Semi Year SNAPS 2022 - Should We Be Timing The Market?

1st Jul '22 by Edward Croft

35 comments 8942 reads

It’s one of those years. The market has had one of its worst starts to the year in history. The excesses of the last few years need to purge their way through the system. Perhaps, if you listen to the bears, the excesses of the last few decades need to purge their way through the system. Of course, I listen to bears far too much as I have a fatalistic attraction to doom. But in my experience bears do need to hibernate regularly and will only go back into their cave when they get bored… so don’t encourage them!

Nonetheless, after a wonderful 2021 +28%, the NAPS Portfolio (my “no admin portfolio system”) is having a reversal, with a -16% fall year to date, closing for the half at a low for the year. It’s well behind the FTSE All Share which held up rather well due to its heavy loading on value sectors. Of course the NAPS is significantly outperforming the FTSE AIM All Share which is down 28% year to date - which may be more reflective of many private investor portfolios given their AIM bias.

The long run returns look a bit leaner now than at the start of the year, but still up a handsome 14.5% annualised or 175% (before dividends & transaction costs) and beating the FTSE All Share pretty convincingly.

If you compare the performance of the NAPS against the 247 Funds & Unit Trusts in the IA All Companies Sector - it would lie in fourth place over the last 5 years. That’s in the 98th Percentile of performance. Given the NAPS takes an hour per year to manage, but those other 243 Funds & Unit Trusts seem to take full time management by teams of expensive fund managers - it does make you wonder what they do all day.

Nonetheless, two funds from the excellent Slater Investments would have beaten the NAPS. It’s worth noting that we co-acquired Company REFS with Slater Investments several years ago and know them well. REFS was the product devised by Jim Slater and the inspiration for Stockopedia.com. Slater’s Zulu Principle is a wonderful read that espouses the importance of rigorous selection rules based on quality, value, growth and momentum. I’m not surprised that an investment firm that sticks to…

Unlock with Email

Unlock with Google

Already have an account?
Login here

WarrantStar

4th Jul '22

Ed. Thank you for your elaboration. Of course, most of us do not invest in a way which is market cap weighted. Mostly we invest specific amounts of money into the shares that we have selected. What I would really like to see is two charts similar to the ones that you have provided above. Chart 1) All Stockranks 0 to 100 with no market timing. 2) All Stockranks 0 to 100 with market timing. This would provide some really clear evidence about how effective market timing can be. Also please can you clarify whether the horizontal lines on the charts indicate 1) Still holding the existing shares, but not buying any more. OR 2) Selling all existing shares and not buying any more.

Edward Croft

@WarrantStar - in answer to your question - the horizontal lines are 'out' of the market - i.e. completely uninvested.

I don't have the full 0-100 results... I'll do more on this but here's a couple of indicative examples in a few deciles:

Once again:

Blue Line - is the decile
Orange Line - is the 'strategy' of either fully invested when decile > 200d MA of the decile, and uninvested when below

0-10 StockRanks

30-40 StockRanks

80-90 StockRanks

@Caulkhead

"Rather than the 200 day MA have you looked at the 90 day? I remember research from long, long ago that demonstrated that most market return was achieved when the market was above the 90 Day. When I have looked at it the 200 day MA is too slow moving to be of much use"

Yes the 90 day looks "better"... but of course may be unrealistically optimistic given the likely transaction costs of switching in/out.

This is using the 90 day MA as a timing signal on the 80-100 quintile. It "improves" the transaction costless return on the 80-100 group.

And here's the chart showing the 'trades'... every time the blue line is at 1.0 its fully invested, every time it's at 0, it's uninvested. There are a lot of thick blue lines below which essentially means the quantile 'price' is moving above/below the MA daily during that period.

