How should I use stop losses?

4th Sep '19 by WarrantStar

40 comments 2121 reads

Hello everyone

I hope that you are not looking for advice on how to use stop losses! Obviously that is because I am looking for advice!

I have been investing successfully in shares for 30+ years and had never considered stop losses until now. But recently I read an article in Investors Chronicle and then I bought "The art of execution" both of which strongly advocate the use of stop losses as a way of minimising losses. I use Stockopedia's great new Stop loss alerts, which are of the trailing variety.

I would really like to hear from subscribers that use stop losses. My questions are:

1) Are your stop losses fully automatic or do you have to instruct a trade? Why?

2) Do you treat the alerts as discretionary or mandatory? Why?

3) If discretionary, then do you sell immediately or do you wait to see if the limit is still broken at the end of the day? Why?

4) Do you make your decision to sell purely based on on the stop loss, or do you only sell if other factors such as a profit warning or a bad news release arrive? Why?

5) How successful or unsuccessful do you consider your use of stop losses to be?

I would really appreciate your advice on this because it looks like it could be a good step forward, but only if it is done in the right way.

Thanks.

Unlock with Email

Unlock with Google

Already have an account?
Login here

Howard Adams

5th Sep '19

Hi WarrantStar

I am a recent experimenter with stop losses and have found a marked improvement in my recent portfolio performance (it is quite a short period of experimentation so far, but very encouraging).

Below are a set of trading rules I am experimenting with. Using only small amounts of capital until my stats demonstrate they have validity.

These rules only focus on the actions of buying, scaling up and stop losses (a combination of quant rules and process behaviours). I think it important to align stop loss rules with buying rules as in combination they result in outcomes.

The back-drop to them is that they are enacted within the Wyckoff structures Jack Corsellis explains (brilliantly I must say) in his Stocko thread and accompanying vids https://www.stockopedia.com/content/technical-analysis-comment-stocks-for-video-review-489681/

Thus, some would label these momentum trading I believe.

• First buy £600 (even if it’s a re-entry). Buy in round numbers of units i.e. 10 or 15 (close to £600) (NB: adjusting this up/down means I am raising/lowering my risk/reward ratio)
• Scaling up second buy - same units as first buy (ideally once first buy positive %pct gain, but not always)
• Scaling up third buy – holding in positive %gain. Either same units as first buy, or double units of first buy if %pct gain has been strong and/or price is rising fast
• Scaling up fourth buy- same units as third buy

• All holdings excluding Unit Trusts have stops
• Holding < £1k (& is negative %pct gain) -6% stop
• Holding > £1k < £2k & is negative %pct gain -4% stop
• Holding > £1.2k and is close to positive %pct gain (but not quite) -3% stop
• Holding is positive %pct gain (irrespective how large holding is) -2% stop

• Recalculate stops midday & close of business (just one might be sufficient)
• Re-set to new calculated stop if new stop value > current stop value
• Re-set to higher stop if a holding has a noticeable gain in a trading session e.g. around +2% or more
• Behaviours – I MUST NOT trade at market opening no matter what positive RNS or other trigger seems to be present (risk/reward is often a failed trade)
•
• Scaling down TBC ****

Early indications suggest that although the numbers of my positive trades versus my negative trades are at similar levels. Importantly, the £values of my positive trades exceed the £values of losses by 5.7 times. That suggests my exits are fast and clean whilst my scaling up and then capturing gains are working reasonably effectively.

If these stats continue to work/improve then I will increase my initial entry value to say £700, thus increasing my investing risk/reward by 16.7%.

I hope this suggests ideas.

Regards
Howard

William K

Interesting questions WarrantStar. I used to use stop losses on all my trades. I'm not convinced they saved me any money. Today I virtually never use them - except to lock in profits. The reason being that today I favour less volatile assets and invest for the long term. In other words, if I think something needs a stop loss I don't buy it. Having said that, stop losses are probably essential for speculators who. are riding a trend in which case the 20% trailing stop loss is statistically the best bet as it avoids getting stopped out of winning positions. It's worth noting that the great investors like Peter Lynch never used stop losses. Whilst aggressive traders like Martin Zweig sometimes set their stops at 3%. It all depends on the style of investing.

WarrantStar

For further clarification, my general approach is to buy and hold quality growth companies with a minimum market capitalisation of £50m.

Wilkonz. Thank you for your reply. Did your stop losses require you to manually instruct a trade? Mine do require this and are set at 20%.

Howard. Thank you for such a detailed reply. You have certainly given me a lot to think about! Do your stop losses require you to manually instruct a trade? If so , do you ruthlessly follow your rules? Please can you give a rough estimate of your average holding period?

