What's wrong with us forex traders?

I'm feeling the urge to rant a bit about the forex community and where I see bad habits that will eventually affect people's pocket books. There is always a danger that this kind of righteous indignation will make some people angry but I'm really doing it out of concern. I want everyone here to be successful (unless you are on the opposite end of one of my trades) but I see some key habits in forex forums across the net that are scary. It's time for a little self reflection.

 

1. Reporting results in pips and not percentages - If other types of traders from the futures or stock world saw us doing this, they would choke on their pretzels. First of all, and probably most importantly, pips have different monetary values depending on the pair traded. But time and again, I see system results report Total Pips across diverse pairs. Now I understand that forex is very different from stocks in that it is traded on margin. This makes it fuzzy what exactly one should use as the basis for return. % of what?

One obvious answer might be margin. The problem there is that margin means different things depending on the account leverage used. Therefore, my suggestion is to use USD-exposure as the basis (or if your are EUR-based then EUR-exposure, etc.) What do I mean be this? Well, you know those crazy numbers we trade all day long like 1.5800 and 2.0001? Guess what? They are exchange rates! I know it's easy to forget that we are trading exchange rates because it isn't often necessary to think of them as such when trading but doing so is key to understanding your returns. When you buy a lot (100000) of EURUSD, how many USD worth of EUR are you buying? Well, simply multiply the units by the rate of the currency on the left in terms of the currency you want. In this case 100000 * 1.58 if the EURUSD quote is 1.58. This gives you 158000 USD. Now if you make $1000 on the trade, you've made ~0.63%. Note that this return is independent of leverage and I think it should be. Leverage is a personal choice of the trader. System results shouldn't be published with leverage in them. The result should tell traders that for every dollar exposure they take on, they would get what return from it.

1. I've mentioned leverage several times but I want to mention now that most forex traders I read about both misuse the term and use way too much of it when you put it in perspective. First of all, the brokers offer various account leverages from 1:1 to 400:1 in some cases. But account leverage has little to do with real leverage. Real leverage goes right back to the issue of exposure that I mentioned above. If you have positions equivalent to $1,000,000 and your account is only $100,000, then you are trading 10:1 leverage. I don't care if your account says 400:1 and you are paying only $2500 to open the position. Leverage is the ratio of your investments to your account balance. The margin (often called "leverage") is just an amount the broker wants you to have on hand and has nothing to do with your position. If you change your margin level, your position stays the same and you just have to pony up more or less money to keep it open.

As for using too much, that is of course open to debate. But let me put this into perspective. Those hedge funds you always hear about blowing up right and left? They typically use leverage in the 5:1 to 30:1 range. In other words, too much. I would bet that most forex traders would look at leverage like that and laugh at how "small" it is and how rich you could be if you just jacked it up on their great system. But let me tell you that survivors in the world of trading would hesitive to go past 3:1 or 4:1. It's just too dangerous. At 5:1, a 20% drawdown would wipe you out. Think 20% is never going to happen to you? Well there is a rule of thumb in trading that your worst drawdown is likely to be as big as your typical expected annual return. I would venture to guess that most forex traders wouldn't be bothering if they didn't expect to make 20% annually ergo.... I would seriously consider keeping your positions at 2:1 to 3:1 if you can and 1:1 would be totally great if your system can still perform well at that level. And, yes, you can use less that 1:1 leverage. It's call keeping cash reserves by trading positions that are smaller than your account size. In other words, if your system needs 100:1 to make you a living, either your account is too small or your system too foolhardy.

1. So you are using good leverage or none at all and you record your trades in terms of money and not pips. But 20% returns? Come on, I want 100%! 200%! Let me pass on some sobering news. I think that the limit of trading systems that are truly consistent over years using reasonable leverage (see above) is 20-30%. Where do I get this? Well, years of system testing first of all but also the buzz in the industry that even the Soros' and Buffets' pull about 20-30% over all their years of trading on average. Yes, I know of the guy who made 17,000% on that one option trade or the real estate mogul who flipped a house for 300%. But I'm talking about long-run expectation here from the best and brightest. Learn to shoot for 20-30%. You may have your 100% years but if you can find me one person, just one, who has audited records of greater than 30% per year for 10 years, I will never post another opinion about trading again. I'm personally involved in the hedge fund industry and believe me, a 20% annual return is seen as absolutely fantastic. You will have money thrown at you and these are the financial big boys. So do you really think that spending a few weeks with MetaTrader is going to get you a system that makes 100% without being very risky to your account? It might in 6 months of backtesting but don't expect it to last. I know this is not what the guy with $5000 who wants to quit his job next month wants to hear but that's the truth as I see it. It takes money to make money I'm sorry to say.
2. Ah, backtesting. Can't live with it and can't live without it. A system will never be proven by backtesting. I think most people know that. But I'm always surprised that people are willing to put real money on the line with a system that was backtested on 1 year of data. But isn't 1 year a long time? Not in trading it isn't. I've seen systems that looked like the holy grail over 4 years and flopped in the fifth. Your system needs years and years of backtesting. 5 at least and 10 if you can get it. 20 would be fantastic. Even then I would discount my results when starting with real money. Markets change but they change slowly. If you test one year, I can bet you are only capturing one type of market regime and it will change on you the minute you start trading. Also, no one ever seems to standardize their results. They always report actual returns but never detrend them. If the whole period was a bull and your system goes long alot that's not a winner. You must detrend the results by dividing by the average return per trade of a random system (i.e. dart board). There are mathmatical techniques that help with this and I can share them with people interested and there are books on the topic. I bet if most systems were put through the hoops they would find that the results aren't statistically significant but simply a results of market luck.

