Hi guys,
I have been using reddit for years in my personal life (not trading!) and wanted to give something back in an area where i am an expert.
I worked at an investment bank for seven years and joined them as a graduate FX trader so have lots of professional experience, by which i mean I was trained and paid by a big institution to trade on their behalf. This is very different to being a full-time home trader, although that is not to discredit those guys, who can accumulate a good amount of experience/wisdom through self learning.
When I get time I'm going to write a mid-length posts on each topic for you guys along the lines of how i was trained. I guess there would be 15-20 topics in total so about 50-60 posts. Feel free to comment or ask questions.
The first topic is Risk Management and we'll cover it in three parts
Part I

• Why it matters
• Position sizing
• Kelly
• Using stops sensibly
• Picking a clear level

Why it matters

The first rule of making money through trading is to ensure you do not lose money. Look at any serious hedge fund’s website and they’ll talk about their first priority being “preservation of investor capital.”
You have to keep it before you grow it.
Strangely, if you look at retail trading websites, for every one article on risk management there are probably fifty on trade selection. This is completely the wrong way around.
The great news is that this stuff is pretty simple and process-driven. Anyone can learn and follow best practices.
Seriously, avoiding mistakes is one of the most important things: there's not some holy grail system for finding winning trades, rather a routine and fairly boring set of processes that ensure that you are profitable, despite having plenty of losing trades alongside the winners.

Capital and position sizing

The first thing you have to know is how much capital you are working with. Let’s say you have $100,000 deposited. This is your maximum trading capital. Your trading capital is not the leveraged amount. It is the amount of money you have deposited and can withdraw or lose.
Position sizing is what ensures that a losing streak does not take you out of the market.
A rule of thumb is that one should risk no more than 2% of one’s account balance on an individual trade and no more than 8% of one’s account balance on a specific theme. We’ll look at why that’s a rule of thumb later. For now let’s just accept those numbers and look at examples.
So we have $100,000 in our account. And we wish to buy EURUSD. We should therefore not be risking more than 2% which $2,000.
We look at a technical chart and decide to leave a stop below the monthly low, which is 55 pips below market. We’ll come back to this in a bit. So what should our position size be?
We go to the calculator page, select Position Size and enter our details. There are many such calculators online - just google "Pip calculator".
​
https://preview.redd.it/y38zb666e5h51.jpg?width=1200&format=pjpg&auto=webp&s=26e4fe569dc5c1f43ce4c746230c49b138691d14
So the appropriate size is a buy position of 363,636 EURUSD. If it reaches our stop level we know we’ll lose precisely $2,000 or 2% of our capital.
You should be using this calculator (or something similar) on every single trade so that you know your risk.
Now imagine that we have similar bets on EURJPY and EURGBP, which have also broken above moving averages. Clearly this EUR-momentum is a theme. If it works all three bets are likely to pay off. But if it goes wrong we are likely to lose on all three at once. We are going to look at this concept of correlation in more detail later.
The total amount of risk in our portfolio - if all of the trades on this EUR-momentum theme were to hit their stops - should not exceed $8,000 or 8% of total capital. This allows us to go big on themes we like without going bust when the theme does not work.
As we’ll see later, many traders only win on 40-60% of trades. So you have to accept losing trades will be common and ensure you size trades so they cannot ruin you.
Similarly, like poker players, we should risk more on trades we feel confident about and less on trades that seem less compelling. However, this should always be subject to overall position sizing constraints.
For example before you put on each trade you might rate the strength of your conviction in the trade and allocate a position size accordingly:
​
https://preview.redd.it/q2ea6rgae5h51.png?width=1200&format=png&auto=webp&s=4332cb8d0bbbc3d8db972c1f28e8189105393e5b
To keep yourself disciplined you should try to ensure that no more than one in twenty trades are graded exceptional and allocated 5% of account balance risk. It really should be a rare moment when all the stars align for you.
Notice that the nice thing about dealing in percentages is that it scales. Say you start out with $100,000 but end the year up 50% at $150,000. Now a 1% bet will risk $1,500 rather than $1,000. That makes sense as your capital has grown.
It is extremely common for retail accounts to blow-up by making only 4-5 losing trades because they are leveraged at 50:1 and have taken on far too large a position, relative to their account balance.
Consider that GBPUSD tends to move 1% each day. If you have an account balance of $10k then it would be crazy to take a position of $500k (50:1 leveraged). A 1% move on $500k is $5k.
Two perfectly regular down days in a row — or a single day’s move of 2% — and you will receive a margin call from the broker, have the account closed out, and have lost all your money.
Do not let this happen to you. Use position sizing discipline to protect yourself.
​

