Is a stop loss good or bad? What is the best stop-loss strategy? Are there alternatives to a stop-loss? How many different stop loss strategies are there? When you read books and blogs about trading, one mantra is repeated: you must always have a stop-loss. Why this obsession with stop-loss? Perhaps there are stop loss options?
Our backtests reveal that a stop-loss mostly makes your strategies perform worse because stop-losses are a trade-off between ruin/survival and profits. We also explain why we don’t include stop-loss levels in our monthly Trading Edges. At the end of the article, we give you several alternatives and stop loss strategies.
To have a fruitful discussion about the pros and cons of stop-losses, we need to understand and define a stop-loss:
(Before we go on we’d like to mention that we have a backtesting course that covers all aspects of how to backtest.)
What is a stop-loss? (Stop loss meaning)
Risk management is a very important factor when determining whether a trading strategy is profitable or not. Stop-loss is a much-used technique by traders.
We define a stop loss as the price where you want to realize a loss in order to limit your losses. We can say it’s a defined price level – mostly set in advance of your purchase of the shares or the asset.
A stop-loss can be both a “physical” order in the market or a mental one. For example, if you buy Microsoft at 100 you might want to enter a stop-loss order at 90 immediately after purchase. If the price goes down to 90, your shares are sold automatically at the market or as a limit order.
Put short, stop-loss is a limit where one simply liquidates the position at a certain level. Before entering a position, the max loss is determined, and if the trading vehicle reaches that limit, the position is closed. This is to limit the potential loss (because we never know how far the position can go against us).
A trailing stop might take you out with a gain and thus might not be a stop-loss per definition. This is because a trailing stop “trails” the price and increases as the price goes up (if you are long). If you bought at 100 and want to use a 5% trailing stop, the stop is increased to 104.5 if the price goes up to 110 without hitting the trailing stop.
Usually a trailing stop is never lowered, only increased.
Don’t use a stop loss – do this instead (video)
The video below gives you an example of how you can improve returns and reduce volatility without using a stop loss (we have plenty of videos on our YouTube Channel):
Where to set stop-loss/What kind of stop-loss strategy
We will briefly mention some different methods in setting stop losses.
A predetermined stop loss strategy: percentage or fixed price
This can be either a fixed dollar amount or a fixed percentage amount.
A percentage level, for example 5%, can be placed below your entry price. Another option is to use a fixed dollar amount. If you only want to risk 1 000 for the trade, then you calculate the stop price that equals this amount.
Does such a stop-loss strategy make sense? We believe it only makes sense if you have a predetermined risk/reward calculation. If not, we can’t see much benefit in just setting a random percentage as a stop-loss.
Volatility based stop-loss strategy
This is, we believe, a more rational approach than the one above. Financial assets change volatility depending on the market, and you want to make sure that you don’t get stopped out from the noise in the market.
Why use stop-loss – the pros:
Stop-loss is for survival – not against losing money
If you lose your capital, you can’t trade, obviously.
Likewise, if you lose 50% of your capital, you need to gain 100% to recover to the previous peak of your equity.
A stop loss is foremost about avoiding the risk of ruin and staying in the game. This is by far the best advantage with stop loss.
Stop-loss is “impersonal”
If you put in a stop-loss order after you have purchased, you can simply “forget” the position.
Traders are prone to behavioral mistakes and when you put in an order early you are done and can move on to the next trade.
Trailing stop locks in profits
A trailing stop can both be a stop loss and a stop profit, depending on the movement prior to execution.
One of the worst mantras in the stock market is “you don’t go broke by taking profits”. It’s hard to argue about that, but at the same time, you will most likely never get very wealthy either. Hence, make sure you give your trades room to move in your desired direction.
But unfortunately, this doesn’t mean stop losses are automatically good – quite the contrary:
Why not use a stop loss strategy – the cons:
A stop-loss can’t be fixed or predetermined
Many traders put a stop-loss at for example 5%. This doesn’t mean you are risking a max of 5%. Frequently the loss gets much bigger than that.
When you place a stop order you are not guaranteed it will be executed at that price. If it gaps down more than the pre-determined stop-loss, the order will be executed below the 5% level.
This happens when there is adverse news. You simply have no opportunity to get out at a 5% stop loss.
Further down in the article we provide you with an example of stop-loss levels. If we used a 1% stop level, it means that 67 trades of 164 are stopped out. But 14 of the trades are bigger than 1% and executed on the opening print if the open was lower than 1% from the entry price. Those 14 trades were executed at an average price of 1.95% and the median was minus 1.85%.
Stop-loss strategy has backward logic
If you are trading the stock market, which is highly mean reversive, an ETF or stock gets “cheaper” the more it falls. The statistics are clear: when the price goes down, it means the risk premium rises. This has “always” meant higher returns in the future.