I'm not saying these thing because I think I'm better than anybody. I struggle with the same things every trader does. But I do think my survival will depend on keeping these things in mind. They are more important than any system. Systems come and go. I don't want to do the same.

A great no-nonsense post.

Well maybe we shouldn't be so concerned about what the next person is doing, and just focus on ourselves. This rant has been brought up over and over again, to no avail. It seems people don't care. It's really about how many pips you can gain. whether the pip had a nickel value, or a 30,000 dollar value is really no one else's business! If I say I netted 40 pips off a trade, you can fill in the value for yourself. All you need to see is the entry point, and exit point, and on the real, you could figure out the pip gain or loss using that info. Anything more than that is just plain NOSY!

That's why I suggested %. It still hides money value. And yes pips are useful if you are talking about one pair or pairs all denominated in the same quote currency like EURUSD and GBPUSD, but I see systems that trade everything from xxxUSD to xxxJPY and then lump the pips.

I understand the the frustration with bringing up something that involves other people but what is the forum for if not sharing ideas and helping each other become better traders? We might as well all go away and keep to ourselves.

Re: Reporting results in pips and not percentages

Reporting pips for a specific pair will allow readers to see potential exits, wont it? These pages are after all about learning !!

Railing against how posters write their experiences is not ( in my book ) the best way to go here. If we dont like the way some folk write their work then dont read them, there is an "ignore" function available to us all. It is this sort of criticism that makes me hit that ignore button.

I am here to learn and to help ( where I can ), not to criticise or to point the percieved errors of other people.

I am not going to read this thread any further ...

fairwind:

LOL.

I think your interpretation of my words is quite unfair. There is a big difference between ridiculing others and pointing out that pips isn't the most accurate way to report results in mixed pair situations. It has the potential to be misleading, even unintentionally. I never said they weren't worth reading. I'm not trying to turn people away from reading others' posts. I offered a constructive suggestion of a different more accurate approach. Take it or leave it but I think pips suck as a performance metric. That's all I meant. The whole world outside of forex forums reports results in % return and we use pips. Why? Probably because we are a bit lazy. It's not easy to convert pips to %. I understand the resistance. But that's no reason to not encourage change. I'm sorry you were offended.

Your first sentence shows that you really didn't get it. I'm not saying you don't use the prices for entry and exit. Obviously you do that. I'm saying that it isn't helpful to report your system results when they involve mixed pairs as +/-X pips. It's apples and oranges if the pairs have different pip values.

I too, am guilty of reporting in pips with little or no regard for % 'earned' (as opposed to luck) Recently I've begun to realize that a truly successful system focuses on % just like my savings account and 401K. I've made almost 1000 pips since Feb, but my account is up only 15%. 5% per month sucks when you account for the taxes owed, overhead and the fact that my 401K does that well with very little risk involved. I realize now that my next step should be to up the % of equity used on high percentage moves. (easier said than done) but necessary neverless.

Reality is often a bitter pill to swallow.

Thanks for making me think

agreed the number of pips made shows what caliber the trader is.
not the amount because one couldve made 100k in 10 pips risking his whole account and saying he made 100% return for example,
but we wouldnt take him seriously because he made 10 pips with a stupid risk

thats y we need to assess a trader by the number of pips made and according to the timing of his entry.

great thread

Nice post, but I think you contradict yourself in the back testing part. As you say, a system may work, then stop as the markets change. By back testing we are interested in the recent past on the basis that no market change has occurred recently, trying to back test a system over the last 20 years is as pointless as doing over the last 2 days, a balance needs to be found which is dependent on the type of trading you are looking to do.

But all the stuff on de-trending data is good, that's the first time I've seen it mentioned in these parts.

