Thoughts on the risks of shorting

I'd like to discuss shorting for a second. I only recently started really digging into shorting and how it operates, but I'd like to clear my head of a few notions and see if anyone else comes to the same conclusion. I'll number my thoughts:

1.) Price can go up to infinity, and can only go down to zero. Price cannot ever become negative in magnitude.
2.) Taking a long position, you assume a reward of potentially infinity, and a risk of your account going to zero.
3.) Taking a short position, you assume a reward that approaches some limit as the underlying asset's price goes to zero, and a risk of potentially infinity.

Let me state #3 in another way: with a short position, your gains are limited to how far down a price can go before it hits zero, and ITS RISK IS INFINITY. This is inherent behavior of taking short positions, but I somehow don't think many people fully grasp the magnitude of this:

If you short a $100 stock, and its price suddenly quadruples over night (very rare possibility, but let's entertain the thought for just a moment), then you owe the brokerage $100 (the $100 you were given as proceeds when the short was initiated) and also an additional $300 that you may not have had in the first place.

Let's say we have a decent algorithm that has been running for years, and you've managed to amass nearly $100,000 from taking alternating long and short positions. This is fantastic. However, be aware that if you fully leverage your account towards a short, you are making yourself extremely vulnerable to losses when flash price shifts occur. If, for example, you set a trailing stop on a short at 2% loss, then a massive price shift occurs in the market to increase the price of the underlying asset by 50%, your algorithm will do a simple comparison and say: "hmm, 50% loss is higher than 2%, so we'd better stop these losses". It then cashes out the short and you're on the hook for paying out 50% of your account's value.

This is an extreme case to be sure, but not so extreme when you find yourself with a large-ish amount of money and alternating long and short positions using full leverage on a stock that is volatile in nature (micro cap companies, etc.). For these volatile stocks, its not unheard of for pump and dump schemes to conduct a hit and run, drive the price up by 10x, then leave. By that point, your algorithm's trailing stops might have inadvertently screwed you over by realizing those losses in an attempt to save your account, but failed to do so quickly enough. If you fully leveraged $100,000 on a short in THIS scenario, you'd be on the hook for over a million dollars!

I just want to plant the seed in everyone's mind that, if you've crafted a clever algorithm that makes use of shorts, to be very careful about full leverage of your accounts. As your portfolio value increases, so does the danger.

Thanks for listening :-)

Link for consideration:

Update Backtest

Thanks for sharing. I think you captured the biggest danger with shorting very nicely. I didn't look at your link initially but immediately an article came to mind that I read a while back. I typed in "short lost 100k" in Google and it was the first link. I only realised you also posted that link after coming back here :)

But yeah, that is a notorious exampe of a 'short went bad'. Proper risk management and staying away from the small- to micro-caps should go a long way in preventing something like this from happening to you. But in any case, "Learn from the mistakes of others. You can't live long enough to make them all yourself". I hope you prevent somebody's biggest mistake with this lesson :)


Thanks! And yeah, that $KBIO trade is the stuff of nightmares. Amassing a large portfolio value with a clever algorithm can lure one into a false sense of security. Ironically the more successful you are, the more vulnerable you are to an errant short. With longs, at least you know you can't do worse than just losing your portfolio. With shorts, you can go into debt.


As an option to shorts, I've been looking into using Inverse ETFs as an alternative -- especially for accounts (like IRAs) which cannot trade on margin. They have some shortcomings themselves, of course.


I wrote down your comment on a sticky note and put it with the rest of my "good to know" notes, Jonathan. Very good advice.

I wonder if anyone in the trading universe has identified nearly anti-correlated ETFs that one could use as a functional short position? That would be incredible.

A cursory google search shows that XIV seems to operate fairly antagonistically to VIX, for example.


I wonder if anyone in the trading universe has identified nearly anti-correlated ETFs that one could use as a functional short position? That would be incredible.

Yes, Inverse ETFs. ;)

A cursory google search shows that XIV seems to operate fairly antagonistically to VIX, for example.