Rusty2

The other issue is dividends, if you are out of the market you will be missing out of those, some of which are rather attractive at the moment, compared to cash.

cmpeckham

Interesting to see market timing discussed constructively on stocko, instead of the more usual statement I've often read on these pages that "it's impossible". I've been using my own market timing system for 22 years now. It started out in a pretty rudimentary way (based originally on the Coppock indicator) and has developed over the years as I've progressively found better ways of doing it. It's helped me enormously, not least this year where all manner of signals were available to signpost an upcoming fall in the market.

In the past I used the 200 day average discussed in this article, which has been analysed for ever. Paul Tudor Jones is often associated with it. One comment I would make on the 200 day average in these systems is that its main value is in keeping you out of the market during major downturns. It's not there to navigate every twist and turn. Consequently, it's a mistake to take signals too frequently. You'll just end up being whipsawed up and down. In Ed's analysis above he appears to use daily signals, which I think is a mistake. It's better to take monthly signals. If you want something more responsive than that, you could, for example, split the portfolio in to 4 tranches of 25% each, with a signal given for each tranche on a rolling 4 week basis.

The monthly version of the 200 day is the 10 month moving average. Meb Faber gained a name for using it in multi-asset strategies, which are worth reading about (when you have an hour available) in this:

https://papers.ssrn.com/sol3/p...

Also, similar "dual momentum" strategies to that described by Ed above, but based on monthly signals, sprung out of a book by Gary Antonacci and the concepts are summarised well in this bite sized video:

https://www.youtube.com/watch?...

I'm not saying that these are the ultimate in market timing but they are well worth the time and should get you thinking.

Caulkhead

That's really interesting. I guess there are options these days for much reduced trading costs but I fear you are correct. It's also apparent that the bulk of the outperformance came from being out of the market for the Covid crash. Early in my career I spent quite a lot of time playing around with moving averages without any real success, I'm sure methodology is much more sophisticated now though.

googlyeyes

5th Jul '22

Rather than a binary in/out of the market Ed, would it not be better to have a more tiered approach - reducing exposure/position sizing in weaker market conditions and exiting fully in the worst case. A traffic light based system (e.g. Green = Fully invested; Yellow = 1/2 or 1/4 positions, proceed with caution, Red = move to cash/reduce/sell any losing positions). I don't think it's possible to time the market exactly but it is possible to gauge the general direction/temperature and avoid some of the worst. Timing is not about prediction but about being ready and reacting as conditions change.

As you say even though the FTSE All Share was making ground and not showing as much weakness, the underlying breadth was poor (same was true in the US markets where this correction/bear market really started last year). Using net new highs/lows, Advance/Decline and other breadth indicators (McClellan etc) can give a better view of the market than the index as often stocks will peak before the indices ("the generals are last to be shot in a bear market" and as the indices are often driven by the largest companies...i.e. the generals, waiting for that can be too late).

Below is a screenshot showing the Net new highs/lows for the S&P 500 on top of the S&P 500 ETF (SPY) - the red areas are signals to be out of the market (so you can see 25th Feb 2020, and 22nd Nov 2022 were key areas signalled in the US markets, there was a few other shorter periods in between).

and in the case of just the Nasdaq, this was starting to look weak from last summer:

I do believe that the overall trend/key moving averages are important too though, to me it looks like the FTSE-All share is only starting to roll over now (the 10 week MA is declining under a 30 week moving average which is only now turning down) - the all share has missed the worst of the declines so far (compared to other markets) as it has been cheap in relative terms but if earnings start to miss, it will not be immune either (I am not predicting anything, but I am being very cautious right now)

Other books which I believe have useful concepts about timing/gauging the market (either individual shares or market generally) are Stan Weinstein's "Secrets to profiting in Bull and Bear Markets" (his concepts around stage analysis are very useful imho) and William O'Neill's "How to Make Money in Stocks" (distribution days and follow through days). Both also cover entry points for buying individual shares but have chapters on overall market conditions.