You make a good point but I what I mean to say is not that a new system necessarily needs to be created every few years but that a system needs to be proven over many market types. Too often a system is backtested on one type of market, like a trending one, and the system designer thinks they have a real winner. But in that type of market it's easy to win with just random trade selection. That's why, as you've noticed, detrending is useful.

Also, by backtesting over 20 years vs. 2 you learn much more about the inherent weaknesses or strengths of a system. Even if markets are different now, you have more data that can be compared to current market conditions and interpreted accordingly.

Ok, I think I need to clarify because this is the second post implying that I meant we should switch to amounts of money vs. pips. I certainly don't. I'm actually suggesting something that very much doesn't depend on the risk or leverage that the trader uses. In fact, as I hope to demonstrate, pips actually disguise risk in much the same way you describe in some cases.

For example:
Let's say the EUR/USD is trading at 1.5631 and a trader decides to buy. The trader then closes the trade at 1.5661 for a 30 pip profit. This is how we typically see things described. And this is perfectly clear because we are talking about one pair.

Now let's say the trader also trades the USD/JPY on the same day short at 104.40 and closes at 104.00. A great day for our trader who grabbed 40 pips on this trade.

Total for the day: 70 pips!

But hold on. The pips for the EUR/USD and USD/JPY have different values in USD. The EUR/USD pip is worth $1 for every 10000 EUR traded. But the USD/JPY is trickier. The pip value is variable and is worth about $0.96 per 10000 USD traded.

That's why I suggest that the results be based the return achieved by $1 unleveraged dollar of exposure to the pair.

For the USD/JPY this is simple 1 unit is $1 unleveraged and the return on that $1 is (104.40 - 104.00) * (1 / 104.40) * (1 / 1) = $0.0038 or 0.38%

For the EUR/USD the number of EUR $1 will by is the inverse of the entry price or 1 / 1.5631 = ~0.64 so our return is (1.5661 - 1.5631) * (1 / 1) * (1 / 1.5631) = 0.0019 or 0.19%

Want a short cut? Just take the ratio of the entry price and exit price:
104.40 / 104 = 1.0038
1.5661 / 1.5631 = 1.0019

Now this tells you unambiguously what $1 of exposure returned on the trade. This leaves it to the investor to determine what exposure they are taking. If it's $100,000 then the return will be 0.19% of $100,000 for example. And if their account is only $10,000 the that will be 1.9% on their account. But we don't and shouldn't care about the account size when reporting results. We want to know how the trade performed and keep risk appetite out of it.

By thinking of trades as a return on exposure (which can be scaled to the traders liking) we end up with a comparitive measure that avoids the issues of differing pip values and is independent of leverage used or account size.

The important thing to remember is that when trading fix lots of 10K or 100K that you aren't trading the same size of your base currency in all cases. A 100K lot of EURUSD us actually only about 64K USD worth of trade where a 100K USDJPY is 100K USD. The 2 trades sound like the same size but the JPY trade is really 1.56 times as large as the EUR trade! How often though are these pairs just thrown together and talk about like they are the same in terms of risk exposure. This is the key concept that really hit it home I hope.

I too hate that people report systems or trades in pips. It shows a fundamental disregard of basic risk management ideas.

I think the largest strength of Forex is the ability to vary your position size and have your commissions increase or decrease proportionally. I realize that the commission tax is fairly large compared to a properly capitalized futures, but I don't have $100k USD to go trade the SnP.

Everyone should report gains in terms of R instead of pips or percent.

If my SL was 100 pips and I booked 50 positive pips. I earned .5 R .

Or if my SL was 10 and I booked 50 pips, I earned 5 R.

It standardizes different traders because I only like to risk .5% to 1% per trade whereas some people are much more aggressive risking 2 to 3% per trade. If I know my R is .5 % and someone shows me a system which averages 10R a month with 4R drawdowns... I know I stand to make 5% on equity with 2% drawdowns.

Reporting in Pips alone does not show who's a great trade and who's not. Ya'll need to go read some Tharp about percent volatility position sizing.

I can grab 100's of pips days out... hell anyone can do it, if their stops are large enough... That's the problem... you might have the best entry in the world, but if your stops are so large that you either 1) don't stand to profit very much since you have buy or sell with smaller lots... or 2) you're risking an insane amount of money on a single trade idea with large lots.

Go google "risk of ruin"

Reporting solely in Pips is useless unless each trade had a fixed SL amount...

I agree with reporting results in %. There are many systems on forums that can even falsely give the impression that they are profit in terms of pips but actually they are either just breakeven or losing. Sometimes when stop loss is hit for certain pips, loss in % is big because many lots were closed but when it hits TP, gain is small coz fewer lots hit TP. Also the position size may vary from trade to trade depending on the set up or quality of signal. So 40 pips profit may be less than or more than 40 pips loss.