Kelly Criterion

If you’re wondering - why “about 2%” per trade? - that’s a fair question. Why not 0.5% or 10% or any other number?
The Kelly Criterion is a formula that was adapted for use in casinos. If you know the odds of winning and the expected pay-off, it tells you how much you should bet in each round.
This is harder than it sounds. Let’s say you could bet on a weighted coin flip, where it lands on heads 60% of the time and tails 40% of the time. The payout is $2 per $1 bet.
Well, absolutely you should bet. The odds are in your favour. But if you have, say, $100 it is less obvious how much you should bet to avoid ruin.
Say you bet $50, the odds that it could land on tails twice in a row are 16%. You could easily be out after the first two flips.
Equally, betting $1 is not going to maximise your advantage. The odds are 60/40 in your favour so only betting $1 is likely too conservative. The Kelly Criterion is a formula that produces the long-run optimal bet size, given the odds.
Applying the formula to forex trading looks like this:
Position size % = Winning trade % - ( (1- Winning trade %) / Risk-reward ratio
If you have recorded hundreds of trades in your journal - see next chapter - you can calculate what this outputs for you specifically.
If you don't have hundreds of trades then let’s assume some realistic defaults of Winning trade % being 30% and Risk-reward ratio being 3. The 3 implies your TP is 3x the distance of your stop from entry e.g. 300 pips take profit and 100 pips stop loss.
So that’s 0.3 - (1 - 0.3) / 3 = 6.6%.
Hold on a second. 6.6% of your account probably feels like a LOT to risk per trade.This is the main observation people have on Kelly: whilst it may optimise the long-run results it doesn’t take into account the pain of drawdowns. It is better thought of as the rational maximum limit. You needn’t go right up to the limit!
With a 30% winning trade ratio, the odds of you losing on four trades in a row is nearly one in four. That would result in a drawdown of nearly a quarter of your starting account balance. Could you really stomach that and put on the fifth trade, cool as ice? Most of us could not.
Accordingly people tend to reduce the bet size. For example, let’s say you know you would feel emotionally affected by losing 25% of your account.
Well, the simplest way is to divide the Kelly output by four. You have effectively hidden 75% of your account balance from Kelly and it is now optimised to avoid a total wipeout of just the 25% it can see.
This gives 6.6% / 4 = 1.65%. Of course different trading approaches and different risk appetites will provide different optimal bet sizes but as a rule of thumb something between 1-2% is appropriate for the style and risk appetite of most retail traders.
Incidentally be very wary of systems or traders who claim high winning trade % like 80%. Invariably these don’t pass a basic sense-check:

• How many live trades have you done? Often they’ll have done only a handful of real trades and the rest are simulated backtests, which are overfitted. The model will soon die.
• What is your risk-reward ratio on each trade? If you have a take profit $3 away and a stop loss $100 away, of course most trades will be winners. You will not be making money, however! In general most traders should trade smaller position sizes and less frequently than they do. If you are going to bias one way or the other, far better to start off too small.

How to use stop losses sensibly

Stop losses have a bad reputation amongst the retail community but are absolutely essential to risk management. No serious discretionary trader can operate without them.
A stop loss is a resting order, left with the broker, to automatically close your position if it reaches a certain price. For a recap on the various order types visit this chapter.
The valid concern with stop losses is that disreputable brokers look for a concentration of stops and then, when the market is close, whipsaw the price through the stop levels so that the clients ‘stop out’ and sell to the broker at a low rate before the market naturally comes back higher. This is referred to as ‘stop hunting’.
This would be extremely immoral behaviour and the way to guard against it is to use a highly reputable top-tier broker in a well regulated region such as the UK.
Why are stop losses so important? Well, there is no other way to manage risk with certainty.
You should always have a pre-determined stop loss before you put on a trade. Not having one is a recipe for disaster: you will find yourself emotionally attached to the trade as it goes against you and it will be extremely hard to cut the loss. This is a well known behavioural bias that we’ll explore in a later chapter.
Learning to take a loss and move on rationally is a key lesson for new traders.
A common mistake is to think of the market as a personal nemesis. The market, of course, is totally impersonal; it doesn’t care whether you make money or not.
Bruce Kovner, founder of the hedge fund Caxton Associates
There is an old saying amongst bank traders which is “losers average losers”.
It is tempting, having bought EURUSD and seeing it go lower, to buy more. Your average price will improve if you keep buying as it goes lower. If it was cheap before it must be a bargain now, right? Wrong.
Where does that end? Always have a pre-determined cut-off point which limits your risk. A level where you know the reason for the trade was proved ‘wrong’ ... and stick to it strictly. If you trade using discretion, use stops.

