Long-Short Strategy: A 3 month experiment

Saturday, Oct 05 2019 by

Over the last 3 months I have conducted an experiment to mimic the long-short strategy of the early hedge funds. I selected 15 long shares and 15 short shares and invested as close as possible to £2000 in each company on a spreadsheet. The long and short shares were matched by sectors.

Criteria for going long:

Price > 200MA

Low debt (<1x net profit)

Stock rank >80

Criteria for going short:

Price <200MA

High debt (>3 x net profit)

Stock rank <50


My bench mark was the FTSE All Share Index. In the course of this experiment it fell from 4093 to 3933 (a drop of 4%).


The long investments now have a mean value of £2,010 (range £1,353-£2,944) whilst the short investments now have a mean value of £1,582 (range £908-£2,651).

The long investments have beaten the benchmark by 4.6%.

The short investments have fallen 16% below the benchmark.


The probability that these results could have arisen by chance (unpaired Student-T test) is 1.2%.


Whilst this experiment doesn’t actually prove anything at all since the time frame was short and the samples were small, it suggests that predicting price falls is easier than predicting gains, especially in a volatile market overshadowed by Brexit and the Trade Wars. I plan to repeat this work with somewhat more rigorous academic discipline and over a longer period.

Unlock this article instantly by logging into your account

Don’t have an account? Register for free and we’ll get out your way


As per our Terms of Use, Stockopedia is a financial news & data site, discussion forum and content aggregator. Our site should be used for educational & informational purposes only. We do not provide investment advice, recommendations or views as to whether an investment or strategy is suited to the investment needs of a specific individual. You should make your own decisions and seek independent professional advice before doing so. The author may own shares in any companies discussed, all opinions are his/her own & are general/impersonal. Remember: Shares can go down as well as up. Past performance is not a guide to future performance & investors may not get back the amount invested.

Do you like this Post?
17 thumbs up
0 thumbs down
Share this post with friends

14 Posts on this Thread show/hide all

cicerosflipflop 5th Oct 1 of 14

most interesting.

| Link | Share
EdmondJ 6th Oct 2 of 14

Good initiative! Albeit with a caveat how a model portfolio can be one thing, a real portfolio quite another. How is the short side to be implemented?

| Link | Share
wilkonz 6th Oct 3 of 14

In reply to post #519141

Agreed EdmondJ.  The model portfolio implies that all investors are rational and ignores the impact of behavioural finance. Long-Short is not for the risk averse, although in effect it reduces risk by negating market swings.

With regard to implementation, a hedge fund would implement the short side by 'borrowing' shares from a broker, selling them and using the money to buy long shares. (In effect this doubles the profit margin - if one excludes the trading costs and interest.)

A retail investor simply spread bets. There are two issues with spread betting. One is that there may be high spreads on volatile shares and small cap shares. The other is that some spread betting companies won't let you go short on shares that are already heavily shorted.

| Link | Share
Gromley 7th Oct 4 of 14

Interesting experiment Wilkonz.

I note you make no mention of any market cap limits, so I suspect it is possible your short folio contains a number of un-shortable micro caps?

There is actually no direct relationship between market cap and availability to short,, but the tiddlers do tend to be less likely to be available. Running a real money long short folio I find that construction of the short folio is very much steered by which stocks are available.

I am particularly interested to see though how "well" your shortfolio has performed in a three month period when the low SRs overall have not been particularly discriminating.

From the stock ranks performance charts here is the matching performance overall

0-10 -1.7%
10-20 -3.7%
20-30 -9.1%
30-40 -4.9%
40-50 -3.8%
Simple Average 0-10 -4.6%

So your -21% is substantially "ahead" of the game.

This suggests that your extra momentum and debt filters have been highly effective or your portfolio selection has been lucky.

Probably a bit of both, but I am impressed with the performance so will be looking further at those filters. With my systematic approach, I would normally only rarely move in to shorting anything in the SR30+ range, which can therefore be quite limiting.

| Link | Share
wilkonz 7th Oct 5 of 14

In reply to post #519446

Many thanks for your interest Gromley.

With regard to your query about 'un-shortable' micro-caps, most of the. shares came from the FTSE 350, all had a cap value >£250M and all were shortable with spread betters except for two (Aston Martin Lagonda with a cap value >£1 BN, and Purple Bricks with a cap of £300 M - Purple Bricks actually gained 5% against expectation). My main reason for going with Stock Rank was because it was quick and easy like the other two criteria (thank you by the way for your figures, I was intrigued to see that the danger zone was 20-30).

My prime objective was prove or disprove to myself the Efficient Market Hypothesis, and I wasn't expecting anything like the astonishing and highly significant difference between the two groups over such a short time frame. I suspect these may be freak results because of the current unsettled market conditions rather than my selection criteria. Which is why I shall repeat the work with a different set of shares over a longer period.

| Link | Share
wilkonz 7th Oct 6 of 14

Many thanks for your interest Gromley. The write up was indeed a bit sparse, brevity being the soul of wit and everyone's time being precious. With regard to cap values, all the caps were over £250M, and a few came from the FTSE 100, like Glencore (short) and Evraz (long). Most should have been quite shortable.

Thanks for your figures on Stock Ranking - curious that the danger zone seems to be in the 20-30 range.

Yes I think the results may be a bit freakish despite the high degree of statistical probability - perhaps because of the current unsettled market conditions. I shall repeat over a longer period with larger samples.

