What Works on Wall Street portfolio

1st Jul '20 by JamesS

13 comments 926 reads

Hi All,
I have just finished reading "What works on Wall Street" by James O'Shaugnessy (highly recommended) and have been thinking about how to construct a portfolio based on some of the factors he believes to consistently effective, specifically using some of the composite value factors and 6 month price momentum. However before I dive in headfirst I was wondering if anyone on here had any success building and running such a portfolio?

Additionally quite a specific book related question and a broader portfolio construction question

Firstly the value factor I was looking at was the Value Composite Two Indicator. Considering this is ranking of the ratios: Price/Book, Price/Sales, EBITDA/EV, Price/Cash Flow, P/E and a Shareholder Yield, would you substitute in the Stockopedia Value rank as they are quite similar? (P/B, P/S, P/FCF, P/E, Div and Earnings Yield). If not any tips on setting it up as a screener?

Also O'Shaugnessy recommends using 25 stocks. I see a minimum position size of £800 as smaller than that, the annual dealing makes a severe dent in any profitability (based upon HL fees of £11.95 per trade, 2 trades per year = 3% of position). Do you have to take the plunge and commit £20k? Or is it possible to construct such a portfolio over time? Or take fewer positions without increasing the risk too significantly?

Thanks in advance for any help

James

Unlock with Email

Unlock with Google

Already have an account?
Login here

HumourMe

2nd Jul '20

If you haven't already done so you can look at Stockopedias interpretation from a previous edition here https://www.stockopedia.com/screens/guru/james-oshaughnessy-11/ Should be a good starting point.

A broad value rank system should be similar in my opinion.

Fewer shares will mean greater volatility of returns until you build up to whatever you are comfortable with. After a point adding more will detract from performance. About 15 years ago I did some backtesting which indicated that for value shares, portfolio sizes of around 5-10 was a good balance. This was only on 5 years worth of data though. Statistically as a rule of thumb, to half volatility you need to quadruple portfolio size. This breaks down from time to time as everything correlates, we see a crash, and no portfolio size apart from zero protects you.

Hope that helps.

William K

The book should probably have been entitled 'What Used to Work on Wall Street'. In his early years as a fund manager James O'Shaughnessy had disastrous results when he followed his own advice. Although the book is a monumental piece of research, the principles have become outdated and certainly do not provide a recipe for success today. My advice would be to read some other books before determining your strategy.

ezlifeme

One of our members has an excellent video intro such reading material

This Link SHOULD? Work. https://www.stockopedia.com/co...

jjis

You might want to look at the Tiny Titans in Guru Screens which is one of the top 5 performing screens over 5 years and is quite close to what you say you are looking for. It does use P/S and 1 year relative strength though even though the write up says that he moved to a Composite Value in the 2012 updated book. See here for details.

You can however make a duplicate of the Strategy and amend it to your requirements so you could replace the P/S ratio with say Stocko Value > 60 and change the 1y rel to 6 Months if you wanted then and see what you get. You might also want tot tweak the Market Cap settings as personally I think £15m is a bit too low and £50m might be better, although I guess that could detract from the performance and from what you said it seems liquidity might not be a problem for you.

Here's one I made earlier on the above lines.

Hope this helps and good luck with your investing journey.

Regards

Jamie

JamesS

I think thats a dead link?

Interesting info on the portfolio size for value shares, it does help thanks

Thanks, I'll give them a watch.

Hi Jamie,

Certainly helps, similar to what I have been playing around with. I set the value rank at >70 as he mentions selecting from the top 3 deciles. Also tweaked the market cap as I was similarly wary of the micro-microcaps.

https://www.stockopedia.com/screens/copy-of-tiny-titans-881243?view=screen-view-173058be652

I'm just not sure to what extent using the stocko value rank differs from his value rank. I did try to set up a screen using the individual value factors but it leaves a very small population

https://www.stockopedia.com/screens/o-shaughnessy-880858

Hi Wilkonz,

Interesting point and not one that I had considered. I thought that the principles were still quite valid as the tiny titans screen has performed well over the past few years although the selection criteria are slightly different to mine and is re balanced quarterly.

Additionally the back testing he does shows it doesn't always outperform or give positive performance but will generally do well. Could it be that recently is a period of low performance?

Happy to hear more as its always important to listen to an opposing view

Thanks James. My comments were based on the 1996 version of the book and I haven't read the 2011 4th edition (which I'd guess is the version to which you're referring). It's probably better if you ignore my comments and listen to those who are more up to date than me. On the whole, I stick to large caps so I wouldn't be much help with the Tiny Titans screen either. Best of luck with your investing. William

Thanks James. My comments were based on the 1996 version of the book and I haven't read the 2011 4th edition (which I'd guess is the version to which you're referring). It's probably better if you ignore my comments and listen to those who are more up to date than me. On the whole, I stick to large caps so I wouldn't be much help with the Tiny Titans screen either.

Read both editions. Comments valid!

@James

An issue with reported mechanical small cap strategy performance when translated to the real world is spread.

My link failed due to being sourced on the old site. However, on the new site, browse to the guru screens, then make an adjustment for performance based on spread. When I did this some manual re-ranking was necessary that ended up with me discounting these.

The advice to read about a lot of alternatives is very good. Private investors have distinct advantages over institutions and these are worth exploring (also vice versa). I'd second watching the recommended youtube videos first as they are an excellent introduction, to give you a flavour of what appeals to you. It may well be a mechanical small cap strategy, but if you can find an approach closely aligned with your psychology you are more likely to stick with it.

Hi HumourMe,

Thanks for the advice about the spread, good point.

I have tried to read around the subject, everything from peter lynch and phillip fisher emphasising the importance of researching an individual company to Ben Graham and O'Shaughnessy buying baskets of stocks (may be generalising a bit much). The mechanical element is appealing to me as I dont feel I have enough time to put in the research to get the edge. Any books you would recommend?

3rd Jul '20

Hi James. You might find Burton Malkiel's classic 'Random Walk on Wall Street' useful, even if you disagree with his conclusion, namely that sensible retail investors who put all their money into a tracker fund (e.g. £VUSA) will outperform most active fund managers. If you want an edge and have limited time this is probably the way forward. Trading expenses are minimal. If you want to be a little more adventurous, you could look at an investment trust (e.g. Scottish Mortgage Investment Trust (LON:SMT)) and let someone else do the hard work. If you create a basket of 25 small caps for yourself, you're likely to spend up to 5% of your resources through trading expenses and that will wipe out most of your first year's profits.

Hi James,

Wilkonz views are sensible. I saw your question last night and given your time constraints could not think what books to recommend. I remember reading and even enjoying Malkiel, however, can't remember the contents. It is on my shelf behind me with the intention of re-reading it. A contrasting argument to (complete) randomness is presented in The (Mis)Behavior of Markets by Benoît B. Mandelbrot. However the latter won't help you with a mechanical strategy.

One approach is Stockopedia's NAPS portfolio. I won't give a link as I'm stubbornly on the old site, but you should be able to find it. Not sure how it has held up this year - I'm looking forward to its half year review.

If the guru screens were enhanced with drawdown, correlation of returns, adjustment for spread, dividends etc.. (drawdown used to be available on the old site - haven't checked recently on the new) then they could be used to 'design your own fund' and be the basis for a DIY mechanical strategy. I've requested the additional stats however I perceive that understandably most of the focus recently has been on the new site. