will last is anyone's guess. Part of me feels like I have a winning set of trading signals. But another part of me feels like I'll be right back at the drawing board in a few months. Only the market has the answer to that conundrum. And even then, the answer may have an expiration date: "Signals Only Good Until __" After all, it's happened before. Hmm...maybe I need to start thinking of my trading system as a carton of milk.

Anyway, it's good to be back. I'll try to post more often now that I'm back in the market. Hope your trading's going well...and if it's not, remember the market is telling you something important, even if you don't want to hear it.

Your Trading System Needs Internal Controls

I've said this before, but it's worth repeating - especially after reading about the rise and fall of Soul Trader's Grail forex system, which I always regarded as one of the most advanced automated systems around. Reading his explanation of why this system finally had to be shut down really brought home to me the need for rigorous internal controls on your trading system.

By internal controls I mean the type of tracking algorithms that watch over your trading performance and intervene to shut it down when performance begins to dip. Or to paraphrase Soul Trader, "How to identify when the market conditions cause the system to fail, and what to do when it's failed."

String of losses over X% of trading capital? Click: the system goes off until performance improves. Recent trading odds dip below Y%? Sorry, you're out of the market until things perk up - and if they don't perk up, you'll just have to be patient and appreciate the fact that at least your funds are safe. Preserving your capital in adverse conditions is pretty boring, but it's one of the most important single factors in trading success.

The only reason I'm still trading, and have even bothered to return to trading after a recent hiatus, is because of these built-in trading restrictions. I think of them as a firewall between me and the worst the market can dish out. They're not perfect, and they certainly don't prevent all drawdowns all the time, but they're a lot better than nothing at all. In fact, given the choice I'd much rather trade with no stop-losses and strong internal controls than vice versa.

Disclaimer: check back in a few weeks and see if I've abandoned my trading controls because they failed completely in unexpected market conditions.

Back in the Market After Some Forward Testing, a Bad Carry Trade, and a Bit of Boredom

Hello again, and apologies for this blog's recent suspended animation. For various reasons hinted at above I decided to give forex a rest for a while - but I'm now back and have just made my first foray into live trading the EUR/USD again.

The principal reasons I took a break were:

I was frankly a little bored and disappointed with the progress of my EUR/USD trading system. Out of impatience with its performance, I'd convinced myself it didn't actually work with real trades, it just looked good in backtesting. In fact, I ended up questioning the entire idea of backtesting since it offers absolutely no guarantee your trading rules will work into the future. So I decided to give it a break, and maybe come back in a few months, plug in the new data, and see if my system would've actually made decent trades in that time.

Around the same time, largely because I was disillusioned with my own trading system, I convinced myself carry trading was by far the best way to trade. So I devised what I thought was a very clever hedging system using the USD/CAD to balance the volatility of the carry trader's beloved GBP/JPY pair. But unfortunately I decided this just before the GBP/JPY carry trade experienced a giant melt-down that demonstrated how little I know about carry trading or hedging. So I got burned. Thanks to my stop-losses I didn't get wiped out, but I did get scorched pretty badly. (Did I mention I also broke my own trading rules by taking on way too much risk? Well, I did. And it was dumb.) In case you're wondering what a carry trade meltdown looks like, here's a picture. Congratulations to all the GPB/JPY shorts out there, it must've been a fun couple of months:


But there is some good news after this tale of burnout and reckless trading. As I mentioned, I've been letting my trading system sit idle for the past few months while new EUR/USD price data piled up. This new data was the raw material I needed to effectively forward test my system and see if it actually produced real, profitable trades, not just pretty pictures of historical backtests. When I plugged in the new data earlier this week, I was very pleasantly surprised: my system had racked up a very steady, consistent, profitable trading record while I was ignoring it. Which is why I'm back trading (and posting) again.

One of the trading risks I've described before is the desire to fiddle compulsively with your system, even when it's doing just fine as it is - the "If it ain't broke, don't fix it" problem. Since I'm a bit of a compulsive fiddler, my challenge now is to find something else I can fiddle with to keep me from breaking what isn't broke. One thing I've been wanting to learn for a long time is how to program Metatrader to trade my system automatically using their MQL programming language. I suspect it'll be quite a challenge transitioning all my trading rules from Excel to Metatrader - but if I can get them automated it'll save a huge amount of time over the long term. So if there are any experienced Metatrader programmers out there, any advice on getting started with MQL would be much appreciated!

Current Trading Strategy: Slow and Steady

There's a lot less daily drama and a lot more sitting on the sidelines in my forex trading these days, and so far it's a strategy that seems to be working. Because of the filters and meta-signals I've added to my trading system, it generates a lot fewer signals than it used to, since it's switched off during periods when it's likely to underperform. So in an average week I'll now make 2 trades or so, a lot fewer than the daily trades I used to make (and that ended up leading to some really nasty drawdowns late last year). I'm looking for smaller profits with each trade too, with my limit orders set at levels targeted for both trending and ranging price behavior.

This slow and steady strategy looks for a few high-probability trades and in the process filters out a lot of potential winners but also lots of losers, and seeks out modest profits across many different market conditions, with a focus on playing the likely daily range rather than hoping exclusively for profitable trends (which can be few and far between). And with the signal flipper I've added, there's a built-in ability to recognize when the core indicators aren't working and basically start fading my own signals.

Overall it's a slower, more patient, more boring and thus far, more successful approach to trading. So far this month I've placed three trades using this system and they were all winners, bringing in 31 pips in total profit. Again, not huge profits, but I'll take them over a drawdown anyday. The system also got the two trades before those right as well, making this a 5 trade winning streak. I'm under no illusion it'll last much longer, but I'm sure enjoying it while it does.

Hoping for Whipsaws

Strange as it sounds, my current trading system has me looking forward to days with extreme price swings, or whipsaws, which I used to anticipate with dread. The key to this change in perspective was shifting my strategy from primarily trend-based trading to a system that looks for profits in ranging behavior as well. The result is that a whipsaw no longer looks like one of the nastiest patterns on the chart, but instead is a potential opportunity to grab some easy pips as the market swings within a semi-predictable range.

One essential part of my whipsaw-friendly strategy is a take-profit target of just 33% of the previous day's trading range in whatever direction I place the trade. Using a fairly modest limit order like this dramatically increases the chances that a wildly ranging day will also be a profitable one. By combining this take-profit strategy with a significantly wider stop-loss, my trades also allow for swings in the wrong direction that reverse and come back around to hit the take-profit a lot more often than they're taken out by the stop-loss.

Whether this strategy will work over the long-term is still an unanswered question (just like it is for any trading system) but results in both backtesting and live trading so far look good. I'll provide a periodic update on how the system's working (or not working) after it's logged some more live trades. Now bring on the whipsaws!

Home, Home on the Range

If you've watched the forex markets for any length of time you've probably noticed that they spend an awful lot of time in ranging activity, ping-ponging back and forth between support and resistance levels without breaking above or below them. So if you've come to forex expecting high drama every day, it's a good idea to adjust your expectations and get as comfortable with range-based trading as you are with the trend-following strategies that are often assumed to be the most profitable. Because if you have a trend-oriented mindset and keep looking for price breakouts that don't materialize, you're going to be (a) disappointed and frustrated (b) losing more money than you make.

Advantages of knowing the likely trading range and trading within it are:

• You'll tend to place more conservative, higher-probability trades.
• You'll have a realistic sense of where to place your stop-losses and limit orders.
• You'll be able to make money from the kinds of whipsaws that could otherwise prove very costly.
• You won't be as bored by ranging price behavior because you'll accept it and even expect it as a normal state of the markets.