In the case of O'Neill, his company undertook extensive studies on the US markets over 100+ years (not all of these are in the book but you can find it on their website and podcasts etc) and found that 3 out of 4 stocks will move with the direction of the overall market and 50% of a stocks movement is related to the sector & industry (generally the sector/industry will move as a whole and you will have leaders within the group that outperform) - not sure if this rings true here in the UK market where there is less breadth of stocks in many sectors/industries.

I believe as private investors that being able to move to cash quickly is an edge we have over large institutions/funds which are forced to stay invested during downturns (funny how their advice is that it's "time in the market rather than timing the market" as others have said when that is all they can do if they are a long only fund) - investing in stocks is speculative by nature so undertaking it when the odds are in our favour has a better chance of success. Here is an interesting study I found online (conducted by a Dr. Eric Wish) showing the differences between staying fully invested (All days) vs. missing out on the best/worst days and the difference that would have on your portfolio - he has a blog here: https://www.wishingwealthblog....

Silver Moon

12th Jul '22

Before properly wetting my toes I've constructed a virtual portfolio following your guide Ed. There are so many ways to the holy grail of capital returns and I'm trying them all. Only trouble is, there being only 10 portfolios I had to dump my virtual Coffee Can for Silver's NAPS:)

Bitbug

28th Jul '22

Thank you for your article and the reference to Clenow's book: "Stocks on the Move". Clenow's book focusses on the S&P500 and I have started thinking on how to adapt the approach using Stockopedia's rankings. It seems like one might create a benchmark index as a substitute for the S&P500 based on a Stockopedia screen such as:

(Ignore the 'Yield TTM >= 0%' rule as this was used to get Stockopedia to include the yields when downloading). This produces some 80 stocks (as of 27/07/2022) which can then be used to create a benchmark index by averaging the returns over some specified period:

In Clenow's approach, one would only buy stocks if the benchmark index was above the 200-day moving average. Strictly, of course, this benchmark index is not an index at all as it is not based on market capitalizations. However, one might be able to use it as a proxy.

Proceeding as in Clenow, from the benchmark index of stocks, one would exclude stocks if there were price changes exceeding 15% on any one day over the past 90 days and if the current price was less than its 100-day moving average. One then computes the annualized exponential growth rate for each stock (I combine this with the dividend) and ranks them accordingly. Again, as in Clenow, the number of shares to purchase is then calculated from the 20-day Average True Range (ATR) and available capital (£100K assumed), and working from the top down, the stocks are purchased until the money runs out:

The distribution over the sectors looks like this:

The portfolio can then maintained exactly along the lines suggested by Clenow.

I suspect that this would work quite well. However, back-testing the strategy is out of the question since access to historical Stockopedia rankings are unavailable.

On the Stockreport page you can go to print and more or less there is a weekly print available, from that you can get the Stockrankings, a time consuming exercise. On the Stockreport page there is the Stockrank under the chart for a rough guide.

30th Jul '22

I appreciate your comments, including the references to recommended books. Where do I get hold of historical data on advancers and decliners for an index? You show this on a couple of your graphs.

jonno

31st Jul '22

Morning Ed,

Interesting article, but I fear all too cerebral for me. My approach is very simplistic and will no doubt not suit all. I endeavour to concentrate on the company and not the market. Within reason I stay fully invested. Also I don't use stop losses, which in a poor market tends to guarantee one a 20% loss, to paraphrase Peter Lynch. My portfolio of 26 shares would have been reduced to 3 or 4 had a 20% stop loss been employed.

If in my view if the company is performing well in relative terms I tend to be sanguine and will accept the paper loss and look to add irrespective of the market, although I will watch the 50 and 200 day moving averages. The cheaper I can buy the share, the less the risk. Thereafter it is a matter of monitoring the company and patience. Waiting for the bears to hibernate as you so eloquently put it.

Off course the above doesn't always work and having 95% of the portfolio in UK small caps and Aim has been distinctly unhelpful. Still I remain optimistic.