Well, yes. That's because it's an Inverse ETN for VIX. ;)

Just curious about this statement you made:

Ironically the more successful you are, the more vulnerable you are to an errant short.

Why do you think that's the case? I think that depends on two things: (1) position sizing and (2) diversification.

IMO, if (1) stays constant (e.g. investing the same percentage of your portfolio), then the only thing that changes is the absolute amount of money you can lose. Theoretically speaking, the risk of such a loss occurring remains exactly the same because it is independent of your performance. This brings me to (2): the oldest lesson in the book is to "not put all your eggs in the same basket". Just by allocating your account 50/50 between two assets already diminishes the chance of going into debt by a factor of 2. Instead of a 100% loss, you would need to have a 200% loss to actually build up a debt.

In any case, there are many ways to protect against this. Not just 'wisely picking your stocks', but simple hedging instruments like options are built for reasons like this. Options (call and put) can also never go below zero (as long as you don't write them). If I want to go short on a company, I'd buy some puts instead of actually shorting the underlying. Same as you described with the inverse ETFs. You can also do a combination of a derivative and the underlying by applying an imperfect hedge. In that case, you create a specific profit margin for yourself and you are still totally covered in case something goes wrong. For example, you can look into "strangle" and "straddle" option strategies. These speculate on volatility opposed to direction: you literally profit if the stock makes a volatile move in either direction, up or down. The downside of these strategies are the transaction costs and the time-value-of-money that options exhibit. However, knowing these combinations means that you can apply a whole range of different trading strategies too. Instead of trying to predict direction one can focus on forecasting volatility. This is still quite difficult, but some say it's not as difficult as forecasting direction (Google "GARCH model"). Trading the VIX, by the way, is also indirectly making a trade on general market volatility. But like you said, inverse ETF's can therefore also be a good choice in case you want to go short but don't want to risk large unexpected gains in the stock. Only problem I have with ETF's is that they target a specific field. Options really cover most individual companies.

In any case, I enjoy having this discussion :)
Always good practice to keep your risk management game strong :)

You are correct, and in this case I was defining "vulnerability" as exactly what you quoted:

"then the only thing that changes is the absolute amount of money you can lose".

I'm thinking of the situation where we create an algorithm that we "set and forget". If we're lucky enough to have it double, triple, or multiply by tenfold our portfolio value, then as we accrue more and more money we are susceptible to being placed further and further in debt due to an errant bet on a short if we fully leverage our portfolio. I.e. the amount you can be in the hole is directly proportional to how much you intend to short.

This is not really an issue for people that conduct good risk management procedures, but I can easily see the automation of a very good strategy leading to some complacency; those who are not really familiar with shorts coming into the game (like myself: I had never shorted a stock before I got here) might think that shorting poses the same risk as taking a long position. I actually thought that way before I did some of the math on risk/reward behavior of longs vs. shorts. Glad I came upon this conclusion!

Thanks for the discussion points! As always, they're great. :-)


One other thing I wanted to mention was that the implementation of trailing stops and other loss-limiting controls of that type can also lead us into a false sense of complacency. Just because we have second-by-second insight into pricing data doesn't mean our loss-limiters can actually SUCCEED in limiting our losses. This could be due to any number of reasons:

1.) Algorithm crashes
2.) Brokerage disconnects (which I hear doesn't happen often, but it does happen occasionally)
3.) Circuit breaker trips on NYSE

Our loss-limiters are not perfect in the least, and I had to break myself of the common thought process: eh, my algorithm will save me when things go south :-P


Agree with most things you say, but I have to say that stop-losses can be quite a good safety net.

This because stop losses (not stop limit) are often placed at the moment of placing the original trade. Therefore they are active the moment you take a position. Your broker will simply keep track of the stop loss and execute it when necessary. This process eliminates most of the points you mentioned in your last comment.

I'm not a big fan of stop losses myself, but theoretically speaking they are quite a good way to prevent disasters.


Sorry, I guess I was only talking about algorithmic implementations of loss control like trailing stops by monitoring portfolio value, etc. :-)


Update Backtest


The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.


This discussion is closed