The Top 20 Clearest Trending US Stocks - With Dominant Returns Over UK Small And Mid Caps

27th Nov by Edward Croft

99 comments 4566 reads

This week I had an email from a long-term subscriber whose SIPP has grown by 82% year to date. He’s very much a “quality and momentum” investor, who expanded his horizons to North American markets a number of years ago when he felt the growth prospects were poor for UK stocks. It’s been tremendously successful for him, with his biggest SIPP winner year to date being Celestica (NYQ:CLS) - a Canadian provider of Cloud Infrastructure for Google and Facebook - returning 743%. “I would never have found those companies without Stockopedia”.

I have to admit I have as much home bias as anyone, and haven’t participated enough in the US market gains. Knowing the UK markets well, and how to craft returns from them has worked, and I am cautious with the high valuations inherent in US stocks. Nonetheless, I do keep kicking myself when I hear of growth company stories like this. What our companies struggle to achieve in a decade, it seems like American companies can achieve in a year.

So I thought I’d finish up this triptych of articles on Trended Momentum with a look at the USA.

How have the May list of 20 US stocks performed?

Back in the “Didn’t buy Warpaint?” article published on May 30th, a few asked me to run the same process that I’d run on UK stocks, on the USA. The list was shared in the original comments. Since that date, the S&P 500 has appreciated 15%, but the US Trended Momentum list has significantly outperformed with a 23% gain - and this was a list that included General Motors not Tesla!

Since that date, there have only been 3 losers in the list, with 17 winners. The biggest winner +93% has been the $12bn Sprout’s Farmers Market, which appears to have been a classic Peter Lynch style multibagger, followed by the airline SkyWest +60% and Interactive Brokers +51%. It’s not all been about Artificial Intelligence and Cloud Computing in US markets.

This is past performance over an extremely limited timeframe of course, and the near future may not see such strong results - but who knows. Momentum breeds momentum, and the broad Russell 2000 Index has just broken out to a new all-time high.

Reviewing the history

I’ve backtested a similar process to the one…

Unlock with Email

Unlock with Google

Already have an account?
Login here

Paul Walton

3rd Dec

SFM is just an example of a strong trend, doesn't matter what it is

Dave R

Wed 5:24pm

Good point regards getting back to the trend clarity question. I read Ed's bullet "Calculate the trend clarity for each. (Linear regression → R2)" as similar to Clenow's Stocks on the Move book. In that Clenow gets a 90d R^2 figure and multiples that by the annualised 90d return. Appreciate Ed didn't use the return multiple, but I used a single R^2 figure based on a particular lookback period (e.g. 90d, 6M, etc). In your graphs you show R^2 over a period of time, so assume you're measuring it on a daily basis rather than a single point in time (today-lookback period)? Are you then using the daily R^2 chart (including R^2 trend line) for the basis of your selections? Also what lookback period are you using for R^2?

Also the other thing to note is that Cernow and others have all said that for this type of strategy to work you need to work with "total returns" data not plain "share price" data normal found on the internet. The reason is that dividend paying stocks will have a "false" price variation (the standard post dividend share price drop) that impacts R^2. Whereas a non dividend stock that has the same yearly total returns will have a better chance of a good R^2 score as it's not hit with the post dividend dip.

Wed 6:14pm

I'm using 90 days like Clenow does for consistency. I then compare stock price data over 252 trading days + 90 days and compare it against a fixed 1:252 range (gives a straight line at 45 degrees left to right ,a proxy stock with a perfect trend if you like) in a spreadsheet to get the R^2 plot. I had nothing to compare it against until I discovered indicators in Tradingview for R^2 and the adjusted slope and I get the same result thankfully (see screenshots) so I must be doing it right. Clenow says to buy based on the highest adj slope then work down until you run out of money but I find breakouts to be more intuitve where the R^2 and adj slope are rising indicating a developing trend e.g. On Beach group (LON:OTB) which is in my multibagger screen + travel is a hot sector (IAG, IHG for instance).