The stock market has over the last 30 years shown strong evidence of mean reversion and thus any drop has been followed by better odds of increased returns in the nearest future.
Most argue the case for stop loss to prevent large and uncontrollable losses. If you’re not using one it’s just a question of time before a large loss wipes out your position – even your account.
But most uncontrollable losses happen overnight when an asset gaps against you, so-called overnight strategies. You can’t limit your losses because of this, but perhaps, on the contrary, limiting your profit potential if you sell or cover big gaps.
To avoid this you need to trade lower size.
Stops can make you “bleed to death”
If you set the stops too tight you risk facing many small losses in a row. We are all prone to behavioral mistakes, and you might give up the strategy too early because of this.
We strongly recommend using backtesting in order to find strategies to trade, but they have a weakness in curve fitting. This also involves stop-loss.
For example, you might use types of stop losses that work in certain environments, and thus force you out at the wrong moments when markets change.
Why we don’t include a stop loss strategy in our Trading Edges
We’ll try to answer in short why we don’t include stop-loss levels in our monthly Trading Edges:
Stop losses strategies are “personal”
Stop-loss levels are all about risk management. We are not investment advisors and all our Trading Edges are not regarded as investment advice. We have no knowledge of your account, your wealth, or your risk level, and thus you need to calculate your own tolerance for risk.
Secondly, stop-loss needs to be evaluated based on the overall risk of your portfolio or your portfolio of strategies. If you are trading many strategies, it might not make sense to use a stop at all.
Further down we argue what you can substitute a stop-loss with.
An example of a stop-loss strategy
A few months back we published a trading strategy in the ETF with the ticker code XLP as part of our monthly Trading Edges (April 2021).
Let’s test a few stop-loss strategies and see how the strategy performs. Our Trading Edges are never published with stop-loss levels, and hopefully, you’ll understand why after reading this article.
The strategy performs like this without stop-loss:
The average gain per trade is 0.75% over 141 trades. Max drawdown is 10%. All in all, we believe this is a pretty solid strategy with good risk-adjusted returns: the profit factor is 3.5 and the Sharpe Ratio is 2.7.
Let’s look at some stop loss strategies:
A fixed stop-loss strategy
Let’s implement a stop loss of 1.5% that is either executed intraday at the exact level or on the open if it gaps adversely. What happens to the strategy’s performance?
The average per gain drops to 0.37% but the max drawdown is hardly reduced at 9%.
Why is that? One reason is that with a stop loss you increase the number of losers: the number of max consecutive losers goes up from 3 to 5. You simply get more losers because many trades show a loss before they improve.
A stop-loss acts as insurance and it costs money to insure!
A volatility stop-loss strategy
Let’s test a volatility stop of 2 times the 20-day ATRs (we exit if the price drops twice the 20-day average at the time of the purchase):
The average gain per trade is 0.55% and the max drawdown is 8%. The smaller drawdown comes at the cost of lower overall profitability. The results improve compared to the fixed stop-loss simply because the volatility stop has more wiggle room for the position to work your way.
All in all, a stop-loss lowers the expected average gain per trade.
However, we believe XLP is one of the ETFs that performs the worst with a stop loss, so you better test yourself in other instruments.
Amibroker code for stop-losses:
Amibroker is a very powerful tool despite its cheap price. It works both for backtesting and live trading, especially with Interactive Brokers. How you can learn to code, do backtests, and live trading is described in our Amibroker course.
Quantified strategies and stop-losses strategies
This website is all about quantifying trading strategies, but we are not alone in being skeptical about stop losses. Larry Connors and Curtis Faith have tested stop-loss as well:
Larry Connors and stop-loss strategy
Larry Connors is famous for his focus on mean reversion trading strategies in stocks and ETFs. In his book called Short Term Strategies That Work he even dedicated a full chapter to stop-losses which discusses how Larry’s research team ran hundreds of tests to try and find optimal stop levels.
He arrived at the simple conclusion that an optimal stop-loss doesn’t exist. Moreover, stop-losses tend to hurt system performance much more frequently than the opposite. They came to the conclusion that for the trades they were looking at, the optimal stop was consistently none at all. In every case, they found that instituting stops hurt system performance.
How about trend-following and stop-loss? Let’s look at Curtis Faith, a Turtle Trader:
Curtis Faith and stop-loss strategy
Curtis Faith has written a very good book about trading, The Way Of The Turtle, one of the few books we recommend.
Curtis Faith arrived at the same conclusion as Larry Connors. On page 145 he states the following:
Note that the zero case, which means no stop at all, has the best MAR ratio numbers. In fact, the test with no stops is better for all the metrics: CAGR%, MAR ratio, Sharpe ratio, drawdown, and length of drawdown – every single metric. The same things holds true for a test of the Triple Moving Average system: Every single measure was worse with any stops. The same test of stops applied to the Donchian Trend with time-based exit system yields similar results except that for very large stops of 10 ATR or more, the results are about the same as those for a test with no stops. This certainly goes against the common belief that one must always have a stop.
(For a better understanding of risk metrics, please read our article called trading strategy and system performance metrics.)
DE Shaw and stop-loss strategies (David Shaw)
David Shaw, the king of quant, was (is) a successful hedge fund manager. Here is what he said about stop-losses:
We believe that stop-losses are generally ineffective as a risk management tool and thus almost never use them, relying on other solutions instead….Stop-loss protocols are simple, and simplicity is a virtue. But the simplicity comes at a price because stop-losses require trading on the basis of stale or incomplete information. …Stop-losses are set in advance, which accounts for the staleness, and they are a function of only one form of information – price. This means a lot of other crucial market dynamics are ignored…There are other stop-loss implementations as well , especially withtheir most extrem form (used by retail investors), whereby a trader leaves the stop order with a trading counterparty. As an execution tactic, this is unlikely to produce good results.
Only huge stop-loss limits work
Our strategies tend to work with only very huge stop-loss levels, like for example 10%. When a stop level is so big, then in reality there is no stop loss but you avoid getting stopped out too early.
Instead, we recommend alternatives to stop-loss:
Alternatives to stop-loss – the best stop loss strategy (stop loss options)
As you might have understood, we are no fans of stop-losses and we prefer alternatives to stop-losses.
We have previously indirectly written about alternatives to stop-losses: trade many asset classes, trade different time frames, employ different types of strategies, vary position sizes, and look for uncorrelated strategies.
- Which time frame is best in trading? What time frame should you trade?
- Is this the Holy Grail of trading?
- Why build a portfolio of quantified strategies
- What does correlation mean in trading?
The whole idea is to offset losses in one strategy or asset with gains in another. This is a substitute for a stop-loss.
Trade many asset classes:
Why limit yourself to for example just stocks? We are trading both stocks and stock indices, but we also trade commodities like gold, gasoline, Treasuries, Dax, silver, oil, and heating oil. We even trade some currency pairs.
The reason is to have uncorrelated returns. If we lose in ES-mini, we might gain in USD/EUR. If we win in gold, we might suffer in stocks.
We hope you get the idea. The main idea is to smooth returns. We have covered this topic in many articles.
- What does correlation mean in trading? (Trading strategies and correlations)
- Uncorrelated assets and strategies – benefits and advantages (examples and backtests)
- Does your trading strategy complement your portfolio of strategies?
- Why build a portfolio of quantified strategies (including two strategies)
Trade different time frames
There is no best or worst time frame, and we look for strategies that hold for decades and as little as 1-4 hours. Why limit yourself? One dollar made in day trading is the same as one dollar made in swing trading.
The return you get on a day trade can be completely uncorrelated to a swing trade of two weeks.
Trade both long and short
We are mainly trading from the long side, at least in stocks, but we love adding short strategies, which are very useful to smooth returns. The downside is that shorts are more prone to erratic moves, and they are also very hard to find. Short selling is difficult but only a few strategies can make a huge difference for risk mitigation.
We like to use time-based “stops”. Why? Because trading edges tend to “fade” as time goes on. The best edges are either very short term or long term (we believe the long term edge is getting better currently because of all the noise in the markets). A time bases stop is if you (for example):
- Exit after N-minutes, N-hours, N-days, etc.
- Exit at a certain time, for example, close, 1500, 1400, etc.
If you test strategies, please try at first without using stop-loss, but only an exit signal. You’ll be surprised to see how many edges simply “fade away” after a certain number of days. Likewise, when to exit a trade is important.
Employ different types of strategies
Mean reversion and trend following strategies exhibit different characteristics. Mean reversion has many winners and occasionally big losers, while trend following has many more losers but the “random” big winner that offsets all the small losers.
Employing both types of strategies is something we recommend. You can read more here about the two different types of strategies:
- What are negatively skewed trading strategies?
- Trend following strategies and systems explained
- How to create a mean reversion trading strategy
Vary position sizes
Instead of using stop-loss you might consider using a lower position size.
We have one general rule in our trading: always trade a lower size than you would like. This way you are a lot less prone to behavioral mistakes.
Remember that the size of your positions will have an impact on the overall risk level of your trading account. With bigger positions, eventual losses will become bigger as well. Make sure that one position doesn’t make up a too big share of your total account balance!