Here's a chart of some ranging activity from the week of March 21-26 that could've proved challenging to the trend-following trader and profitable for the ranging strategist. Looks kind of like a giant, week-long whipsaw, doesn't it? Depending on your trading mindset, this could be a very pretty or a very nerve-wracking picture:

Of course, forex markets do move in trends as well as ranges, and being able to switch to a trend-based strategy quickly will be a key factor in developing a consistently profitable trading system. Let's say you currently have a range-oriented trade in progress, and all indicators are suddenly pointing to an emerging trend. What do you do? First off, if you're on the wrong side of the trade, don't panic - the best strategy may be the simplest, just letting your current trade end however it's going to end based on your original parameters, and then place your next trade based on the new market conditions.

If you're lucky enough to be on the right side of the trade and you've got a limit order placed at a point of likely resistance (and it hasn't been hit yet) move it outward in regular increments in the direction of the trend, thereby increasing your potential profits. You'll also want to move your stop-loss up, maybe even setting it at the break-even point so at the very least you won't lose money. Also consider switching to a trailing stop-loss that will follow the new trend up as far as it goes, locking in profits all the way.

And after that? Be ready to shift back to your range-based trading strategy. Following a dramatic market trend you'll often find yourself back home on the range - and that's not necessarily a bad thing.

Put Your Trading System on a Diet

Losing weight is one of the most common New Year's resolutions, though probably also one of the first to fall by wayside as the mealtime routines and caloric temptations of everyday life reassert themselves. A more achievable resolution may be to slim down your currency trading system - after all, there's a lot less exercise involved, and you're likely to see benefits right away.

What do I mean by putting your system on a diet, which is probably one of the weirder trading analogies I've come up with so far? Well, just as some people overeat in reaction to stress, it's very easy to overburden your trading system with bells and whistles in an effort to feel safer, more confident,and generally better insulated from the hazards of the market. The result can be a tangle of too much information and too many redundant moving parts that can bog down your trading, leave you confused, and even lead you to trade when you shouldn't.

To give an example from my own trading system, I just finished cutting some excess weight out of my EUR/USD short signals. My system now generates 16% fewer short signals than it used to, while at the same time showing gains in trading odds (up to 69% from 66%) and overall profitability in backtesting. I did this by adding a new trade filter based on internal feedback, which basically is acting as a kill-switch on those 16% of unprofitable short trades. Assuming this improved performance continues into the future, I'll make more per short trade, spend less on spread costs to place bad trades, and worry less about trades that go wrong. All good things.

Here are some ways to approach slimming down your trading system:

• How many chart indicators are you using? Do you need multiple chart windows to display them all? If so you probably have at least a couple redundant indicators that are taking up space on your screen and in your head that could be put to better use. Can you clearly state what each indicator adds to your trading performance? You should be able to. If you can't come up with a compelling (= profitable) reason, cut that indicator from your charts. See if you can consolidate all your trading in a single currency into a single chart.
  
  
• Keep in mind some people don't place trades based on charts at all. Me, for instance.
  
  
• How many currencies are you trading? Do you have a good reason for trading so many? If not you could probably cut back to fewer currencies, save on spread costs, and reduce the risk of overtrading.
  
  
• I guarantee that you are currently placing too many unprofitable trades. Don't ask how I know, I just do ;-) Visualize a filter or switch that could help you avoid these bad trades. How would it work, what would it look like? Now apply that concept to your trading data to see if it works. Repeat as needed.
  
  
• How many hours a day are you trading? Do you know if some are more profitable than others? If you can identify the most profitable times of day and focus your analysis and trading resources on those, you could make a lot more money in a lot less time. Two hours of well-timed trading combined with two hours of intensive strategy and analysis each day could yield far better results than 8 hours of trading combined with scattered analysis when you find the time between trades.
  
  
• What periodicity do you trade in: 5 minutes, 15 minutes, 1 hour, 1 day? Did you pick that particular time period because you know it's most profitable for you, or for some other reason, like it's most exciting and appears to offer the most action? If you didn't choose it for its profitability, you must have chosen it for some other reason, and it may not be a good one. Which means there's a good chance you're placing trades you shouldn't.
  
  
• Are you getting into and out of the market in the same direction all the time? If so you could probably combine some of those trades, save on spread costs, and reduce the risks of bad trade execution costing you even more pips.
  
  
• Do you subscribe to signal services that clog your inbox, take time to read and process, send contradictory signals, and leave you more confused than before they arrived? Dump 'em.
  

You get the idea. The goal here is to identify and remove all extraneous distractions, unprofitable habits, lazy assumptions, time wasters, useless moving parts, and other assorted clutter that can add to your risks and cost you money. It could be one of the best New Year's resolutions you make...as long as you stick to it!

Related topics:
Signs You May Be Overtrading
Simplify Your Trading
Stop Wasting Pips!

The BunnyGirl Trading System

One of the best-known free trading systems available was invented by the (charmingly named) BunnyGirl, who posted her strategies in the MoneyTec forums and in so doing started one of the liveliest forex discussions ever.

The core of her trading system is a 5 period x 20 period weighted moving average (WMA) cross on a 30-minute chart. The main currency pairs traded are EUR/USD, GBP/USD, and USD/CHF, and she goes both short and long on all pairs. However, it's the filtering that goes into identifying each trade that really makes the system successful...and according to BunnyGirl it's been extremely successful, to the tune of 90% trading success and a 45 trade winning streak at its peak performance.

So how does the filtering work? Here's what BunnyGirl says to look for:

• WMA crosses around the key trading period of 00:00 GMT (Greenwich Mean Time)
• A breakout of at least 20 pips for the EUR/USD and 25 pips for other pairs, plus spread if going long. This helps minimize those nasty whipsaws that can saw away your funds in a hurry. (Get it...whipsaw, saw away...I just made that up. Oh, the wit. Please, save your applause for the end of the post.)
• Avoid trades when the 100 period WMA is near the cross and the price is heading in that direction.
  
• Don't trade in the last 5 minutes of a 30 minute chart period - wait for a breakout in the next 30 minute period.
• Don't enter trades near points of strong price resistance or if they're going against the prevailing daily trend.
• Multiple lots: BunnyGirl traded a sophisticated mix of 4 lots with trailing stops set at various levels. I won't summarize this system here but you can read all about it in her complete strategy, which you can download below.

There are a lot of other rules and filters you'll want to review before trying this system, and rather than copy them all verbatim here, I'll just cut to the chase and show you where to download it. I've come across two versions of the system which differ in a few details, and one of which has a lot more helpful charts. On the other hand, the other one's half as long. So I'd download them both and compare them to see which works best for you:

• BunnyGirl Bunny Cross Trading Framework (Longer, More Graphical Version from MoneyTec.com)

• BunnyGirl's FAQ (Shorter Version from ForexBasic.com)

Related topic:
Chart-Based Trading Systems

Starting up live FX Engines GBP/USD trading

I had a really mediocre November, with very few trades going my way...and that was before I got caught flat-footed by the huge declines in the dollar around Thanksgiving. In the meantime one of my automated FX Engines was quietly cranking away earning hundreds of (demo) pips trading the British Pound. It managed to catch a lot of the action I missed while I was traveling and sleeping off turkey dinners. So I've decided to give this engine a shot at the big time (well, the pre-big time), and am allocating 20% of my trading capital for it to buy pounds with.

This particular engine looks for upward breakouts of the GBP/USD pair at the opening of US East Coast trading, and jumps on them when they happen (invariably while I'm fast asleep). Interestingly it only racks up fewer than 40% positive trades, but its risk/reward ratio works out very profitably - over 3 years of backtesting its trading performance was over 5000 pips.