With M&G (LON:MNG) you could use a bit of Technical analysis and say its performing a higher 2nd peak then 3rd peak, ok the trend is getting stronger and R^2 stays high

On the other hand,with Pan African Resources (LON:PAF) for instance you can see the peaks are getting lower and R^2 is pointing down and going below 0.7, so perhaps the trend is weakening

Screenshots for M&G (LON:MNG)

Rusty2

Wed 8:25pm

M&G (LON:MNG) l don't think so, maybe Morgan Sindall (LON:MGNS) ?

Re Pan African Resources (LON:PAF) could be a Volatility Contraction Pattern (VCP).

Parsley

Thu 4:53pm

Very grateful for this post - thank you

Fri 7:58pm

Thanks Paul for the guide to Tradingview scripts, very useful for me. I need to spend a bit of time understanding the Exponential Regression Slope Annualized with R-squared Histogram aspect.

Fri 10:53pm

I think there's a total return data option on Tradingview. I prefer to look at weekly charts initially so divis wouldn't make much difference imo. So long as the R^2 is at a certain level or maintaining one then you get an instant idea of how strong the trend is compared to others. You're not just relying on your eyes to get a feel for the trend

Sat 12:52am

Ed, what has happened to the folios that contain US stocks? My old one is based in £, but the rates of exchange, after the close are wrong, seems to be using around $1.32-1.36 to £. Before close they were correct, l wondered why my profits dropped after the close.

Also on my recent US folio, it is in US$s, neither folios currencies can be changed.

Sat 6:01pm

Thanks again Paul. I never knew Trading View had that functionality. For reference it's the ADJ button in the bottom left of the chart screen.

Saucepan

Sun 11:08am

I found this discusson very interesting, especially the helpful comments regarding the possibility of setting up in Google Sheets. The attached image is a screenshot of my attempt to do this, and seems to be working well. I used weekly closing prices over 6 months. The two columns on the extreme right compare (R2 and slope values for) the most recent 3-months with the whole 6 months: to gauge whether trend is strengthening/smoothing or not. I got a lot of help with formulas from ChatGPT, as I am no expert!

For R2 and slope I used conditional formatting to highlight cells where R2 and Slope were greater than 0.8. I'm happy to share the formulas used for each cell in Row A2, if of help to others. Of the examples shown, ROAD ticks green on all six criteria.

Sun 5:14pm

As promised here's how I calculate the math needed for the functions, outlined in Clenows book, in google sheets using googlefinance data for stocks and a few neat formulas. Apologies if Im teaching you to suck eggs. I'll try and keep it as brief as possible as some formulas require further explanation if you're not familiar with them.

My formulas are in A column and U thru AA, this is purely historical and you can use any cols you like. All formulas begin with the equals sign. SHEETS.NEW opens a new sheet on the Chrome browser. NOTE: you dont need all the formulas but there for reference. Also, I'm using a 90 day lookback period

In column A you need numbers starting 252 thru 1 starting with 252 (number of trading days) in cell A2. Quite a nifty formula for this is:

=QUERY(ARRAYFORMULA(ROW(A1:A252)),"SELECT * ORDER BY Col1 DESC")

Next Column U contains the ticker for the stocks you're interested in, preceded by LON: for UK stocks e.g. LON:BT.A for British Telecom

V column holds the googlefinance function to give the close price in U:

=GOOGLEFINANCE(U2) then copy down for other tickers in the column

W holds the Log number function:

=LN(V2)

or use the function to get the Log for the whole column U at once:

=ARRAYFORMULA(ifna(IF(U2:U="","",LN(V2:V))))

X column gives the exponential slope (this is where Column A comes in):

=ifna(IF(U2="","",SLOPE(ARRAYFORMULA(iferror(LN(query(QUERY(GOOGLEFINANCE($U2,"CLOSE",$T$4,$T$3,"DAILY"),"SELECT Col2 ORDER BY Col1 DESC LIMIT 90",0),"SELECT * limit 90")))),$A$2:$A$91)))