(PS - I thought I had replied to your post earlier but Stockopedia seems to have mislaid it. So if you get two replies, no need to worry about my short term memory.)

| Link | Share | 1 reply
Gromley 7th Oct 7 of 14

In reply to post #519506

I thought you had responded earlier too, but when I go here to look it had vanished!

I wouldn't pay to much attention to 20-30 being the under-performer, there can be quite a lot of volatility over short periods but over longer periods the deciles normally fall out broadly the correct order.

In fact looking at the data for market cap > £250m the data in this case looks better behaved and had bigger falls than the overall data I showed above. In fact the 0-10 range fell by an impressive 33.8%. However I suspect that is a very small sample (higher market caps tend not to have the very lowest SR) and was probably influenced by a single stock that took a major tumble on 17th September.

From a quick scan of my background data I'm pretty sure it was Sirius Minerals (LON:SXX) which fell by 54% on the day (which suggests to me that there were only 3 stocks in the SR0-10 Mcap >250 bucket).Sirius Minerals (LON:SXX) might well have qualified under your criteria so perhaps you were unlucky not to have picked it!

| Link | Share
Nick Ray 7th Oct 8 of 14

I would want to check the beta of your short portfolio.

If it has fallen 16% for a 4% fall on the index, there is a risk that it has a beta of 4 and if the index rises by 4% it will rise by 16%!

Matching the beta of the long and short portfolio is one of the problems with running long-short portfolios. If you don't do that then you are not market neutral and so you might as well just stay long and accept the contribution of beta to your returns.

On the occasions when I've looked into it I always come away thinking that it might be easier to just short the index and use leverage on the long side, rather than try to run a long and a short portfolio and balance both; especially because the candidates for shorting tend to have higher volatility than the long candidates, which makes things even more fraught.

However, I have never had the nerve to actually try it with real money, so by all means please feel free to ignore my mumblings!

| Link | Share
wilkonz Tue 5:09am 9 of 14

In reply to post #519551

Nick Ray, many thanks. The outcome defies both my understanding of beta and the Efficient Market Hypothesis. Whilst I haven't actually checked the beta of my short portfolio, I very much doubt if it was anything like 4 before I started - although as you rightly point out it was 4 by the time I finished! I append below a list of my short shares - one or two didn't quite fit my criteria but I had to give myself a bit of leeway in order to match the sectors.

AMLAston Martin Lagonda
ORD GBP0.001
DLARDe La Rue plc
Ordinary 44 152/175p
FCHFunding Circle PLC
ORD GBP0.001
GLENGlencore plc
Ordinary Shares USD0.01
INTUIntu Properties plc
Ordinary 50p
PPSProton Power Systems Plc
Prd Gbp0.01
PUREPurecircle Ltd
Ordinary 10p
PRTCPuretech Health plc
Ordinary 1p
PURPPurpleBricks Group plc
Ordinary Shares 1p
RTNRestaurant Group
Ord 28,1/8p
SDRYSuperdry plc
Ord GBP0.05
TEDTed Baker
Ordinary 5p
Ordinary 1p
WMHWilliam Hill plc
Ordinary 10p
Ordinary 10p

| Link | Share | 1 reply
wilkonz Tue 5:18am 10 of 14

In reply to post #519536

Thanks Gromley. There were quite a few prime suspects that didn't make it to my Shorts short list including £SXX and £TCG (Sirius Minerals and Thomas Cook). From memory I don't think either of them were easily shortable when I started my study, either that or there were no straightforward sector matches.

| Link | Share
Nick Ray Tue 2:10pm 11 of 14

In reply to post #519556

I make the beta about 0.8 (on a fairly crude check) so not unreasonable.


That "waterfall" of stocks falling by 50% one after another does seem to sum up what 2019 has felt like sometimes!

| Link | Share
Richard Clifton Tue 3:54pm 12 of 14

I to have studied and put into play long/short hedging speads and intend to to so in the future but not now. Just by chance I caught a line in a video on youtube with Jim Simons the mathematician cryptologist for US intelligence that cracked the market with maths and is head of one of the biggest and most profitable firms.....He said and I quote "Long/Short strategy is ok, but the trouble is with Long/Short hedging is that it requires and needs a lot of volatility to work and be effective" ........And obviously at present with all the global uncertainty across the board, volatilty is a commodity that very much needed by all in the markets just look at the VIX...

| Link | Share
wilkonz Tue 3:58pm 13 of 14

In reply to post #519716

Thanks for the chart - fascinating. Amazing what software can do these days! I'm not sure what to make of the beta aspect - it certainly has limitations in usage, especially over the short term. I guess it's one of the problems inherent with using historically derived data. I shall remain cautious before drawing any conclusions, but the next three months will be interesting. 

| Link | Share
wilkonz Tue 4:46pm 14 of 14

In reply to post #519751

Absolutely Richard. I agree with Jim Simons. The VIX is creeping towards 20 from a low of 15 in April. It was around 24 when the markets started plunging about this time last year. If the VIX continues rising, we could see perfect conditions for Long-Short strategy - which for obvious reasons doesn't work quite so well in a sideways market. Investors are already shunning debt laden companies.

| Link | Share

Please subscribe to submit a comment

Stock Picking Tutorial Centre

Related Content

Let’s get you setup so you get the most out of our service
Done, Let's add some stocks
Brilliant - You've created a folio! Now let's add some stocks to it.

  • Apple (AAPL)

  • Shell (RDSA)

  • Twitter (TWTR)

  • Volkswagon AG (VOK)

  • McDonalds (MCD)

  • Vodafone (VOD)

  • Barratt Homes (BDEV)

  • Microsoft (MSFT)

  • Tesco (TSCO)
Save and show me my analysis