Trade Execution, Position Sizing, and Risk Management

In a previous article we talked about the importance of stop losses as a trade exit. Today we're going to talk about entries, also known as trade execution or "pulling the trigger".

A lot of traders (including myself) had trouble with this in particular at first. So if you struggle with execution, you're not alone. You already can perform analysis and have a basic idea of whether you want to be a buyer or be a seller. Your analysis gives you a gut feeling, and often a sense of urgency to act before the opportunity passes by. But the specifics of actually opening that trade are crucial; it's frustrating when you lose money due to a poor entry but ended up being right about the main trend all along.

We're going to solve that problem today. We'll use the following checklist:

1. Context (Pressure) in a higher time frame
2. Unambiguous entry criteria in a lower time frame
3. Initial Stop Placement
4. Perfect position sizing
5. Stop Management
6. Recent Failure, and Persistence

 

Context (Pressure) in a higher time frame

 

Understand that markets move in trends and countertrends. You've heard "the Trend is your Friend", but countertrends are your friend too. Your goal as a trader is to enter a position in the direction of the primary trend only during a countertrend, and to do so at the point in time when that countertrend is likely to have just completed.

This article picks up where that analysis leaves off. You're looking at a chart, you've done your analysis, and have determined that now is (approximately) a good time to buy. Now what? Read on.

 

Unambiguous entry criteria in a lower time frame

 

Your primary analysis should be done on a single time frame (and have knowledge of all higher time frames), but your entry occurs on one time frame (or at most two time frames) lower than that. The time frames I use go like this:

Monthly -> Weekly -> Daily -> 4 Hr -> 1 Hr -> 15 Min -> 5 Min - 1 Min

If Step 1 (above) was performed on a Daily chart, you should think about making your entry on a 4-Hour chart, or at the very most a 1-Hour chart.

The Context analysis determines the "market pressure". Think of this like throwing a ping-pong ball as hard as you can into a strong headwind. It will initially continue to move away from you, but the wind is exerting pressure on the ball to cause it to change direction. A falling market with an upward pressure is similar, and you want to make your entry shortly after the inflexion point as momentum moves from negative to positive.

The best way to do this is with unambiguous entry criteria. The method I use is that of a "reversal bar". It looks like this:

The analogous structure exists for a downside reversal bar, which occurs at resistance. It has a bar close lower than the bar open (black candle), bar close lower than the previous close, and a high "spike" with quick retracement is preferable.

In either case, a so-called "reversal bar" is meaningless if it does not occur both at a support/resistance level and in reversal context of a higher time frame.

The first thing to do after performing your higher-time-frame analysis is to then switch to the lower-time-frame view and wait for price to get close to the execution target region (usually your support/resistance level or cluster of support/resistance levels). This is signified by the red support line shown in the above graphic.

Eventually, price hits that level, and probably pierces it. You're still waiting, and you should see a healthy bounce. Continue to wait for that bar to close. If it does close as a reversal bar, you're ready to make the entry.

But don't click "Buy" just yet - we need to know how much to buy, and for that we need to know our initial stop loss.

 

Initial Stop Placement

 

We cannot know "how much" until we know "how far". Fortunately, we can know exactly where our stop should go.

Your charting package should have an ATR (Average True Range) indicator. ATR(14) is fairly standard and represents the average range of price movement over the past 14 bars. Measuring the ATR(14) from the recent low extreme gives a sane initial stop. By definition, the ATR(14) provides enough "wiggle room" to allow for random price movements.

In the above example (1H chart, and I'm rounding numbers) BTC was $800.00, the low was $790.00 and ATR(14) was $12.00. This puts our initial stop at $778.00 (that is, $790.00 - $12.00).

If we're buying at $800 and have an initial stop at $778, then we're risking $22 per bitcoin purchased.

 

Position sizing

 

A future article will discuss the mathematics behind position sizing. For now, assume that we will risk 1% of our portfolio per trade. This is a fairly middle-of-the-road risk percentage (neither conservative nor aggressive). You can choose anything from 0.5% to 2% initially depending on your nature.

If we have a $10,000 trading account, and want to risk 1% for this trade, then we're looking at a loss of $100. We calculated previously that with an entry at $800 and an initial stop at $778, we're risking $22 per bitcoin. A loss of $22 isn't enough - we want to lose $100, so we need to buy 4.545 BTC to make that happen ($100 / $22 = 4.545).

So, we put in an order for 4.545 BTC at the current price ($800), with a stop at $778. This costs $3636 out of our $10,000, making us 36% leveraged in this trade. (In other words, we're actually 64% cash)

If we wanted to buy more than 4.545 BTC (in order to potentially make more money), then we need a tighter stop loss. But it's unsafe to arbitrarily tighten the stop loss just to buy more. Instead, you do your analysis on a lower time frame, and enter on a lower time frame still. Short time frames, such as a 5-minute chart, offer tighter stop losses.

One thing to keep in mind though is that in an illiquid (crash, or possibly short squeeze) market, it's difficult to get out without slippage. It's also more difficult to perform analysis in shorter time frames, because you need to always include awareness of all higher time frames from the level of Context upward.

 

Stop Management

Stop management is really a large component of position management and not directly related to entry execution, so it will be covered in more detail in a future article. However, what needs to be understood now is that your initial stop using this technique is quite far away. The only time it should be hit is when price goes immediately against you. In most cases if your analysis was good, price will bounce at least short-term after your entry, and you will want to move your stop up to the bottom of the reversal extreme ($790 in our example).

You can then sell a portion of your holdings (1/3 is quite common) in order to lock in profit, and depending on how much room there is between price and your entry level, you might move the stop all the way up to your break-even point. The most important thing is that you've eliminated risk on principal and you're playing with the market's money. There's a good chance it'll go up and a good chance it'll go down, but you don't care as much at this point - you've gotten your principal out, and if you get stopped out you can always re-enter without much harm done.

 

Recent Failure, and Persistence  

 

A single investment opportunity isn't over once you've been stopped out. Persistence is key. If you felt that the initial 0.5% to 2% risk was low, it's because it may take several entries into the same trend to get one that really takes off. So, you might accumulate three or four 1% losses just trying to get a good entry, but it's usually worth it in the end.

The fact is that the odds of success go up after you've been stopped out but the overall context for the trade remains intact. This is simply because the average trader isn't very persistent, and take losses to mean that they did something wrong. The market actively tries to push prices around in ways that will hit your stop loss. When the "stop loss fuel" is used up, the attractive targets are now on the other side of the market, and price goes back that way instead. This is so-called "wash and rinse" action, which was a focal point of this article.

More generally, if last time something failed, then next time that same thing is more likely to work. A naive statistical analysis shows that the overall probability of some technique might be 50%, but when you've got skin in the game and can feel what the different market participants are doing and thinking, your odds of success increase as you identify when a particular technique is likely to work, and when it's not.

The moral of the story: after being stopped out, do your analysis again. If it's still telling you the same thing, enter again on the next execution target. Rarely will this happen more than three times before making all of the losses back and then some.