Stop-loss strategy and mean reversion
One certain type of strategy rarely works with stop losses: mean reversion.
Mean reversion means that any swings in one direction are likely to “revert to the mean” and either reverse or lose momentum. A drop in the price is mostly followed by a snap in the other direction.
If you go long in a mean reversion strategy you do so in the expectation it will turn around. If it drops further and your stop gets hit, you are selling when the edge has become stronger. This is certainly not optimal, as you can imagine!
So, what to do about your stop-loss strategy?
We end the article by giving you some food for thought.
We believe a stop-loss strategy can’t be judged solely on its own merit but has to be looked upon in the bigger picture. By the bigger picture, we mean your overall portfolio. If it’s a 10% loss for the strategy, what is the damage overall? Are your strategies likely to be correlated and create damage at the same time? Are you trading many markets, time frames, and both long and short?
Keep this in mind:
- Can you live with the drawdown of the strategy?
- Test by using a time exit and measure how the strategy “fades away” with time. Check the drawdown.
- Test your strategy and see how it performs with and without a stop-loss strategy.
- Does a stop-loss improve the drawdown of the strategy? If so, why, is it outliers? If no, why not?
- Keep in mind that the worst drawdown is highly likely yet to come.
- How will the strategy’s drawdown affect your overall portfolio?
Why stop-loss doesn’t work
Do stop-losses really work? We are “brainwashed” to believe that you should always have a stop in place. But, perhaps surprisingly, stop-losses normally just make a strategy perform poorly, as you might have learned if you have read so far. Let’s look at some arguments for why stop loss doesn’t work.
All in all, a simple stop-loss strategy with an arbitrary stop loss level is unlikely to work, especially if you leave the stop loss order open at your broker. Plenty of traders are waiting for typical stop loss levels to get hit so they can “wipe out” the retail traders and take them to the cleaners.
Many traders swear to stop losses by default, but to us, it doesn’t make sense because:
- If you buy something at 100, why should you just put in an arbitrary stop-loss level? If you bought something at 100, presumably it’s because of value or some quantified reasoning. If you use stop-loss it needs to make sense where to put your stop-loss level.
- If you trade stock ETFs/futures, it means the risk premium in the market rises when the index falls. This means more likely higher returns in the future. Why sell at a lower price than you bought when the expected future return increases?
- We have yet to see many strategies that perform better with a stop-loss. However, we’re mainly mean-reversion traders, not trend followers.
A stop loss strategy makes a strategy worse (example and backtest)
Let’s use an example of a previously published strategy: Nasdaq (QQQ) Mean Reversion
Without any stop-loss, the strategy returns 1.7% per trade, 12.34% annually, and has a maximum drawdown of 19.51%. The drawdown is big given you allocate 100% of your equity to this strategy, but as part of for example a portfolio of 10 strategies the drawdown is much more reasonable assuming the strategies are somewhat uncorrelated.
The table below shows the result for the same strategy using different levels of stops (exit is done at close if the stop level is hit):
|Stop level||% per trade||Annual returns||Max drawdown|
At a 5% stop loss it seems the strategy does not deteriorate so much. We forgot to include the profit factor in the table, but that one also deteriorated with stop losses.
Stop loss strategy- pros and cons – ending remarks
Stop-losses might serve other purposes, for example, your faith in trading a system, but we suggest you look into the other alternatives of stop losses we have mentioned in this article. Always keep in the back of your head that trading is about survival and avoiding ruin: Never cross a river that is on average four feet deep. This is sure to make you drown.
At the end of the day, you need to spend considerable time setting up a trading plan. This includes testing for stop-loss strategies and how to smooth your returns. Such a plan might not include stop-losses at all, but rather diversification. After all, there are many stop loss options.
A stop-loss is there to avoid ruin, foremost, not to make you rich. A stop-loss strategy might preserve your capital, but it might also lead to many small losses that ultimately eat your profits. You have to find the balance. A stop-loss strategy comes with its pros and cons.
– How is a stop-loss defined, and can it be either a physical or mental order?
A stop-loss is defined as the price where one intends to realize a loss to limit losses. It can be both a physical order in the market or a mental one, set in advance of purchasing shares or an asset.
– What is a trailing stop, and how does it differ from a traditional stop-loss?
A trailing stop increases as the price goes up (for long positions) and can result in gains. It differs from a traditional stop-loss, which is a fixed price level set to limit losses.
– What are some alternatives to using a stop-loss in trading?
Alternatives include trading multiple asset classes, employing different time frames and strategies, varying position sizes, and using time-based exits. These approaches aim to offset losses in one area with gains in another, reducing reliance on stop-losses.