Picking a clear level

Where you leave your stop loss is key.
Typically traders will leave them at big technical levels such as recent highs or lows. For example if EURUSD is trading at 1.1250 and the recent month’s low is 1.1205 then leaving it just below at 1.1200 seems sensible.
​
If you were going long, just below the double bottom support zone seems like a sensible area to leave a stop
You want to give it a bit of breathing room as we know support zones often get challenged before the price rallies. This is because lots of traders identify the same zones. You won’t be the only one selling around 1.1200.
The “weak hands” who leave their sell stop order at exactly the level are likely to get taken out as the market tests the support. Those who leave it ten or fifteen pips below the level have more breathing room and will survive a quick test of the level before a resumed run-up.
Your timeframe and trading style clearly play a part. Here’s a candlestick chart (one candle is one day) for GBPUSD.
​
https://preview.redd.it/moyngdy4f5h51.png?width=1200&format=png&auto=webp&s=91af88da00dd3a09e202880d8029b0ddf04fb802
If you are putting on a trend-following trade you expect to hold for weeks then you need to have a stop loss that can withstand the daily noise. Look at the downtrend on the chart. There were plenty of days in which the price rallied 60 pips or more during the wider downtrend.
So having a really tight stop of, say, 25 pips that gets chopped up in noisy short-term moves is not going to work for this kind of trade. You need to use a wider stop and take a smaller position size, determined by the stop level.
There are several tools you can use to help you estimate what is a safe distance and we’ll look at those in the next section.
There are of course exceptions. For example, if you are doing range-break style trading you might have a really tight stop, set just below the previous range high.
​
https://preview.redd.it/ygy0tko7f5h51.png?width=1200&format=png&auto=webp&s=34af49da61c911befdc0db26af66f6c313556c81
Clearly then where you set stops will depend on your trading style as well as your holding horizons and the volatility of each instrument.
Here are some guidelines that can help:

1. Use technical analysis to pick important levels (support, resistance, previous high/lows, moving averages etc.) as these provide clear exit and entry points on a trade.
2. Ensure that the stop gives your trade enough room to breathe and reflects your timeframe and typical volatility of each pair. See next section.
3. Always pick your stop level first. Then use a calculator to determine the appropriate lot size for the position, based on the % of your account balance you wish to risk on the trade.

So far we have talked about price-based stops. There is another sort which is more of a fundamental stop, used alongside - not instead of - price stops. If either breaks you’re out.
For example if you stop understanding why a product is going up or down and your fundamental thesis has been confirmed wrong, get out. For example, if you are long because you think the central bank is turning hawkish and AUDUSD is going to play catch up with rates … then you hear dovish noises from the central bank and the bond yields retrace lower and back in line with the currency - close your AUDUSD position. You already know your thesis was wrong. No need to give away more money to the market.

Coming up in part II

EDIT: part II here
Letting stops breathe
When to change a stop
Entering and exiting winning positions
Risk:reward ratios
Risk-adjusted returns

Coming up in part III

Squeezes and other risks
Market positioning
Bet correlation
Crap trades, timeouts and monthly limits
​
***
Disclaimer:This content is not investment advice and you should not place any reliance on it. The views expressed are the author's own and should not be attributed to any other person, including their employer.

Thanks for all the upvotes and comments on the previous pieces:

• Risk management parts 1,2,3
• News trading part 1

From the first half of the news trading note we learned some ways to estimate what is priced in by the market. We learned that we are trading any gap in market expectations rather than the result itself. A good result when the market expected a fantastic result is disappointing! We also looked at second order thinking. After all that, I hope the reaction of prices to events is starting to make more sense to you.
​
Before you understand the core concepts of pricing in and second order thinking, price reactions to events can seem mystifying at times
We'll add one thought-provoking quote. Keynes (that rare economist who also managed institutional money) offered this analogy. He compared selecting investments to a beauty contest in which newspaper readers would write in with their votes and win a prize if their votes most closely matched the six most popularly selected women across all readers:
It is not a case of choosing those (faces) which, to the best of one’s judgment, are really the prettiest, nor even those which average opinions genuinely thinks the prettiest. We have reached the third degree where we devote our intelligences to anticipating what average opinion expects the average opinion to be.
Trading is no different. You are trying to anticipate how other traders will react to news and how that will move prices. Perhaps you disagree with their reaction. Still, if you can anticipate what it will be you would be sensible to act upon it. Don't forget: meanwhile they are also trying to anticipate what you and everyone else will do.
​
Part II