Edit: I just realized that the point I was making was similar to above post by longhornxtreme.

Pips are fine by me. I've been conversing in terms of pips for years - and I fail to see what the big deal is.

Yeah, we know that pips have different values from pair to pair and their value is magnified by the number of lots traded. So what? Pips are a universal language.

At the end of a day, week or month - I'll calculate percentages. And that seems plenty often enough for me.

In the meantime, I'm talking about the number price interest points I win, lose or risk on the next trade.

brent

I think pips are ok... ONLY when reported when in a format such as "I risked XX pips to win YY pips"

Too many people simply say, "I made XXX pips on that trade. WOOHOO! I'm the world's greatest trader!" It doesn't show if the risk was reasonable for the trade... I could care less if he made XXX pips if his risk was XXX pips times 5.

And that's called the risk/reward ratio. Again, just standard terminology.

I think that current/common language already meets our needs. This Percentage Crusade is all about creating a distinction without a difference. In fact it smacks of political correctness with a tinge of elitism.

Forex is always going to have it's fair share of people who don't have a clear understanding of what they're doing or talking about. But they need to grow into having and exhibiting that understanding.

If somebody is more comfortable in dealing with percentages - that's fine. But there's no need to browbeat the rest of us about "what's wrong with us forex traders" just because we disagree.

In the final analysis - this crusade has a long, hard road ahead of it. And I don't believe it will prevail in the end.

So I don't intend to be adopting this form of Newspeak.

So it's newspeak to encourage the use of industry standard ways of reporting results that were in existance long before the forex revolution? If anything, pips are the newspeak.

How do you figure that this is an issue of political correctness when it's about math? That's a wholly inappropriate use of the term. The fact remains that different pairs have different pip values and base currency exposure and that is not a distinction without difference. The difference is real and important.

I am a forex trader, not an outsider. I think it is important to keep these concepts in mind or it's easy to put weight where it isn't deserved. That's how scams get run in forex. But that's the worst case. In the best case, a trader night simply underestimate the exposure they are taking. Like longhorn said, "It shows a fundamental disregard of basic risk management ideas." Well said.

This does seem to be a bit of a personal obsession with you. But I find the whole thing pretty tedious.

I told you that I was pro-choice on the issue. If you're more comfortable dealing in percentages, do it. But I wasn't interested in being browbeat into submission. And that's still true.

So good luck with the crusade. I still believe you'll fall short in converting the whole world to your way of thinking.

P.S. Let me clue you into something that is important. My position on this issue has absolutely, positively NOTHING to do with, "...a fundamental disregard of basic risk management ideas."

That's preposterous!

And you need to stop long enough to let that fact sink in. I'm finished with this discussion about your rant.

Merely dealing in percentages is not adequate either.

Take a fictitious trade with SL of 50 pips and a TP of 100 pips.

So we have a 2:1 Reward to Risk ratio.

Trader A decides that risking 1% of his account to gain 2% is acceptable and does so.

Trader B decides he's much more aggressive (stupid) and risks 10% of his account to win 20%.

BOTH traders only booked 100 pips. But one was being damn stupid and ignored fundamental risk management principles.

The whole point is that reporting merely in pips or percentages without reporting what size of position taken is bunk data.

If you're in a thread where the system always uses a fixed stop loss and only trades a single pair... sure... go ahead and report pips.

But for the majority of FF discussions... the SL changes, or the strategies are diverse, or there's alot of discretion etc...

Pips just don't cut it as an accurate method of reporting profit / loss.

And for the record... I trade currencies exclusively.

longhorn,

All of your points are correct but I was clear that I think issues of account size and risk-taking should be handled independent of the inherent system characteristics.

When you look at a new system and are trying to decide if the system is good or not, you shouldn't care about the risk level the system designer decided to use. You should decide for yourself. If the system designer used fixed lots maybe you you'll decide to use dynamic lots that fit your changing equity or use fewer lots relative to your account.

But that means you need an objective comparison technique that shows the power of each trade independent of the risk relative to your account. In other words, what is the quality of the trade? The only way to do that without getting lost in differing pip values etc is the % of exposure. A 100K lot trade that makes 100 pips in EUR/USD and a 10K lot that makes 10 pips are the same trade in different scales. I'm proposing the % as a scale-invariant method of comparison.

Risk-managment is vitally important, as you point out. But a system's inherent goodness or badness is not related to issues of money managment. The size of the position or leverage you eventually use is.

I hope this clarifies. Basically I think you need various scale-invariant methods of comparing a system then you apply position-sizing as a post-processing step to see if the system will work in your risk tolerance parameters.

Of course, I don't think % return is all that matters. There are other ratios and measure I want to know about a system before I can understand its characteristic feel, like it's expectation (ratio of average trade to average loss), profit factors etc.