HumourMe

For some strategies stop losses make sense, others less so. I would not use them for small cap,illiquid shares; I don't buy these. For a portfolio of value shares, if I had the skill to put together, the argument for using them is mixed (re-balancing periodically may be better). For a factor based strategy where you are betting on probable outcomes of a basket of shares, again they may be less relevant. The trouble with any rule of thumb is it is valid only in a specific context.

They are a form of risk management and position sizing. Stop losses may be the answer. If using start with "plan to risk no more than x%" before you admit you were wrong. 2% is frequently quoted.

Using this guideline if the stop requires 10% then you can risk up to 20% of your principal, falling to 10% of principal at a stop of 20%. However, this really depends on personal risk tolerance and assumes a positive expectancy from your system. Personally I'm currently using levels far below this.

So my context is quality/momentum/breakout/trend follower aspirant.

Entry stop at 7-10% below 'breakout' or 2-3% below a 'low' or trendline for a swing entry.

Rational: I expect it to go up. If it doesn't I'm wrong. It is easier to re-evaluate when not committed.
Trailing generally 15%-20% below high, or last reaction low (support), whatever is higher. Will raise stop to 5% below a range bound share (Supply/demand in balance - which will win?).

Change to trailing stops occurs mostly on Saturday (so I'm not constantly evaluating/second guessing myself).

15%-20% determined by volatility. Designed to try to keep me in steady gainers while allowing some correction. At 10-15% gain, depending on price action (generally if a share has stalled), I'll initiate a break even stop (in reality it is when I switch from the entry stop to the trailing stop). I've tried values below this and found them detrimental.

If you believe something should keep on going up I've learn't to give it room to breathe. I'm currently getting my head around volume which modifies everything.

Stops are just executed.

Rational: Avoid emotional entanglement. However, I do use alerts to flag key buy price points for shares in downtrends. I generally ignore these, reset them lower and benefit from doing nothing.

Not discretionary. If a stop is broken I'm on the wrong side of supply/demand (or I've set the stop too tight). Time to re-evaluate. Mine are automated. I understand the arguments for alerts but on balance I can't react that fast and I've 'saved' more than I've lost this way.

I will try to sell or tighten stop at euphoria, when price goes parabolic. This is hard to master/judge.

A profit warning will generally hit my stop and the first I know is the sell notification. On a few occasions my stop has been hit in the weeks before any profit warning - chance or just price action is difficult to know.

Mixed. I was initially much too keen to get to break-even and got stopped out on shares that went on to perform. I changed broker 18 months ago after liquidating everything and using stops more extensively made getting any traction more difficult. However, this year my draw down has been a maximum of 4% but the context is that most of the time I've been in cash and I'm at break even. I've been reading trading books and as Howard mentioned learning about Wyckoff and volume.

Snoo

If your investment style is fundamental and your outlook is long-term, then IMO stop losses are not really that appropriate. Most quality growth companies in the past have endured large draw-downs that would have taken out trailing stop losses at least. Share price growth is rarely linear.

A stop loss does not offer any protection against profit warnings as most likely the price will gap down past it.
And general volatility of small-caps means that unless your stop loss is large you may be stopped out on the general drift of the markets. Something like Air Partner (which is in the Buffetology fund) has quite a wide share price even intra-day. In August the share price has gyrated from a low of 68p to a high of 85p on no news.

Stop losses have their uses, that is for certain but I feel the varying amount of opinion leads to confusion, because the context in which they are applied is more important than the stop loss itself.

So a technical-led trader trading without a stop loss would be quite bad, but then a fundamental-led trader trading with a stop loss is also bad (I speak from experience).

Not trading with a stop loss is not necessarily a free invitation to HODL everything regardless, just that on price declines you have to make a manual call. Which means to me for any holding you have to have an idea of what you think the intrinsic value of the share is, and how the share price stacks up against that. That is the main factor, not what the percent gain/loss on holding is. For a technical-based investor, those factors are reversed.

Yes. I set a price for selling at whatever seemed appropriate - usually 15-25% below the purchase price. (My broker is Hargreaves Lansdown and they don't do trailing stop losses as far as I can tell.)

Hi WarrenStar

Yes I manually set stop losses in Hargreaves (its clunky, and quite time consuming).

Yes I ruthlessly follow the rules (in the short few weeks I've figured them out).

For example Future (LON:FUTR) was up 15% on open this morning. I was already in profit. I therefore immediately increased my stop (to -2% of sell price because I was already in profit on my holding).

Not long after, I got stopped out. GGgrrrrr you might think.

But often a rapid riser stock falls back quite far within a short time (often within a day). So in this case my quick lifting of my stop locked in a further 3% gain which I certainly would not have normally captured.

If I do want to be in Future (LON:FUTR) I will now only enter after a pull-back and then see a reversal upwards occurring (this is Wyckoff disciplines - Technical analysis + fundamentals).

'Rough estimate of my holding period' .... I do not have that empirically yet as the stop loss experiment is quite recent.

