When to Sell a Position

20th Jul '20 by Jack Corsellis

2 comments 861 reads

Let's look at what is optimal in terms of average percentage win vs average percentage loss across different win rates, assuming a 2:1 gain vs loss ratio.

Note the table in Fig 1. This table shows the optimal 10 trade compounded return on investment for average percentage win vs average percentage loss assuming a 2:1 gain vs loss ratio.

Fig 1:

To put this table into the context of investing, imagine you went to the horse races and there were 10 races to bet on. A magical genie appears and tells you you're going to win on 4 out of 10 horse races (40% win rate).

The magical genie says to you that you can pick how much you'd like to win in percentage terms (0% - 100%) on your 4 winning races. The only caveat is you'll close half that percentage on the 6 losing horse races (2:1 gain vs loss ratio).

So, what do you pick in percentage terms to win on your 4 winning races... 30% ... 50% ... 100%?

Due to the maths involved there must be an optimal answer.

As you can see on the table in Fig 1, the optimal response to the magical genie would be "I'd like to win 20% on my 4 winning races and therefore lose 10% on my 6 losing races".

The magical genie is intrigued, he asks "that's very specific, why did you pick 20% and 10%?".

You reply "because anymore or less at a 40% win rate with a 2:1 gain vs loss ratio and the results deteriorate from the 10.20% compounded return on investment".

Now we're getting somewhere.

So what does any of this mean and how can you use this new knowledge?

Often times we get confused for when to sell out of a position and lack a clear rationale for as to why. Surely it's a better strategy to target what is optimal and reverse engineer.

By knowing how you can be profitable at a 40% win rate, you can start to build failure into your trading process. I'm not saying you're going to commit 100% of your capital to each trade, that's not sound risk management nor optimal position sizing (which I'll cover in future posts). It is important to start from a place of optimisation.

Remember, the timeframe you trade on…

Unlock with Email

Unlock with Google

Already have an account?
Login here

gattis01

20th Jan '21

I like the math but it doesn't seem to help me deciding how to trim my portfolio. I've held RIO for many years and now it is my most significant holding. The question is, given the current circumstances, if I sell will I find it easy to find a better home for the proceeds.

LeeA_uk

21st Jan '21

Interesting post and it certainly is a great idea to have a committed target portfolio strategy.

I would say the method is more relevant to short term trading rather than longer term holdings as I like to measure when to sell more based on the performance and long term value of a company, rather than purely on a pre-determined %. As Gattis mentions below, for a longer term position I'm not sure this method would help determine if it is actually a good time to sell.

Can the above method also have some additional factors built in to determine if the position has potential future gain and also incorporate some of your volume trading methods etc? This would then become a really useful tool to actually determine if positions are forecast for more gain or might be flattening out.

Like if you could plug in the data of your portfolio and set the recent volume data, trading levels, price point, rating (your own assessment plus stockopedia ratings), etc? Then this could forecast a target sell amount and/or time?

Obviously this would need to be essentially based on your own forecast for the stock but would be a very handy tool if it could include some of the rationale you are getting at.

When to Sell a Position

Unlock the rest of this article with a 14 day trial

About the Author

Jack Corsellis

2 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...