The fact that this engine can help hedge against losing EUR/USD trades is also appealing - the GBP and EUR are highly correlated, so on days when my primary system gets a short EUR/USD trade wrong, this engine can help limit the damage and grab some pips going in the opposite direction. I always used to enjoy trading the GBP/USD pair and lately I've felt I'm missing out on the potential profits offered by its higher volatility by confining myself solely to the EUR/USD...so this engine could help improve that situation as well.

Of course, its success depends to a huge degree on continued downward pressure on the dollar - for the time being this seems likely to continue, but if the dollar suddenly begins to strengthen I may find myself eating these words. Yum.

Related links:
FX Engines
FX Engines Update

Forex Trading Systems

I recently added new directory categories for Carry Trading Strategies and Chart-Based Trading Systems, so there are now three categories devoted to trading systems, which I thought I'd highlight here:

Automated Trading Systems

Includes: FX Engines, MetaTrader 4, ProSignal, TradeBolt, OneDayTrades, Strategy Runner.

Carry Trading Strategies

Includes: Information about currency carry trades from Investopedia, The Economist, and Oanda forums.

Chart-Based Trading Systems

Includes: BunnyGirl Cross, RickyD System, Ichimoku Kinko Hyo, Trendfollower, Simple Candlestick System, Woodie's CCI System.

Since there are dozens if not hundreds of systems available, I'm sure I've just scratched the surface of what's out there. So any suggestions for ones I've missed would be much appreciated!

More directory topics...

Layering Your Trading Signals

One of the core features of my trading system is the layering of multiple signals into a single meta-signal. Rather than focus my attention on a few indicators, I decided to use every indicator that crossed my path that showed a positive return in backtesting. I then layered these with every other indicator on my trading menu so that if even if one wasn't firing, another one probably would be.

I'm not a programmer, so all my signals are generated via straightforward Excel functions. The syntax I use to construct my long and short meta-signals is based on the OR and AND functions, with an OR function at the highest level generating the final trading signal when any of the sub-signals have been fired. Here's what just one portion of the code looks like:

=IF(OR(AND(OR(J12=1,L12=1,H12=1,M12=1,N12=1,O12=1,Q12=1),
F12=1,AK12=0,T12=0,S12=0,AN12=0,AO12=0)...

In this long meta-signal, each number/letter combination represents a different sub-signal occupying a particular cell. A "1" represents a long sub-signal that's fired, and a "0" represents a short sub-signal that must be inactive in order for the long meta-signal to be activated. (More about that in my Trading by Process of Elimination post.)

By layering multiple signals into an automated meta-signal, you save yourself the headache of trying to decide which signal to pay attention to, sorting out contradictions between signals, and getting psychologically swamped by a dozen different indicators on your charts. It's a way to squeeze stressful decision-making and discretionary concerns out of the trading process by delegating them to a logical script. This allows you to focus more of your energies on refining that script by combining signals in different ways to achieve optimal results. And for me, that type of analysis has proved to be far more satisfying, and confidence-building, than biting my nails trying to make trading decisions without a consistent, logical process to guide me. It's not a strategy that will work for everyone...especially if you have an aversion to spreadsheets...but without it, my trading career would've ended a long time ago.

Related topics:
Trading by Process of Elimination
Designing a Trading System

Signal saturation: a good problem to have. (Or, what to do with your overlapping signals)

I don't even know if I'd call this a problem - it's more of a confirmation that you're on the right track with your trading system. What I mean by signal saturation is when you add new and promising signals to your trading system, but they don't add much to the bottom line. I had this problem today with the Ichimoku indicators discussed over at Forex Project. Intrigued by their possibilities, I wrote up a few Ichimoku equations and backtested them on some EUR/USD data, and came up with what looked like excellent results. But then I plugged my new Ichimoku signals into my trading system, ran the same backtests and found they added only a few new pips to the results - and in a couple configurations, actually subtracted pips!

What's going on here? Well, it's a problem I encounter most of the time when trying to introduce new signals to my system: the existing signals are already covering that territory, so there's no additional trading edge to be gained. Or, to put it in a more positive light, the new signals are simply confirming the validity of your existing signals, since they overlap with them almost completely. Having two sets of signals that are saying the same thing certainly suggests you're doing something right. When I come across this type of signal saturation or overlap, I will usually just put the redundant signal in storage with my other backup systems, in case market conditions change someday and it's no longer redundant.

Another strategy for dealing with overlapping signals is interpret them as reinforcing each other: if two or more signals are in play, it's an even stronger trading indicator than one signal. So it's possible that the combination indicates an even higher-probability trade than a single signal, something you'll want to test out by statistically comparing the outcomes of single signal vs. multi-signal trades.

Yet another weird quirk of overlapping signals are the areas where they don't overlap. There are sometimes significant opportunities to be found in the 10% of disagreement between signals that agree 90% of the time. For example, some of my most powerful short signals are generated when two of my highly-correlated long signals part ways. For whatever reason, at these points of disagreement, the market tends to go the other way entirely and what's usually a long signal becomes a short one. One example I mentioned above is when those Ichimoku indicators were actually subtracting pips from my long trades.

You can get caught up in all kinds of explanations for why this happens, most of which will lead you nowhere - but the important thing is it happens. (But if you really want an explanation my personal theory is it's a form of divergence.)

So, to summarize: in the worst case, signal saturation is a minor annoyance. In the best case, it's a way to generate even stronger signals, and even some unusual new signals. If it shows up in your trading system, congratulations, you're on the right track!

Related topic:
Signals aren't set in stone...so don't be afraid to fine-tune them

The Trendfollower Trading System

I keep a number of trading systems on file as backups in case the one I'm currently using goes haywire and I need to shift strategies. At some point I may also start trading with one of these backup systems concurrently with my main system to spread out my risks a bit, so I have fewer trades in one basket, so to speak. One of these trading systems I've always thought highly of was devised by Trendfollower, a trader who used to frequent the MoneyTec forums (and maybe still does, haven't checked lately).

You can access a copy of his system for free at ForexBasic.com. It's the first one listed, and you can download it as a Word doc by clicking on the little Word icon.

To briefly summarize how it works, the Trendfollower system looks for a cross of the STARC band below the Bollinger Band for short trades, and the STARC band above the Bollinger Band for long trades. STARC bands are an indicator you don't see enough of, in my opinion, and in fact they're not available on all charting platforms. I know GFT's DealBook software has them, as does Oanda's system; if they're anywhere on MetaTrader I haven't found 'em yet. STARC is an acronym for Stoller Average Range Channels, and you can read more about STARC bands in this Investopedia article.

Here's a picture of a trade specified by this system, which took place over about 2 months (!):



I won't go into all the detail of how the system works, since Trendfollower's explanation tells you everything you need to know. But here are some of the things I like about it:

It uses a multi-day or even multi-week timeframe, which means you don't have to spend hours every day staring at your charts or make trading decisions on a minute by minute basis. As Trendfollower puts it, "The nice thing about trend trading, is that you don't need to be at the computer all day long. 3 looks a day is enough."

It uses a very straightforward entry signal. There's no ambiguity as to whether it's happened or not.

It has very clearly defined exit strategies.

It uses Bollinger Bands, one of my favorite indicators, and STARC bands, which are an interesting indicator in the same genre as Bollingers, since they also form a statistical envelope around the price trend.

Trendfollower says he traded with this system successfully for 3 years, which I consider a pretty good track record.

It gives you the option of trading several different currencies without worrying too much about spread costs, since the trades are large and fairly infrequent. To quote Trendfollower again: "Because of the long term nature of my trades pip spreads are not important to me so I look for new trend breaks amongst 17 odd currency pairs." This means that your "inventory" of potential trades to choose from is quite substantial.

So if you're looking for a trading system that doesn't require constant attention, high spread costs, lots of short-term trades and all the stress that ensues, this might be the one for you. Good luck if you trade with it!