Subjectively I have some of the longer-uptrenders for several months. For the rapid risers coming out of a pull back I might only hold for a few days if the reversal fails to be sustained, OR my gains have become positive and then a > -2% fallback occurs. I am satisfied as I have then captured profits.

But, very importantly, irrespective of my holding periods, I biais my stock selections towards trading repeatedly stocks I am familiar with (as I have a sens of the stock and can thus exploit changes of character in the stock price & volume patterns, the recent moves in Ab Dynamics (LON:ABDP) being a specific example). So I might move in and out of the same stock over several years - Games Workshop (LON:GAW), Fevertree Drinks (LON:FEVR), London Stock Exchange (LON:LSE), Ab Dynamics (LON:ABDP) to list but a few.

I'm not sure this truly answers your question as my rules are driven by .... ' a 'best allocation of capital into investments with greatest probability of providing a positive outcome in the shortest time possible, thus reducing my risk exposure to the down turns of a stock or the market'.

Regards
Howard

Rusty2

The problem with stop losses is most reports focus on losses not the profits that got away. Ab Dynamics (LON:ABDP) is a good example, I first bought this at about 178p, got stopped out at 158p, now is is 2800p. IG Design (LON:IGR) also went up 2x soon after I got stopped out. For me stop losses have not worked. Also you can sell you don't need a stop loss. Mark Minervini uses stop losses but he also sells many well before the stop loss is reached. Auto stop losses can take you out of shares do to tree shakes or just thin trading and volatile price movements which is common in UK smaller shares. You also have to consider trading costs especially if you holdings are small, ie around £1,000, broker costs in the UK are high and there is also stamp duty. You can easily lose 5% just buying and selling a share.

Nick Ray

The trouble with stop-losses is that they are also "keep losses" since they lock in the loss. There is already a stop-loss at -100% anyway whereas the upside is potentially much greater than +100%.

So an alternative way to approach this is to "keep profits" by top-slicing whenever the price rises by some amount (maybe +25% or +50%). The realised profit can then be put into a different stock to help diversify the portfolio. You have not exited your successful stock but you have banked some of its gains so if it falls back it will lose less than when it was on its way up. The trick is to top-slice and not sell out completely with successful stocks.

[If you are "volatility harvesting" then the flip side of this rule is to top-up when the price falls by a certain amount. Personally I find this does not work too well with stocks (although it might make sense with collective investments where a period of under performance is more reflective of a whole sector or market rather than an individual company.) So rather than top-up I would wait until the next rebalancing period where it will be sold if it is not performing as well as the rest of the portfolio.]

Part of the aim is to try to avoid building in a negative bias to portfolio performance in the strategy. Selling when a stock is falling tends to create a negative bias from the volatility of the stocks. Selling when a stock is rising tends to create a positive bias from the volatility. (Also it is easier to sell into a rising price than a falling price!)

Gromley

Hi WarrantStar,

With that clarification in mind I would suggest that some of the thoughts given so far, relating to shorter term trading approaches are probably not directly applicable and in any case I don't think there is a definitive "right" answer.

FWIW, I have modelled the use of stop losses on Ed's NAPS portfolios and on similar portfolios that I have constructed. These portfolios may have some similarities with yours.

The bottom line from this is that I have never been able to find a level of stop loss that actually "works" . By "works" I mean to produce a portfolio performance that is superior to sitting on your hands.

I have though generally found that stop losses in the region of 15-25% perform the least badly (generally lowering portfolio performance by somewhere around 5% pa - plus extra dealing charges). If the stop losses given you the peace of mind to stick to your guns that level of adverse impact might be a price worth paying.

One thing that has been suggested is to set stop losses at a level related to the volatility of each individual stock (eg it would be dangerous to set a tight stop loss on a stock that routinely wobbles about a lot.) That is something I do intend to look into further, but I need to find a meaningful measure of volatility - I don't think the measure used here on Stockopedia does the job.

For example Walker Greenbank (LON:WGB) , where I am running a "when to sell" which includes a stop loss, is described as "speculative" putting it in the 40% most volatile stocks. However when I look at the history most of the bigger moves that probably give rise to this rating are due to actual news, on which you probably would want to act. Absent news and the WGB share prices seems to travel in relatively straight lines. (The 10% stop loss I have set out at the start of the experiment is probably still too tight, but given that I am using it as an "advisory" for the sake of the experiment I want to put myself in the position where I have to make decisions that I can perhaps learn from).

Any suggestions from anyone on an appropriate measure of volatility that differentiates between "moves" and "wobbles" would be gratefully received.

hi again WarrenStar

Adding to my previous post #9.

You might want to take a look at my post #34 on this thread.