• Preparing for quantitative and qualitative releases
• Data surprise index
• Using recent events to predict future reactions
• Buy the rumour, sell the fact
• The trimming position effect
• Reversals
• Some key FX releases

​

Preparing for quantitative and qualitative releases

The majority of releases are quantitative. All that means is there’s some number. Like unemployment figures or GDP.
Historic results provide interesting context. We are looking below the Australian unemployment rate which is released monthly. If you plot it out a few years back you can spot a clear trend, which got massively reversed. Knowing this trend gives you additional information when the figure is released. In the same way prices can trend so do economic data.
​
A great resource that's totally free to use
This makes sense: if for example things are getting steadily better in the economy you’d expect to see unemployment steadily going down.
Knowing the trend and how much noise there is in the data gives you an informational edge over lazy traders.
For example, when we see the spike above 6% on the above you’d instantly know it was crazy and a huge trading opportunity since a) the fluctuations month on month are normally tiny and b) it is a huge reversal of the long-term trend.
Would all the other AUDUSD traders know and react proportionately? If not and yet they still trade, their laziness may be an opportunity for more informed traders to make some money.
Tradingeconomics.com offers really high quality analysis. You can see all the major indicators for each country. Clicking them brings up their history as well as an explanation of what they show.
For example, here’s German Consumer Confidence.
​
Helpful context
There are also qualitative events. Normally these are speeches by Central Bankers.
There are whole blogs dedicated to closely reading such texts and looking for subtle changes in direction or opinion on the economy. Stuff like how often does the phrase "in a good place" come up when the Chair of the Fed speaks. It is pretty dry stuff. Yet these are leading indicators of how each member may vote to set interest rates. Ed Yardeni is the go-to guy on central banks.

Data surprise index

The other thing you might look at is something investment banks produce for their customers. A data surprise index. I am not sure if these are available in retail land - there's no reason they shouldn't be but the economic calendars online are very basic.
You’ll remember we talked about data not being good or bad of itself but good or bad relative to what was expected. These indices measure this difference.
If results are consistently better than analysts expect then you’ll see a positive number. If they are consistently worse than analysts expect a negative number. You can see they tend to swing from positive to negative.
​
Mean reversion at its best! Data surprise indices measure how much better or worse data came in vs forecast
There are many theories for this but in general people consider that analysts herd around the consensus. They are scared to be outliers and look ‘wrong’ or ‘stupid’ so they instead place estimates close to the pack of their peers.
When economic conditions change they may therefore be slow to update. When they are wrong consistently - say too bearish - they eventually flip the other way and become too bullish.
These charts can be interesting to give you an idea of how the recent data releases have been versus market expectations. You may try to spot the turning points in macroeconomic data that drive long term currency prices and trends.
​

Using recent events to predict future reactions

The market reaction function is the most important thing on an economic calendar in many ways. It means: what will happen to the price if the data is better or worse than the market expects?
That seems easy to answer but it is not.
Consider the example of consumer confidence we had earlier.

• Many times the market will shrug and ignore it.
• But when the economic recovery is predicated on a strong consumer it may move markets a lot.

Or consider the S&P index of US stocks (Wall Street).

• If you get good economic data that beats analyst estimates surely it should go up? Well, sometimes that is certainly the case.
• But good economic data might result in the US Central Bank raising interest rates. Raising interest rates will generally make the stock market go down!

So better than expected data could make the S&P go up (“the economy is great”) or down (“the Fed is more likely to raise rates”). It depends. The market can interpret the same data totally differently at different times.
One clue is to look at what happened to the price of risk assets at the last event.
For example, let’s say we looked at unemployment and it came in a lot worse than forecast last month. What happened to the S&P back then?
​
2% drop last time on a 'worse than expected' number ... so it it is 'better than expected' best guess is we rally 2% higher
So this tells us that - at least for our most recent event - the S&P moved 2% lower on a far worse than expected number. This gives us some guidance as to what it might do next time and the direction. Bad number = lower S&P. For a huge surprise 2% is the size of move we’d expect.
Again - this is a real limitation of online calendars. They should show next to the historic results (expected/actual) the reaction of various instruments.

Buy the rumour, sell the fact

A final example of an unpredictable reaction relates to the old rule of ‘Buy the rumour, sell the fact.’ This captures the tendency for markets to anticipate events and then reverse when they occur.
​
Buy the rumour, sell the fact
In short: people take profit and close their positions when what they expected to happen is confirmed.
So we have to decide which driver is most important to the market at any point in time. You obviously cannot ask every participant. The best way to do it is to look at what happened recently. Look at the price action during recent releases and you will get a feel for how much the market moves and in which direction.

