Test 1: Do Targets Help

Do adding simple profit targets help our trades or hurt them when compared to holding to expiration?

It seems simplistic, but I want to start out simple with these tests to see what particular adjustments add the most value. I want to test the very basic management techniques of setting profit targets and stop losses in isolation and in combination to see whether they actually provide value. Today I start with profit targets. My assumption is that profit targets will help reduce the risk of a trade due to exiting sooner, but the expected return will be lower. I am not sure what overall effect on the profit factor this will have. Lets dig in.

What Method Do We Choose

(I present a side discussion on why I choose my target and stop loss calculations. Feel free to move on to the next section to see results)

When testing targets or stop losses, there are three simple ways to calculate them:

  1. Return on Risk – The return on investment (ROI) or return on risk(ROR) is the gain divided by the actual cash at risk in the trade. For example, a trade that took in a $2 in premium on a 10 point wide condor, would be risking $8; if we were targting a 10% return, we would exit when we reached $0.80 in premium.
  2. Return on Margin – This is a percentage return on the full margin, that is all the cash that is indisposed during the trade. For example, a trade that took in a $2 in premium on a 10-point wide condor would have a margin requirement of $1000 per contract (even though we collected $200); if we had a target of 10%, we would be aiming for a return of $1.00 in net premium.
  3. On Premium – We can also set targets as a percentage of the premium collected on the trade. For example, again a trade that took in a $2 in premium on a 10-point wide condor, a 50% target would aim for $1 in net premium.

There is no incorrect method for a trader. But for my tests, I have to choose one method. For running the tests, On Premium actually provides the best granularity, but ROR is generally what I use for my own trades. Look at the following table as a comparison between the different methods. When we have a trade that takes $1 premium:

Profit Target   Stop Loss
On Risk On Margin On Prem   On Risk On Margin On Prem
          0.45           0.50           0.05 5% -0.45 -0.50 -0.05
          0.90           1.00           0.10 10% -0.90 -1.00 -0.10
          1.00           1.00           0.15 15% -1.35 -1.50 -0.15
          1.00           1.00           0.20 20% -1.80 -2.00 -0.20
          1.00           1.00           0.30 30% -2.70 -3.00 -0.30
          1.00           1.00           0.40 40% -3.60 -4.00 -0.40
          1.00           1.00           0.50 50% -4.50 -5.00 -0.50
          1.00           1.00           0.75 75% -6.75 -7.50 -0.75
          1.00           1.00           1.00 100% -9.00 -9.00 -1.00

Notice how quickly the increasing the target shows fruitless with both On Risk and On Margin. Regardless of how high the target is, we have reached are maximum profit. Alternatively, on the Stop Loss side, On Risk and On Margin do not reach their full loss until at or near 100%. But On Prem has the same issue the other two had with Profit targets; you would actually need to increase the stop loss to 900% to reach the full loss.

But if we change the premium collected to $3.00:

Profit Target   Stop Loss
On Risk On Margin On Prem   On Risk On Margin On Prem
          0.35           0.50           0.15 5% -0.35 -0.50 -0.15
          0.70           1.00           0.30 10% -0.70 -1.00 -0.30
          1.05           1.50           0.45 15% -1.05 -1.50 -0.45
          1.40           2.00           0.60 20% -1.40 -2.00 -0.60
          2.10           3.00           0.90 30% -2.10 -3.00 -0.90
          2.80           3.00           1.20 40% -2.80 -4.00 -1.20
          3.00           3.00           1.50 50% -3.50 -5.00 -1.50
          3.00           3.00           2.25 75% -5.25 -7.00 -2.25
          3.00           3.00           3.00 100% -7.00 -7.00 -3.00

On Risk and On Margin have far larger ranges. This presents another challenge, depending on the premium collected for each trade, the effective target range for both the On Risk and On Margin calculations fluctuates. This actually means that many test scenarios will end up duplicating results which will result in misleading statistics.

So On Risk and On Margin have fluctuating ranges when used for Profit Targets, but their ranges are more appropriate when used as Stop Losses. On Premium has fluctuating ranges when used for Stop Losses, but On Risk doesn’t.

Using On Premium for Profit Targets and On Risk for Stop Losses allows me to use a full range values (10% – 100%) for every scenario, eliminating needless duplications in tests. While this may or may not make intuitive sense to a trader, programmatically it is the correct course of action. It is quiet easy to convert from one form to another.

On a side note, I personally use many of these for different situations. I pay attention to both stop losses as a percentage On Risk and On Premium. I am comfortable setting a 2x  stop loss based on the premium I receive.

On to the Test

The test constructs different scenarios for a variety of deltas and profit targets for the first month, second month and third month expirations.

The test created 270 different permutations. The results were automatically compared to the baseline test (you can view the difference against the baseline with columns named “Diff”) and summerized below:

1 Month 2 Months 3 Months All
Improved PF 26(29%) 79(88%) 88(98%) 193(71%)
Improved MaxDD 90(100%) 90(100%) 90(100%) 270(100%)
Improved ExpRet 26(29%) 59(66%) 68(76%) 153(57%)

Adding just profit targets lead to a large improvement for the longer term tests but less so for the front month scenarios. In every scenario it reduced our maximum drawdown, and surprisingly lead to higher expected returns for the majority of the tests.

If we look at a histogram of the improvement in the Profit Factor, we can see a noticeable pattern. The cells in the table show the difference between the current test run when compared to the baseline. A positive number would represent an improvement in the profit factor. The front-month tests perform worse than simply holding to expiration for the majority of the cases. But all other scenarios generally perform better.

ta-hist-diffPF

Another way to view this data is by graphing the Profit Factor improvement against the delta:

ta-diffPF-vs-delta

Next we look at the Expect Returns of each scenario. The heatmap highlights the best performing scenarios. There is no surprise: holding trades to full profit results in larger expected returns. But notice that the high delta trades all perform poorly.

ta-hist-expRet

Finally, I look at one more interesting chart. The following chart shows the number of trades and their profit factor for each test. Notice how the front month trades were able to rack up many trades. These are very low profit target trades. They were able to achieve their profit targets very quickly, and put another trade on during the same month. But interestingly these scenarios that had lots of little profit trades also had huge losses and wound up with negative expected returns.ta-pf-vs-trades-split

Conclusion

Adding profit targets alone may improve the performance of longer duration trades so long as the Delta is below 30. However, for front month trades, the number of smaller profitable wins does not outweigh the fewer massive losses experienced when holding to expiration. In these cases while the downside risk is improved, the upside return is reduced far more leading to a negative scenario.

Test Stats

Permutations 270
Profitable Scenarios 198 73%
Total Trades 65071
Profitable Trades 30801 47%
Expired 6377 10%
Target Met 58694 90%
Test Duration 1.61(min)

Test Files

Target-Only Excel Data

Advertisements

1 thought on “Test 1: Do Targets Help

  1. Pingback: Test #2: Stop Losses | Option I/O

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s