As it goes some way to illustrate how introducing stop losses into my tool koit does seem to be assisting my outcomes.
https://www.stockopedia.com/content/small-cap-value-report-thur-5-sep-2019-boo-tough-markets-mpac-alt-bks-510226/?comment=35#35

Regards
Howard

unwise2

Stop losses are a complicated subject. If you are a 100% fundamentals investor and also buy shares in a downtrend (value) stop losses are unlikely to help your performance.

If you include technical factors in your buying decisions then stop losses will be beneficial. Stop losses are best used by swing traders or buying breakouts out of bases.

jonesj

1 We should not expect a single investing or trading methodology to consistently outperform over the short and long term. If this were starting to happen, then the majority would adopt the methodology and it would become the average.
I don't think adopting simple rules are sufficient to outperform. Combining some simple rules with very hard work and good judgement might outperform.

2 Many great investors do not operate stop losses at all. Other traders do and are still successful. There is no one right answer.
I have never used stop losses on share purchases. However, I have used them on shorts, since the liability would be infinite if I got it wrong.

Investor & Trader

Stops are a must IMHO.

They limit losses and maximise return on risk, however the stop loss should be used within the structure of the price action too, not a set % or abituary figure.

A naked 10k position with a 1k increase is only a 10% return on risked capital.
whereas a 10% stop loss on the same 10k position would give a 100% return on risk (or ROCE)
Also freeing up capital to risk on other positions and diversifying the portfolio.
See my fantasy fund trades (open & close) to see examples of the weekly consolidation breakout system I use with stop loses within the structure of price action.
Due to my risk management I'm able to use 2:1 margin which doubles the return (positive or negative).

Stops are a must IMHO.
They limit losses and maximise return on risk, however the stop loss should be used within the structure of the price action too, not a set % or abituary figure.

A naked 10k position with a 1k increase is only a 10% return on risked capital.
whereas a 10% stop loss on the same 10k position would give a 100% return on risk (or ROCE)
Also freeing up capital to risk on other positions and diversifying the portfolio.

Whoa !

Whilst I am not by any means in the "never use leverage" camp, there is a dangerous fallacy in there if you're not careful.

In your case of the "naked" 10k position (ie 10k from a normal brokerage account without a stop loss) there is a high risk that you could lose 10% of your capital. Perhaps a moderate risk that you might lose 20% & perhaps a low risk of losing 50% and 'very low' risk of losing 100%.

However, if you set a 10% stop loss and therefore consider your risked capital to be only 1k - there is now a high risk that you lose 100% of your risked capital.

Certainly you could use this leverage in an attempt to diversify away some of your risk - 10 individual positions with a high risk of total loss on each one may present a relatively low risk of total loss on all 10. - However this all goes out of the window if the market as a whole takes a sharp downturn.

Add to this that unless you have got 'guaranteed' stop loss with a SpreadBet or other instrument then the stop loss does not in fact limit your potential losses to 10% , and even with a GSL the cost will be more like 12%+ (dependent on the SB co.)

-------------------

I know I am somewhat extending your argument and that you go on to say that you actually use 2:1 leverage rather than 10:1, but I did want to avoid people being unduly bullish; and even at 2:1 don't forget that the market can a does halve from time to time.

Anyone who had stop losses when the Brexit vote happened may remember what a nightmare they were. With my broker I could not cancel as they were pending, you had no control over what sell price you would get and did not know if your shares would be sold or not and when and for how much.

pka

Hi WarrantStar,

This video doesn't answer your questions, but you might find it amusing:

https://www.youtube.com/watch?v=LiE1VgWdcQM&list=RDLiE1VgWdcQM&start_radio=1&t=2

Everything is relative and has to be put into perspective, but you are right to express concern on the face of my approach summary.

For me using a 2:1 leverage, and lets say an initial 10% stop loss on a 10k account:

2:1 ratio = leveraging to a 20k account.
4 x 5 k positions
4 x maximum £500 losses
= A 20% loss on cash/account should all 4 stocks hit the stop loss.

However, my strike rate is 58% and my risk reward ratio 4 to 1 (over a large sample), my average loss is actually 6% when taking into account raising/trailing stops.
The volatility of my stock selections is low and the stops are placed in strong support, coupled with strong fundamentals and a momentum catalyst etc.

Without such stats I agree the approach could be seen as risky, but with the right approach the reward outweighs the risk which is always the trade off. In this system its a play on probabilities and playing enough that the edge materializes.

I generally hold 8 or more positions so I'm better diversified than the example above, should all positions hit the maximum stop loss and I lose 20% of account, this kicks me out of the market until the market improves. Therefore a large unrecoverable drawdown is unlikely.

We have seen a volatile 2019 and you will note the volatility of my fund is low in comparison.

I also hedge the portfolio with 20% gold which counteracts volatility (usually...)

How should I use stop losses?

Unlock the rest of this article with a 14 day trial

About the Author

WarrantStar

40 comments

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

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