Trimming or taking off positions

One thing to note is that events sometimes give smart participants information about positioning. This is because many traders take off or reduce positions ahead of big news events for risk management purposes.
Imagine we see GBPUSD rises in the hour before GDP release. That probably indicates the market is short and has taken off / flattened its positions.
​
The price action before an event can tell you about speculative positioning
If GDP is merely in line with expectations those same people are likely to add back their positions. They avoided a potential banana skin. This is why sometimes the market moves on an event that seemingly was bang on consensus.
But you have learned something. The speculative market is short and may prove vulnerable to a squeeze.

Two kinds of reversals

Fairly often you’ll see the market move in one direction on a release then turn around and go the other way.
These are known as reversals. Traders will often ‘fade’ a move, meaning bet against it and expect it to reverse.

Logical reversals

Sometimes this happens when the data looks good at first glance but the details don’t support it.
For example, say the headline is very bullish on German manufacturing numbers but then a minute later it becomes clear the company who releases the data has changed methodology or believes the number is driven by a one-off event. Or maybe the headline number is positive but buried in the detail there is a very negative revision to previous numbers.
Fading the initial spike is one way to trade news. Try looking at what the price action is one minute after the event and thirty minutes afterwards on historic releases.

Crazy reversals

​
Some reversals don't make sense
Sometimes a reversal happens for seemingly no fundamental reason. Say you get clearly positive news that is better than anyone expects. There are no caveats to the positive number. Yet the price briefly spikes up and then falls hard. What on earth?
This is a pure supply and demand thing. Even on bullish news the market cannot sustain a rally. The market is telling you it wants to sell this asset. Try not to get in its way.

Some key releases

As we have already discussed, different releases are important at different times. However, we’ll look at some consistently important ones in this final section.

Interest rates decisions

These can sometimes be unscheduled. However, normally the decisions are announced monthly. The exact process varies for each central bank. Typically there’s a headline decision e.g. maintain 0.75% rate.
You may also see “minutes” of the meeting in which the decision was reached and a vote tally e.g. 7 for maintain, 2 for lower rates. These are always top-tier data releases and have capacity to move the currency a lot.
A hawkish central bank (higher rates) will tend to move a currency higher whilst a dovish central bank (lower rates) will tend to move a currency lower.
A central banker speaking is always a big event

Non farm payrolls

These are released once per month. This is another top-tier release that will move all USD pairs as well as equities.
There are three numbers:

• The headline number of jobs created (bigger is better)
• The unemployment rate (smaller is better)
• Average hourly earnings (depends)

Bear in mind these headline numbers are often off by around 75,000. If a report comes in +/- 25,000 of the forecast, that is probably a non event.
In general a positive response should move the USD higher but check recent price action.
Other countries each have their own unemployment data releases but this is the single most important release.

Surveys

There are various types of surveys: consumer confidence; house price expectations; purchasing managers index etc.
Each one basically asks a group of people if they expect to make more purchases or activity in their area of expertise to rise. There are so many we won’t go into each one here.
A really useful tool is the tradingeconomics.com economic indicators for each country. You can see all the major indicators and an explanation of each plus the historic results.

GDP

Gross Domestic Product is another big release. It is a measure of how much a country’s economy is growing.
In general the market focuses more on ‘advance’ GDP forecasts more than ‘final’ numbers, which are often released at the same time.
This is because the final figures are accurate but by the time they come around the market has already seen all the inputs. The advance figure tends to be less accurate but incorporates new information that the market may not have known before the release.
In general a strong GDP number is good for the domestic currency.

Inflation

Countries tend to release measures of inflation (increase in prices) each month. These releases are important mainly because they may influence the future decisions of the central bank, when setting the interest rate.
See the FX fundamentals section for more details.

Industrial data

Things like factory orders or or inventory levels. These can provide a leading indicator of the strength of the economy.
These numbers can be extremely volatile. This is because a one-off large order can drive the numbers well outside usual levels.
Pay careful attention to previous releases so you have a sense of how noisy each release is and what kind of moves might be expected.

Comments

Often there is really good stuff in the comments/replies. Check out 'squitstoomuch' for some excellent observations on why some news sources are noisy but early (think: Twitter, ZeroHedge). The Softbank story is a good recent example: was in ZeroHedge a day before the FT but the market moved on the FT. Also an interesting comment on mistakes, which definitely happen on breaking news, and can cause massive reversals.
​