Revisiting Stop-Loss and Profit Target Affect on Win Rate

May 30, 2008: 1:39 PM CST

What is the larger relationship between your stop-loss strategy and your profit target method?  I tested three strategies and evaluated the change in win rate – % Profitable Trades – across 400 tests per strategy and present the abbreviated results here.

First, let’s define the methods:

DIA Daily chart from Jan 1 1998 to May 29, 2008 for almost 10 years worth of data.
TradeStation custom strategies – optimized data imported into Excel Surface Charts.
Comparing the cross-tabulation of 400 studies (per examination) on the %Win Rate (profitable trades).
All strategies were ‘long only’ entry strategies.

First, let’s look at the Momentum Long Entry Strategy (buys when indicator “Momentum” crosses the 0 line and then price appreciates one day after this cross occurred).

Next, let’s examine the MACD Cross Strategy (buys the open of the next bar when the MACD line crosses above the exponential average signal line of the MACD):

Finally, let’s examine the 20 Period Simple Moving Average Price Cross Strategy (buys when price crosses above the 20 day simple moving averages and closes above it for one day).

Be sure to click each chart for the full-size image.

The relationship I discussed earlier is clear and makes logical sense.

The larger your stop, the greater your win rate.

The smaller your profit target, the greater your win rate.

In fact, in all three tests, you could achieve a 95% or better win rate if your stop-loss was $17 to $20 with a profit target of only $1.  This result held true in all three tests (with the Momentum test showing the most robust high win-rates).  The Momentum strategy actually showed a win-rate greater than 90% for any value above an $8 stop-loss (with a corresponding profit target of $1).

Keep in mind that using the DIA, $1.00 corresponds with 100 Dow Points.  This essentially means that with each entry, you play for 100 Dow points with a stop-loss 1,700 to 2,000 Dow Points away.

As expected, the win rate plunged to 6% across all tests with a Stop-Loss of $1 and a profit target of $20.

What was the % win rate of all strategies when the target was $1 and the stop was also $1?

Momentum:      51.5% (361 trades)
MACD:             46.22% (106 trades)
20-MA Cross:   54.3% (151 trades)

Let’s look at the classic “3 to 1” reward to risk relationship (target = $3 with stop = $1)

Momentum:      29.38%  (211 trades)
MACD:             27.27%
(99 trades)
20-MA Cross:   35.11%
(131 trades)

How about the corresponding “3 to 1” with looking at a target of $6 and a stop-loss of $2?

Momentum:      26.00% (100 trades)
MACD:             25.33%
(75 trades)
20-MA Cross:   23.80%
(84 trades)

And for fun, what about the ‘for-test-only’ value of a $1 profit target with a $20 stop?

Momentum:      96.50% (115 trades)
MACD:             95.08%
(61 trades)
20-MA Cross:   94.73%
(76 trades)

Was the system net-profitable for the maximum values tested?  In other words, with a win-rate of 95% or better, what was the overall net-profit?

That’s to be discussed in a later post, but I wanted to pose the question:

What do you Think?

Do you think any (or all) of the strategies will yield net profitable results with such a high win-rate?

Stay tuned to find out!

7 Comments

7 Responses to “Revisiting Stop-Loss and Profit Target Affect on Win Rate”

  1. Cedric Says:

    If you risk $20 to make $1 then with a win rate of 95%, out of 100 trades you will lose 5 times for a loss of $100 and you will win 95 times for a gain of $95. I guess you can go broke being right.

  2. Corey Rosenbloom Says:

    Right on! But I’ll wait for a few more thoughts before revealing the results.

  3. Eyal Says:

    It’s a good thing that you present it that way because so many traders on the internet tell you that they have a X% of profitable trades, but they don’t tell you their bottom line – which counts the most !

  4. Wes Says:

    Isn’t it just:
    (Win %) * (Profit Target) – (100% – Win %) * (Stop)

    So for the $6/$2 with the Momentum (roundest numbers):
    0.26 * 6 – (1 – 0.26) * 2 = 0.08 (Net Profitable)

    A tiny edge, but an edge nonetheless. Although most brokerages would eat that tiny edge up in fees.

  5. Corey Rosenbloom Says:

    Eyal,

    Absolutely true. Even win rates of 70% or better might not lead to instant profits. It depends on how much profit you take in when you get a win and how much you give back when you take a stop-loss. Excellent point.

  6. Corey Rosenbloom Says:

    Wes,

    Right. That’s a formula for Expectancy, or the profit factor you can expect per each dollar risked.

    In the upcoming example, I am going to take out commissions and slippage from the net profit per trade and likely report pre-commission and post-commission numbers.

    Small edges can indeed be profitable, but they need to still have edge once all fees are factored in or else it’s a net losing game.

  7. Craig Pritchard Says:

    Corey,

    The interesting thing I noticed about the three charts you posted is that except at the extremes they all look the same. This seems to support the notion I first saw in Van Tharp’s book “Trade Your Way To Financial Freedom” that entry method is one of the least important factors in trading success. Your analysis seems to show that regardless of entry method, the most important factors are choice of stop loss and exit.