Portfolio down 16% YTD

All the best

16th Aug '22

@bitbug sorry, I did not realise there was a question (there are no notifications when someone comments...perhaps one for the Stockopedia team to look at). I use tradingview (just the free version), there are multiple indicators for net highs/lows but I use Matt Caruso's (Caruso Insights). There is also a similar indicator on the marketinout website (I tend to look at other breadth indicators like Advance/decline and McLellan etc):https://www.marketinout.com/ch...

Megagrumpy

23rd Aug '22

The way that the NAPS portfolio is portrayed one would think that every six months you are rebalancing the portfolio and selling off all the stocks that do not appear on the new list and buying the new stocks - ie maintaining the overall number of stocks at 20. is this the case?

Jleavett

7th Sep '22

I’m interested in not buying new positions if the price is below 200ma however this is not the same as exiting all positions when the price is below 200ma. It just means don’t buy new positions. In practice I don’t think I would sell up every time the price goes below 200ma rather I would probably sell at a defined target and not add to anything or buy new shares until the market recovers. That makes sense to me.

Munday

10th Oct '22

I have managed to keep may portfolio at a loss of around 10% by selling on every favourable opportunity I have found on the way down. Some by timing to leave on ex Divi declarations, and on any bad news on the stock. And any stocks that are close to debt. That is below 50% of book value.

But dealing costs have been high. "Never the less" I am now left with sufficient funds for any future bull market, and what I consider, a base of shares giving a decent return on the divi front.

Benson212

30th Nov '22

Thanks for the article, Ed!

For those in the discussion interested in a 'market timing' mechansim, there's a book called "How to make money in stocks" by William O'Neill (Stockopedia has a screen under Growth Investing). In his book, he states that investors can identify turning points by counting distribution days within the index. When a cluster of distribtion days occur, a market top is likely in and investments should either be sold or hedged. Further details are in his book.

Investor's Business Daily (William O'Neils former company) post these turning point signals on their website for subscribers and have backtested the strategy since October 2005 (chart below).

The rules are:

- When the market is in a 'Confirmed Uptrend' then be 100% invested in the Nasdaq 100.

- When the market is in a "Uptrend under pressure" then you're 50% invested.

- When the market is in a "Correction", then you're 0% invested.

As you can see, the blue line (the etf strategy) has outperformed the S&P500 but not the Nasdaq comp (which is more arguably the better index for comparison). However, what I have found important is the lower volatility.

Looking at the strategy, it has significantly reduced the drawdown of the two biggest declines over this period; the GFC in 2008 and the Coronavirus crash of 2020. However, during 2015 and 2016, there were many signals that led to whipsaw action and underperformance.

However, as of 2022, I am trialling a new strategy. Instead of going to 0% invested when IBD update their signal to "Market in Correction", I open a 50% short position in the S&P 500 as a hedge against my fully invested Stockrank portfolio. This means I get to reap the rewards of staying invested (Ed's blue line) whilst limiting the drawdown when the inveitable bear market ensues.

It would be interesting to see what the outcome would be if Ed re-runs his charts with a hedge in place (i.e. below the 90DMA or 200DMA), or if anyone is able to backtest themselves?

Hi BH1991. You may have seen this already. There is a lot of discussion about timing the USA market in the following Stockopedia thread: Stockopedia | Why are retail investors discouraged from using leveraged ETFs? | By Bitbug.

4th Jan '23

Are the results available now, for SNAPS 2022?

Daniel Rudd

7th Jan '23

Regarding using a timing model. It would be interesting to see how earnings momentum (earnings upgrades>earnings downgrades 1w or 1m) performs with a StockRank of 80-100 compared to the price momentum of >200d MA listed above.

Semi Year SNAPS 2022 - should we be timing the market?

Unlock the rest of this article with a 14 day trial

About the Author

Edward Croft

35 comments

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

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