Related Links:
Trendfollower System Download
STARC Band Definition

Related Topics:
A Nice Bollinger Band Trade
Chart-Based Trading Systems
Simplify Your Trading

Limit Orders or No Limit Orders?

One of the conundrums that I periodically wrestle with is whether to use limit orders (also known as take-profit orders) to exit my trades at certain pre-determined levels. Should I take profits at 50 pips? 60 pips? 75 pips? Usually I start wondering about limit orders anytime the price moves dramatically in the right direction and then falls back close to where it started (or whipsaws), leaving me fuming that I missed out on major profits and wasted a day of trading.

Not coincidentally, this happened yesterday, which is why it's on my mind. I had a EUR/USD short trade in place and when I checked in on it first thing in the morning it showed 30 or so pips in profit. On paper, of course. Then over the next several hours the EUR/USD price drifted back up and I actually ended the day with a small loss. In hindsight a limit order (set at the right level of course - in hindsight they're always at the right level) would've grabbed those profits. So clearly I should be using them, right?

The fact is, though, I only use limit orders in very rare circumstances. If I'm going to be traveling and won't be able to closely monitor my trade, I'll sometimes set a limit order at a very conservative level to exit the trade with some profit without my involvement. But these are uncommon, one-off circumstances. In regular day-to-day trading, I don't use limit orders because in all of my backtesting, the data indicates I actually make more by leaving trades open-ended, allowing the price to rise or fall as much as it wants without setting an artificial limit on where I'll exit. Of course, in situations like yesterday's, I end up seriously doubting this conclusion, and suspect my tests were flawed and that limit orders are the answer. And yet the data continues to support a strategy of (1) not taking profits at pre-set levels and (2) only exiting trades when a shift from short to long or vice versa is indicated. An additional benefit is that this saves on spread costs, since I don't enter as many trades.

One strategy I haven't tried, because it would be a huge pain to put together, is contextual limit orders fine-tuned to a particular signal. So if signal A goes off, a take-profit at 75 pips is indicated, whereas if signal B is triggered, a limit order of 50 pips is deployed, and so on. Since I currently have a few dozen signals at work, this strategy would be incredibly time-consuming, with no guarantee I wouldn't end up with the same conclusion: no limit orders.

On the other hand, I have found there is some additional profit to be made by using stop-loss orders. However, even this is not that significant an amount, and the levels where I set my stop-losses are only reached very rarely. So on most days, I might as well not have a stop-loss in place. That said, I absolutely need them in place, for my peace of mind if nothing else.

But to return to limit/take profit orders: if you're spending a lot of time agonizing about the best place to set them, consider the possibility that you might not need them at all.

Related topics:
Simply Your Trading
Figuring out where to set your stop-losses

Simplify Your Trading

I've found I trade best when I reduce the process to the bare essentials, with as few distractions and decisions as possible. The more moving parts a trade has, the more things there are to go wrong, upset you, make you doubt your trading strategy, and cause you start making changes at the worst possible moment. The more you can reduce the logistical overhead of any given trade, the less likely you are to become emotionally involved in it, and the more time you'll have to focus on the big picture and plan your next trade.

Here are a few strategies that might help cut down on the amount of time and energy you need to devote to a particular trade. (For those who've followed this blog for any length of time, apologies if I'm sounding like a broken record :-)

Are you using a dozen different indicators and signals to initiate trades? Try combining them into a single signal that gives you a simple Yes/No answer about entering a trade. I do this by using MS Excel to calculate signals such as Bollinger Bands and moving averages and then layering them using the =IF(OR...) function, which tells me if any of my signals have fired. An additional advantage of using Excel is that it reduces the amount of time I need to look at charts, which leads me to my next simplifying strategy...

Don't look at charts so much, especially ones with a dozen different indicators jumping up and down. Charts are great for many things, but as I've noted before, it can become very difficult to tear yourself away from them, and if you're not careful you can start spotting patterns in them that aren't there. The fact is, you don't need charts to trade. In many ways an Excel price table will serve you as well or better, if you know how to use it. I still use charts but primarily to identify interesting patterns that I then plug into Excel to test historically. I haven't actually entered a trade based on a chart in almost a year. (Yes, I know, I've discussed chart-based trades using Bollinger Bands, but these were actually initiated out of an Excel formula rather than looking directly at the chart. OK, so I guess charts are good for illustrating points on your blog as well!)

Use a consistent exit strategy that requires as little discretionary input from you as possible. For example, you can combine your exit signals in the same way you combine your entry signals, giving you a simple, unambiguous Yes/No as to when to exit a trade. Or, always set the same fixed limit order to take profit at the same level, and stick with it. Or, consider exiting automatically at certain times of day; this is my exit strategy, and every day at 5:00 PM Pacific Time I either exit my current trade, or roll it over to the next day. Which leads me to...

Combine your trades whenever possible. If you find you keep jumping in and out of the market with trades in the same direction, you'll save a lot on spread costs, and avoid the risks of slippage, bad timing, and poor execution by just trading once.

Use the Fire-and-Forget Principle. Focus on pre-determining and automating all the variables in your trades (exits, stop losses, etc.) so that once you've pulled the trigger, you can walk away and the trade will take care of itself with no further attention from you.

Are you trading multiple currency pairs that are tightly correlated? There's not much point, since they're all likely to move in the same direction at the same time. You might as well just pick one of them and cut down on the distraction of following multiple pairs. This excellent article at Investopedia identifies pairs that are closely correlated, either positively or negatively. For instance, the EUR/USD almost always moves in precisely the opposite direction as the USD/CHF pair. So if you're making long EUR/USD trades and short USD/CHF trades simultaneously, you might as well just choose one or the other, because you're making practically the same trade (and paying more on spread costs, too).

Trade less often. It's much easier to overtrade than undertrade, so odds are you're overtrading.

Hope these suggestions help simplify your trading and boost your profits...if you have other ideas about how to streamline the trading process, please feel free to post them in the Comments below.

Related topics:
Signs you may be overtrading
The perils of chart burnout
Stop wasting pips!
Limiting your emotional exposure to the markets

Tired of Moving Averages? Try a Moving Median

One day I got bored with testing out different moving averages, perhaps the most common indicator in use among traders. So I decided play around with a related statistical function, the median. According to Wikipedia, "In probability theory and statistics, a median is a number dividing the higher half of a sample, a population, or a probability distribution from the lower half." Interestingly, in strict statistical terms the median is actually a type of average, since the term "average" describes a variety of different methods for determining the central tendency or "middle" of a set of data (read more about this in the Wikipedia article on averages.)

What most people typically understand to be an average is in fact the "arithmetic mean," which like the median is one of several measures of the middle of a data set. Going back to Wikipedia, the arithmetic mean is "the sum of all the members of the list divided by the number of items in the list." So if you have five prices from five successive hours, add them all up, divide by five, and you've got the arithmetic mean. Pretty simple - and that's certainly part of its appeal.

But enough with definitions - let's look at results. In my historical testing of the arithmetic mean and the median on forex data, I found both measures yielded consistent trading signals. To start with, I looked at what happens when a short-term moving average crosses a longer-term moving average (a signal familiar to just about all traders). Then I tried the same type of signal with moving medians, testing what happened when a short-term median of a few days' EUR/USD prices crossed below a longer multi-week median. Again, out popped some good results, signals you'd be pleased to have in your trading tool kit.

But the most interesting signals, and those that are currently hard at work in my active forex trading, emerged when I hybridized the arithmetic mean and the median into a single signal. By hybridize, I mean (is that a pun?) that a trade was triggered when the median crossed above or below the arithmetic mean for the same period. By looking at activity of these two subtly different measures for the identical set of prices in the identical time period, I think this turned out to be a remarkably sensitive signal. And the historical results certainly supported that theory. After a lot of further tweaking and recombining of medians and arithmetic means, I ended up with a signal I felt confident going live with.

Unfortunately, I have yet to find a moving median featured in any forex charting software I've looked at, so you'll probably have to generate it yourself in Excel or some other spreadsheet and charting application. Admittedly, this could be a hassle, but just think, you'd be looking at a signal used by almost no one else - unlike the moving average (I mean, arithmetic mean), which everyone and their cat is looking at every day. In fact, my cat's gotten pretty good at interpreting moving averages, and I may turn them all over to him soon.

Mind you, I wouldn't do all my trading based just on medians, but they're certainly an interesting addition to a trader's tool kit. Good luck and happy trading if you try them!

Further reading:
Averages
Arithmetic Mean
Median

Beware of selection bias when designing your forex trading strategy. (Or, how wishful is your thinking?)

An important principle to keep in mind when putting together a forex strategy is the need to avoid bias (aka wishful thinking) from affecting your judgment about what constitutes a signal, a chart pattern, or a trend. In science this is often called the "file drawer problem", referring to the fact that researchers may consciously or unconsciously relegate data that conflicts with their experimental hypothesis to a file drawer, thereby skewing the results of their research toward the conclusion they'd like to be true.

Let's say you're evaluating a chart-based trading system and looking back through old charts to see what happened in the past when it signalled trades. Are you paying equal attention to the times it failed and yielded a loss? Are you adding up these losses consistently and factoring them into the system's overall trading results? Or are you finding subtle ways to discount these failures, and judging them on a different standard than the successes?

For example, do you tell yourself "I would've gotten out of that trade quickly anyway, since it would have been obvious it was going the wrong direction"? Or "This one doesn't really count as a signal because it only crossed over for a very short period." Or "I would've sold when it spiked up like that, before it crashed back down for a loss." And so on...there are an almost infinite number of ways you can shape the data to convince yourself it's a viable forex system. It's human nature - though certainly one of its more troublesome aspects.

You owe it to yourself and your trading account to maintain strict standards, with as little wiggle room as possible, when deciding whether a currency trading strategy is likely to work for you. Believe me, it's a lot better than finding out the hard way.

Related topics:
Common Errors When Designing a Trading Strategy
Don't Let Your Charts Deceive You

A Snapshot of My Forex Trading System's Results

A while back I posted these charts showing the backtested performance of my forex trading system. I've made a number of changes to the EUR/USD system since then, so I thought I'd post an update on its results. Here's a chart of the new and improved system's results over 46 months of testing on historical data going back to September, 2002. Keep in mind only the last 10% or so are the results of real trades; the rest are theoretical trades stretching back long before I'd even heard of forex.



The vertical axis shows number of pips: 1.5 = 15,000 pips, 2 = 20,000 pips, etc. (Click the image for a clearer view.)

During this time the system would have placed 418 long trades and 385 short trades. Total long profits were 10,485 pips, total short profits were 7,754 pips. Shave off about 2000 pips for the spread (probably a high estimate) and you come out with over 16,000 pips in profits, or about 350 pips a month.

Both long and short predictions had a success rate of over 60%: 66.75% for long trades, 63.64% for short trades.

Average gain for a long trade was 25 pips. Average gain for a short trade was 20 pips.

The system combines over 50 different trading variables to generate its predictions. These were narrowed down from hundreds of potential signals via testing and retesting, and then combined in the right "recipe" through still more testing. Hopefully real results over the next four years will look as good!

Related topic:
Charts of My Trading System Results

The Fire-and-Forget Principle

I often find it helpful to think about my trading strategy in terms of vivid, concise analogies or metaphors. These function like mnemonic devices, summarizing key trading concepts in an easily-remembered mental picture. One metaphor I've been finding particularly helpful lately is the "fire-and-forget" principle. While this term was first applied to military technology, many of the concepts involved apply extraordinarily well to the more peaceful world of forex (actually, on second thought it's not always that peaceful...)

Essentially, fire-and-forget means designing and deploying a weapons system that can be launched and that will stay on target without any further input. It's able to function completely autonomously, allowing the launcher to move on to other tasks without worrying about the system and what it's doing. To quote Wikipedia's definition: "The military use the term for a type of missile which does not require further guidance after launch such as illumination of the target, and can hit its target without the launcher being in line of sight of the target. This is an important property for a projectile to have, since a person or vehicle that lingers near the target to guide the missile...is vulnerable to attack and unable to carry out other tasks. Generally, information about the target is programmed into the missile just prior to launch."

What does this have to do with forex? Well, in my opinion one of the most important goals in forex trading is the ability to design and execute a trade with all the key parameters factored in ahead of time, and once the trade's been made, to step back and allow it to follow those parameters (or its "guidance system," to use another metaphor) without any further input. It may miss its target, as trades sometimes will, but instead of having to watch it minute by minute to see if it hits or misses, you'll be free to get to work on your next trade. Following this principle will also help you break the habit of excessive, obsessive monitoring and tinkering with your trades after they've been put in play, which is one of the worst things you can do for your peace of mind and your account balance.

So what are the parameters you'll be building into this fire-and-forget guidance system? A short list could include:

• Currency pair (pretty obvious)
• Entry signal or entry price
• Entry time
• Number of lots to trade
• An exit system, which could include: limit orders, a timed exit strategy, an exit signal, or any other rule-based exit (emphasis on rule-based, not gut-instinct based, fear-based, or impatience-based!)
• A stop-loss, either fixed or trailing

Equally important to have on hand are data on the historical performance of this type of trade. Without that, you're trading in the dark. Remember, the data is your friend.

Once you have your trade parameters in place, the true test of the fire-and-forget principle is whether you can actually "forget" after firing. Are you able to walk away from the trade and accept the results, good or bad, without trying to influence them once the trade is underway? It's very easy to say "Yes, of course, that's the easy part," but the fact is, it's probably the hardest part. You may find you need another set of rules about how often you're allowed to check on the trade, if at all - because it can be very tempting, especially when real money is on the line. But think of all the time you'll free up to improve your trading system once you're able to step back and to let your trades guide themselves!

Related topics:
Limiting your emotional exposure to the markets
The data is your friend

The Martingale System: Some First-Hand Experience...from Las Vegas, of Course

Since I first started trading forex I've been intrigued, impressed, and a little frightened by the concept behind the legendary Martingale betting and/or trading strategy. The Martingale approach works (theoretically) by recouping losses through the exponential increase of bets or trades after each successive loss. So let's say you place a 1 lot long trade with a 10 pip take-profit and 10-pip stop-loss. In this theoretical model, the odds of it going either way are 50/50, and we won't worry about the spread for the time being. (See, I'm already oversimplifying! A roulette table is actually a more straightforward model for this...as I'll demonstrate shortly.)

Now, let's say the trade goes the wrong way and you lose those 10 pips...in response, you place the same long trade (or the opposite short trade...if odds are 50/50, doesn't much matter) but double the size to 2 lots. Now you get it wrong again, and are out a total of 30 pips. Fearlessly, you place yet another trade, doubling the size yet again to 4 lots. Voila! This time you get it right, capturing a 40 pip gain, recouping those lost 30 pips and adding a 10 pip profit. You've just traded a very basic Martingale strategy in an oversimplified theoretical universe. And, heart pounding, you thank your personal pantheon of forex gods that you didn't get that last trade wrong, and the next one, and the next, and find yourself trading 32 lots at the very limit of your margin...and then lose again. Because that would be pretty scary. And you know that while it may seem unlikely, it's definitely possible.

I myself have never dared to apply the Martingale approach to my forex trading. It's that exponential factor that stops me cold. Starting with just 1 lot and doubling your lot size with each loss, it'll take you just 8 successive losses to be trading 128 lots. And sure, theoretically and statistically you'll eventually win out. But you may not have the money to get there and back - or as Investopedia succinctly puts it, "this is assuming the gambler has an unlimited supply of money to bet with."

All those caveats aside, I've always been curious to see the Martingale in action. And since I've been enjoying a few days in lovely, albeit blisteringly hot, Las Vegas, I couldn't resist trying it out at the roulette wheel. I started off with $200 in $10 chips at a table with a minimum bet size of $10. That's steeper than I'm usually comfortable with, but that was my only option at the time, so I took it (though I found $5 minimum tables later). Now, if I were you my first observation would be that $200 isn't enough of a stake. And I would have to agree with you. But it's all I could stomach for this experiment - so already, we see how psychological barriers start working against the Martingale system. If I'd been a truly gutsy bettor, I would've started with $1000 and the willingness to lose every cent of it. But I'm not. So, there I was with $200 in $5 chips.

I confined myself entirely to roulette bets that paid $10 for $10: even/odd and red/black. Which sounds like 50/50 odds, but if you've ever seen a roulette table you'll realize it's not: there are a couple green slots numbered with 00s that worsen the odds slightly. But over time and dozens of bets, "slightly" adds up to "a lot." Yet another reason to remind yourself that all those gigantic Vegas hotels weren't built by winning streaks.

As luck would have it, I started off with a fine winning streak, and was soon up $80. Though I knew it didn't really make a difference and was just a sort of superstition (though a harmless variety), I'd often bet on red after a run of blacks, or even after a run of odds. I also suspect I favored red and even in my bets, just because I liked them better.

After my nice run of wins, I lost a bet, and doubled my bet to $20. I lost that one too, and doubled to $40. And that time I won, taking in $40, recouping my $30 in losses and adding another $10 in profit to the pile. So far, so good! And in fact, that's the way it went for most of the rest of that Martingale trial. At one point, I'd even doubled my stake to $400. "This works, this really works!" I marvelled.

And then I started to lose. Very quickly, I lost $100, and by that time I was hungry and looking for an excuse to leave, so I decided to bail out with my remaining $100 in profits and try again later. I should also mention some incidental costs I incurred along the way, roughly correlating to the spread in forex: $10 in tips to the roulette dealers, and $5 in tips to the cocktail waitress who brought me margaritas. (If only the spread tasted so good!) Overall I was very pleased with how things were going, and enjoyed dinner immensely.

Now in theory, when I returned to the roulette table I should have picked up exactly where I left off, and doubled up after my last loss as if I'd never left the table. That would be following Martingale very strictly. But of course I didn't. What do you think I am, disciplined? Instead, I started from scratch, but with $5 bets at a table with a lower minimum. And again, I started off with a nice winning streak, and found myself up $60 with my well-chosen red/black, odd/even bets. And then things turned ugly. It's impossible to describe the sheer visceral discomfort of a really nasty losing streak, so I won't. I'll just stick to the numbers: 7 successive losses, which required me to double my bet to $320. $320! And what if I lost that bet? I'd be facing a $640 bet. And after that...well, I didn't want to think any further out than that. And, as it turned out, I didn't want to bet that $320, either. It would've required me to dig further into my wallet, throwing more bills on the table for chips, and the prospect of that made me queasy. Psychology had kicked in again, and the simple thought of emptying my wallet to make the required bet - and if that failed, heading to the ATM machine for even more - was almost intolerable. So I did what all bad gamblers and traders do when faced with painful losses: I abandoned my strategy. I just shoved all the remaining chips I had, about $100 worth, onto red and hoped it would somehow all work out. It didn't. The ball landed on black, and I got up and walked away. All my gains were wiped out and I was down to around $100 from my original $200 stake. And that's how things currently stand.

Right now I'm debating whether to head to an ATM and return to the roulette table with a new, larger Martingale fund and a truly disciplined approach that'll really make the system work this time. But you know, somehow I don't think that'll happen. I've learned enough about myself from this Martingale experience to know that when things get ugly, I'm liable to bail out and abandon the only strategy with the potential to work. Just as importantly, I know that I don't have unlimited funds, and few strategies can test the limits of your funds more quickly than the Martingale system. So I think I'll consider this Martingale experiment complete, and go blow my money on something I know I'll enjoy: the Shark Reef. I'm sure I can learn just as much about trading strategies by watching a tank full of sharks, and at a cost of a mere $15.95.

Now, turning our attention back to forex, I've neglected to mention another factor that could dramatically skew those Martingale odds in your favor. It's pretty obvious and you've probably guessed it by now: a winning trading system. If you're an experienced trader and already have a strategy that produces consistent profits, using Martingale principles is potentially a lot less risky, since your strategy boosts your odds of profitability from the get-go. After all, forex isn't like a roulette wheel for those who understand its inner workings. Let's say you trade on news events, and you get a news trade wrong because of a whipsaw in the price, or a sudden reversal, or a stop-loss in the wrong place. Even after this loss, you know that over time you get most news trades right. Since you have high confidence in your system's performance over time, maybe next time you'll decide to double your trade size. I'm not recommending it, but I will say this: if I ever venture into Martingale trading with forex, it will only be in conjunction with a tested trading strategy in which I already have a high degree of confidence.

You may also want to check out the anti-Martingale strategy, which increases bets/trades during winning streaks, instead of losing streaks. Haven't tried it, don't know as much about it, but maybe it works. If it does, please come back and post comments on your anti-Martingale experience!

By the way, if you're planning a trip to Las Vegas, I'd highly recommend the Mandalay Bay resort. Their 11-acre beach is a fine place to get some sun and forget about an expensive Martingale experiment.

And now, I'm off to visit their shark tank. Most posts to come, if I still have all my fingers.

Common Errors When Designing a Trading Strategy

When wrangling spreadsheets and charts with historical forex price data, it can be surprisingly easy to make the most basic types of errors that, if not caught quickly, can lead to all kinds of mistaken conclusions and costly trades. I've compiled a short list of the ones I've been guilty of at one time or another, which I hope will be helpful to you as you design your own trading methodology.

Error 1: Don't correlate your dates and times properly.

I once discovered a forex trading signal that seemed too good to be true, predicting market movements with truly amazing accuracy and racking up huge gains on a consistent basis. Well, it turned out it was too good to be true - I'd been using a signal derived from one day's market activity to predict that same day's activity, rather than activity the following day. By lining up my dates incorrectly, I ended up with a circular signal that essentially predicted itself. No wonder it was so phenomenally accurate!

That said, misaligning your dates can also lead to some interesting discoveries. For instance, I once did a study to discover how London Stock Exchange (LSE) activity correlated to GBP/USD trading the following day. Only I mistakenly used LSE data from two days previous, not one day. As it turned out, this led to some fairly significant predictions.

So, be open to the possibility of an auspicious accident, but make sure to double-check your dates and times to avoid the nasty accidents, which tend to be far more common.

Error 2: Use the wrong metric of success

This is another of those too-good-to-be-true stories - yet again, I found a signal that looked like it yielded amazing profits. Well, it turned out I was adding up the wrong numbers. Instead of totalling all trades generated by that signal, I'd added up a column of all the positive trades it had triggered. No wonder the totals looked so good. The metric I should have been using was one column over, and recorded all trades indicated by the signal. And it was, surprise surprise, quite a bit lower.

Another way of confusing your metrics is to focus on the percentage of positive trades rather than the total gains a signal generates. While a 70% success rate looks pretty good, it could be outperformed by a signal running at just 50%, or even lower. A signal that generates 100 pip profits half the time and 30 pip losses the other half is a lot better than one generating 30 pip profits 70% of the time and 50 pip losses the other 30%. (Sorry if I'm stating the obvious here...but it's not always obvious.)

Error 3: Use the wrong periodicity (time span) for your signals

If you're planning to place day-long trades, you'll probably have more luck using daily charts and price data than 5-minute charts. And if you're planning to do short-term scalping, daily data may not be as helpful as a minute-by-minute chart.

Error 4: Use Too Small a Set of Data

When testing out a potential signal, you'll need a substantial body of market data to determine how it performs over time. A strategy tested on 10 trades over three weeks is likely to be a lot less robust than one tested on 500 trades over three years. The more historical price data you can obtain to put your signals through their paces, the better.

Error 5: Forget to take the spread into account.

I've discussed this before, but it's always worth revisiting. You may have a system that makes you 10,000 pips trading the EUR/USD, but if it takes you 5000 trades to do so, you could very well end up making very little, or even losing money because of the spread. One factor to look at when evaluating spread costs is how many of your trades are sequential: if you tend to have blocks of 10 trades that you can roll into one trade, then you have less to worry about than someone with many sporadic trades occurring in isolation.

Obviously this is far from a complete list of all the errors you can blunder into when creating a forex system -- so stay tuned for more mistakes as I make them.

Related topics:
Don't let your charts deceive you
How I just wasted a bunch of pips for no good reason

Weighting your trading signals to improve performance

One strategy I've considered while designing forex trading signals is assigning a weight to each signal based on its overall performance in predicting market movements. Just as a weighted moving average assigns greater importance to more recent price data with the goal of improving its predictive accuracy, weighted signals would gain or lose importance versus other signals based on certain types of performance data.

In a trading system that relies on several different signals, weighting could help decide between conflicting signals and make trading predictions clearer, removing some of the confusion and uncertainty that are deadly to successful forex trading. Let's say you had a signal based on departure from a lower Bollinger Band that was shouting "Go long," and another signal based on long-term moving averages that was insisting, "Stay short"! Which one should you listen to? It's an ideal situation for signal weighting.

So, how would you go about weighting your signals in a way that's simple, easily understood, and most important, profitable? The best idea I've come up with so far, though I haven't actually put it into practice, is to determine the weight for each signal by calculating the average gain in pips for every trade it predicts. So maybe your long Bollinger Band signal historically yields an average gain of 10 pips per trade, while your short moving average signal yields 15 pips per trade. So, in theory, you could calculate that the moving average signal has 1.5 times the weight of the Bollinger signal, and therefore is the signal to follow.

When you're combining multiple signals to determine a trade direction, weighting can also help by giving you a total score based on the combined signals. Let's say you have three long signals with weights of 1, 1.5, and 3, and three short signals with weights of 1, 2, and 2. Adding up the long signals and then dividing by three for the average weight, you get a weighted score of 1.833, while your average score for the short signals is 1.667. So based on this technique, it looks like you should go long.

Another approach using weighted signals would be strictly additive - say you have 5 long signals adding up to a score of 8.5, and 3 short signals adding up to 10. While you've got fewer total short signals, they carry greater weight than the long ones, and so they win the trade.

These are just a few possible approaches to creating and interpreting your weighted signals. The key to determining the best weighting technique for your particular trading signals will be to test, test, and test some more using as much historical market data as you can get your hands on. Signals aweigh!

Stop wasting pips! Strategies for cutting your trading costs

One of the most important aspects of any forex trading strategy is containing your trading costs, which helps ensure that you're squeezing the most profit out of every trade and aren't drawing down your funds with pointless trades and careless execution. Here are some of the common ways for you to waste your money while trading, all of which I've been guilty of at one time or another; how do you think I discovered them?

Forgetting to take the spread into account: seems obvious enough, but it's easy to forget that every time you place a trade you're paying the spread right up front, and if your trades have a narrow margin of success, a big chunk of your earnings is going to be eaten up by the spread. And if you're not executing your trades perfectly, that chunk is going to be even bigger. Another surefire way to get screwed by the spread is to take a trading strategy from a pair with a low spread, like the EUR/USD, and try to use it for a more exotic pair with a much larger spread. You may quickly find yourself spread way too thin. (Excuse the pun.)

Overtrading: maybe you get a great new trading idea every hour or so. Maybe you change your mind about a trade that seemed perfect at the time but isn't going the way you'd planned. Maybe you feel like getting into 5 different currencies at the same time, for no good reason other than you like variety. All of these undisciplined trades are going to burn up pips like wildfire, and in the long run are unlikely to repay you for those added costs.

Testing an unproven strategy with real trades: whether you've invented it yourself, gotten it out of a book, or found it on a web site, any trading system that you start using in the live forex market without prior backtesting and/or a trial run on a demo system is likely to cost you plenty of pips as you learn its idiosyncrasies and weaknesses the hard way. I lost a lot of money this way - for my first foray into forex, I just grabbed a trading system off the shelf (well, the Internet) and assumed it would work because lots of people seemed to be using it and had testimonials about it all over the web site. I started trading with this system after only a few hours of practice on a demo platform, and quickly discovered that I had no idea how to make it work consistently...except when it came to losing me pips, which it was great at.

Trading exotic pairs with a big spread just because you like the idea. This goes back to point 2 about forgetting the spread. If you're drawn to exotic pairs because you like dropping references to the Thai Baht or Polish Zloty in conversation, then do yourself a favor and just trade them with a demo account. Most people you try to impress with your exotic trading prowess won't know the difference, or care.

On the other hand, if you happen to have worked for the Central Bank of either of these countries and know the currency's behavior by heart, by all means, go right ahead and put real money on the line.

Combine trades whenever possible: don't take profits on a trade if there's a high likelihood you'll just be re-entering the market again in the same direction soon. (Unless your plan is to get out in anticipation of a big dip and then buy at the bottom.) If you don't expect dramatic changes, stay in the trade - you'll cut your spread costs in half. Do it again the next time you're tempted to jump in and out of the market, and you'll cut them by two-thirds. You get the idea. (This is probably just another way of saying don't overtrade.)

Compare spreads on different trading platforms: you may be able to save a pip or more in spread costs with a forex broker offering lower spreads. Even if you're just saving one pip on each trade, it'll add up - and if you're fortunate enough to rack up significant trading gains and begin placing larger and larger trades, that one pip savings could eventually be 5 pips a trade, 10 pips, 50 pips, and more. To the best of my knowledge, the broker offering the lowest spreads at the moment is Oanda, which has a mere 1.5 pip spread on the EUR/USD pair. Oanda will also let you place tiny trades of as low as a dollar, which leads me to my next point...

If you must trade an untested system with actual money, since demo trading just doesn't feel "real" enough (and yes, I know what you mean, there's nothing like real money to focus your attention), then start out placing tiny trades of under $10 with a broker like Oanda. This will allow you to feel like you're making (or losing) real money while preserving your funds and your sanity by reducing the size of your losses. It's a bit like playing the nickel slots at a casino to make your cash last longer - though hopefully the odds will be a lot better.

Finally, at the risk of repeating myself, the best way to save pips is to thoroughly test your trading system and then execute trades with consistency and discipline. Remember, the data is your friend, so treat it like one - spend lots of time with it, get to know it better, check in on it every day, and it may just pay you back handsomely.

Related topics:
Simplify Your Trading
Signs You May Be Overtrading

Sticking to your trading system

One of the hardest things to achieve as a forex trader, at least in my experience, is a level of commitment to your trading strategy that allows you to trade consistently throughout a stretch of losing trades. Every system I'm aware of inevitably generates losing trades, and there's nothing that will sap your confidence in your trading plan than watching your funds trickle (or gush) away day after day.

Since every system needs consistent execution over time to prove itself, bailing out at the first sign of a losing streak, or switching nervously between different systems in a quest for one that's fail-safe, is a great way to wreck your forex career in a hurry. It's important that you give your system a fair chance to succeed before tossing it in the trash, reworking it into something completely different, or moving on to the next great strategy.

Now I'm not saying you should go down with the ship, tossing all your funds overboard on a system that's clearly and consistently failing. But I am recommending that you strike a balance between the importance of long-term consistency and the anxiety of short-term risk aversion. It is entirely possible your system is fatally flawed, and you just don't know it yet. It is also entirely possible that your system is a winner...and you don't know that yet, either. One way to build your confidence in your system's performance over time is through extensive backtesting. Remember, the data is your friend. Testing past trades using historical data will show you the worst drawdowns your strategy has produced, as well as its greatest winning streaks.

But let's say you've done all the above -- thoroughly backtested your system, traded consistently for several weeks, put aside those inevitable panicky reactions and stuck to your trades, and you're still faced with mounting losses. How do you know when to bail out and go looking for a better forex philosophy? One approach is to compare your current losses with the worst drawdowns in your historical testing. Are your losses of a similar magnitude? Are they even larger than those in the backtests...25% greater, 50% greater? I'd say if you're getting into a range with losses are over 50% greater, it's time to start dusting off Plan B. And it's always important to have a Plan B, and a Plan C...just as long as you're not switching between them on a daily basis out of panic or desperation. The forex markets are not known for rewarding either of those qualities; quite the opposite.

Remember, this is all just my opinion, and how useful my advice is to you will undoubtedly vary depending on your risk aversion or tolerance. But I do think it's important to keep in mind, no matter how you trade, that any trading system will (a) generate losses and (b) require consistent execution over time to prove its worth.

Disagree? Then please say so in the comments!

Charts of my Trading System Results

Here's a picture of the EUR/USD system's backtested performance over 23 months of daily trades...I feel like a proud father :-)



On the vertical axis are the total number of pips gained, with 1 equalling 10,000 pips. So this chart shows over 11,000 pips in (theoretical) profit, before the 3 pip spread is factored in. With 503 trades charted, the highest possible spread costs for all these trades would be 1509 pips. However, the real number would've been a lot less, since many of these are successive trades in the same direction that would get rolled into one trade, saving substantially on the spread. I haven't figured it out exactly but I suspect actual spread costs are around 800 pips at most. If I ever get around to switching to Oanda they'll be even less. I also haven't factored in interest, but the way the trades are divided between long and short it wouldn't have added or subtracted much either way.

For comparison and contrast, here's the chart of my GBP/USD system's backtested performance for the same period.



Note that it's significantly choppier than the EUR/USD chart, which makes sense given this pair's higher volatility. Total profit before spread costs comes out to 11,027 pips over 566 trades, with spread costs probably subtracting around 1200 pips.

While these charts may look pretty good, you have to bear this crucial caveat in mind: I haven't traded for this entire period. The vast majority of the data you see here are results from backtests, not actual predictions. In fact, I've only done one actual trade of the GBP from this system, and it didn't go well. The period that I have traded on a regular basis is at the very end of the EUR/USD chart, from about 11/17/05 onwards on the horizontal axis. Note that this hasn't exactly been a period of stunning success. For me, this raises some interesting questions about how to design and test trading strategies that are successful in the real world, not just on a spreadsheet.

For instance, is it possible that I've created a system that very effectively describes the past couple years of market activity, but doesn't actually work well at predicting it in the future? Are the signals I've identified so specific to past activity that they have no value in forecasting new trends? How valuable is a backtest anyway, when the market's dynamics are always in flux and, as one trader recently put it, "it mutates like a virus"? Are there certain types of signals that are simple and true enough that they'll always work, whereas others are transient and illusory? And do I even have enough data here to base predictions on? Would another couple years of data produce a completely different picture? Yes, it probably would - now if only I could get that data cheap (all the data used in creating this system was acquired for free, incidentally).

All good questions, and tough ones to answer...many of the answers will only be revealed through future trades, and hopefully not at an exorbitant cost. Or if you happen to know any of them right off the top of your head, please leave them in the comments!

Related topic:
A Snapshot of My Forex Trading System's Results

Trading by the Process of Elimination

When I started creating and testing long and short trading signals, I discovered a whole class of signals that hadn't occurred to me previously. They're trading indicators generated by a process of exclusion or elimination: if long signals A, B, and C are NOT present, then by default a short signal is generated. What may not be immediately obvious when looking at a forex chart is that the absence of certain indicators can be just as important as their presence.

In some cases, the absence of some signals can amplify an opposing signal -- for example, here's the syntax of one Excel formula I've created that triggers a signal only if one long indicator is present while three short indicators are absent: =IF(AND(W572=1,Z572=0,Y572=0,K572=0), where 1 is a long signal and 0 is the absence of a short signal.

I could give a bunch of other examples, but you probably get the idea. So if you ever set out to design your own trading system, remember that you when you create a long signal, its absence may also be a short signal, and vice versa. Who knows how many pips this could add to your trades!

Designing a Trading System, Tip #1: The Data is Your Friend

When I first started playing around in the forex markets I was using an off-the-shelf system handed me by the account exec at the broker I was then working with (Global Forex Trading). It was Woodie's CCI system, a well-known short-term trading system that uses the CCI signal on a 5-minute chart. While Woodie's system is very straightforward and I have no doubt it works very well in the right hands, mine were definitely not the right hands. The 5 minute chart is a jumpy, jerky, volatile animal, and staring at it for hours on end, far into the night, can make you more than a little crazy. The timeframe just wasn't healthy for a somewhat high-strung and obsessive newbie trader like me.

It was also very hard to determine how well Woodie's system would perform if executed consistently over months and years. How many trades would it get right, on average, and how would the good and bad trades stack up over time? (Assuming the trader knew what they were doing -- another problem, since I didn't.) I just didn't have that data and didn't know where to find it, with the result that I was operating in the dark, not knowing how many trades I could expect to get wrong or right, not knowing what types of gains I could expect from a good trade, not knowing how long it took to get good at this system...not knowing anything, really, except the basic signals and the timeframe.

What I needed, and what I eventually decided was the key preprequisite to any forex system, was data, as much of it as possible. Without data you're left guessing at too many different factors, and the more you're left guessing, the more likely you are to panic and do something stupid: a surefire way to zero out your funds in a few weeks.

After flailing around hopelessly with the CCI system, I set about creating a data-driven system that would operate on a timeframe that was less likely to drive me crazy (24 hour intervals rather than 5 minutes) and that could be thoroughly tested via the best statistical tool I had available (MS Excel). The key element was historical price data that I could use to generate testable signals. But where could I get it, preferable for free? Initially I downloaded a large table of daily data from Oanda's historical price tool. But this table only offered the daily average prices, and didn't have opening, closing, high, and low prices. I wanted more data.

Fortunately I stumbled on a mention of the Metatrader platform in the Oanda discussion forums, so I downloaded it and quickly discovered it offered a huge amount of free historical price data on all the major currency pairs. (If you have the software installed, look under Tools > History Center.) With this data downloaded into a spreadsheet, I was able to begin trying out different formulas and seeing how they performed over almost two years of daily market activity. (Obviously more than two years would be even better but I haven't found this data available for free...yet.) I could determine the total gains from each signal, the average gain per trade, the maximum drawdown, and all sorts of other data points that reassured me I had a system that could work. I could also begin combining signals into a more complex formula that would work under a variety of market conditions.

The result? I now have enough confidence in my system that I can trade it consistently, without panicking, and without staring obsessively at charts trying to second or third-guess or fourth-guess the next market trend. Oh yeah, and it makes money. That's good too. (But who knows, this post may have jinxed the whole thing and I'll have to start over from scratch. Like many traders, I've got a superstitious streak. And no amount of data can get rid of that ;-)

So, to summarize: the data is your friend. Find the data and you'll find the money.