Stop Loss System

 

We arrive at this problem by basing the position size on the ATR and amount you’re willing to risk per position (2 percent, or $1,000). Instead of calculating the position size, you can determine the position size up front by using the dollar amount you’re willing to commit to any given position ($5,000). You can relate this position size to the ATR, calculating risk as your final input. (See Table 1.9)

Position Size	Price	Shares	14 day ATR Value	Risk
5000	145.85	34	2.71	92.90

Table 1. 9 -- Risk

You might be shocked at such a small position size, a mere 34 shares? How about the miniscule amount of risk? You’re only risking $92.90, or roughly 0.2 percent of your $50,000 account. But this is the nature of a volatility based stop. You’re wrapping what you have to work with (your account equity) around the ever-changing volatility characteristics of the position.

You have some decisions to make if this seems out of whack. You can either increase your position size at the detriment of diversification, which is a bad decision, or you can increase your stop size to the extent that it equals your acceptable risk (2 percent, or $1,000). (This is known as a fixed stop.) Increasing your stop size to 2 percent in this example leaves you with a stop loss of about $29 ($1,000 / 34). That’s a pretty big stop for a $145.85 stock. But that’s a possibility, depending on your trading style. For example, you might be a long-term investor who holds positions for months or years, in which case a $29 stop on a $145.85 stock is perfectly acceptable.

To put perspective on calculating risk as the last step, and using the dollar amount allocated to each position ($5,000) to determine the position size, let’s take a look at ten different stocks, including the SPY. (See Table 1.10) Note the varying prices, position sizes, and ATR values across these ten stocks.

Stock	Position Size	Price	Shares	14 day ATR Value	Risk
				(Stop)
SPY	5000	145.85	34	2.71	92.90
RIMM	5000	107.57	46	7.97	370.46
GOOG	5000	633.63	8	23.25	183.47
MSFT	5000	34.09	147	0.92	134.94
BIDU	5000	314.99	16	26.13	414.78
DRYS	5000	89.11	56	10.68	599.26
IBM	5000	104.79	48	2.97	141.71
C	5000	34.00	147	1.79	263.24
EMC	5000	19.64	255	1.00	254.58
XOM	5000	85.10	59	2.43	142.77

Table 1. 10 – Volatility Based Stops

As a reminder, this table is constructed using the 14 day ATR for each stock as the stop loss. Note the varying levels of risk across these stocks. By using the ATR for each stock, you can adjust to that stock’s unique volatility.

Across these ten stocks, the total amount of the portfolio at risk is roughly 5.2 percent. Simply add up the total risk column, and divide the total by $50,000 to arrive at 5.2 percent. This risk is well within your limits (i.e. it’s less than 2 percent per position). But even though the risk totals less than your limits, there’s actually more of your portfolio at risk.

What happens if a stock gaps below your stop after a bad earnings report or other adverse bit of news? Is it possible? Of course it is. It happens all of the time. Take a look at the gap after Google (GOOG) reported earnings. (See Figure 1.11)

 

Figure 1. 11 – GOOG Gap

 

In the example in Figure 11, GOOG gapped lower from the previous day’s close to the next day’s open by approximately $40. What’s stunning is that the stock’s 14 day ATR at the time was around $8.50. In short, you might have done everything right but been blindsided by the market when GOOG gapped lower by nearly 5 times its 14 day ATR.

Granted, the GOOG example in Figure 11 is an extreme one. Gaps like this don’t happen all of the time. But it does illustrate a point and the example reinforces why building in some room for error in your risk management makes a lot of sense.

After viewing the GOOG example, maybe the risk totals from the Volatility Based Stop method start to make more sense. Yes, the position sizes are pretty small, but you’re diversified and you’re protecting yourself against adverse events such as the GOOG gap lower. Most importantly, you’re placing your stops based on the unique characteristics of each position and not a random number. Remember, volatility levels change, so the Stop Loss System is dynamic.