$T$3 holds the formula:

=today() which is todays date, and $T$4 holds:

=today()-365 date a year ago

or alternatively

=T3-365

Y holds the annualised slope:

=(POWER(EXP(X2),252))-1

or to calculate the whole column with one formula in cell Y2:

=ARRAYFORMULA(ifna(if(U2:U="","",(POWER(EXP(X2:X),252))-1)))

Z column holds the formula for R-squared: =

=RSQ(ARRAYFORMULA(LN(QUERY(GOOGLEFINANCE(U2,"CLOSE",$T$4,$T$3,"DAILY"),"SELECT Col2 ORDER BY Col1 DESC LIMIT 90",0))),$A$2:$A$91)

and finally in column AA if you want the Adjusted Slop simply multiply Y by Z*100:

=Y2*Z2*100

or with one formula in AA2 for the whole column:

=ARRAYFORMULA(IFERROR(ifna(IF(U2:U="","",Y2:Y*Z2:Z*100))))

I'll put a cut down spreadsheet together with all the formulas and post a link

Mon 9:44am

Paul: I have replicated your sheet, following your instructions. It works great - so they are spot on and very clear. Big thanks from me!

One question: if, for example, I want to change the lookback period from 90 days to 180 days, which cell(s) do I adjust? Is it only the formula in Cell X2, where I see "limit90"?

Mon 9:55am

Yes that should work, probably too long a lookback tho. You can also reference a cell with the value in it. e.g. in cell T5 put 180 then change the 90 with "&$T$5&"

That way you can test different values

Mon 10:13am

Great, thanks for the prompt and helpful reply!

af20001

Mon 12:03pm

That's a great looking spreadsheet, Saucepan!

It's interesting to see the different approaches people take to this momentum strategy. Having just finished reading the Stocks on the Move book, I found a python script online that tries to replicate the strategy employed by Andreas Clenow, not just in stock selection but also position sizing. The script was a few years old so some things no longer worked, but I've managed to hack it to get it working again. I tried running it on the universe of US stocks (with a lookback of 125 days), and with some further selection (removing stocks where the price is so high the position ends up being less than 10 shares, Stockranks < 75, QM < 85) I ended up with the following:

DASH
ENVA
FDP
EXEL
CBRE
PYPL
PLXS
CSWI
FTDR
CNK
UAL
CORT
VSCO

I've added them to a mock portfolio to see how they perform going forward.

Mon 12:08pm

I realised yesterday when looking at the FTSE100 and SP500 stocks that you need to consider the Adjusted slope too. R^2 might be high e.g. above 0.7 but it doesnt tell you the direction of the trend! Adj slope -ve with a high R^2 indicates a strong trend down e.g. BME

Mon 1:56pm

Will be interesting to see how they compare to Ed's selection, not much overlap. Ed's are off to slow start, down 3% on the S&P500, early days.

Yes some US prices are too high, puts me off buying them. On folios l tend to use $10,000 rather than $1,000. Is Booking Holdings (NSQ:BKNG) excluded for that reason?

Mon 4:25pm

Make that now 5% down, down 1.9% today, some are taking a pasting. Nearly all the top risers this year, in the US markets, are well down today. I don't know why.

Mon 4:28pm

I'm not sure why Booking Holdings (NSQ:BKNG) was excluded, it didn't appear in the final list (before I applied my rules). The script will exclude any stocks that have had a rise of 15% on any given day during the lookback period, as well as any that have crossed the 100MA. It is also limited to selecting the top 50.

For anyone interested in Python, the script I found is here:

https://github.com/skyte/momen...

It doesn't work as it is, some tinkering is required to get it to run without error.

The top 20 clearest trending US stocks - with dominant returns over UK small and mid caps

How have the May list of 20 US stocks performed?

Reviewing the history

Unlock the rest of this article with a 14 day trial

About the Author

Edward Croft

99 comments

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

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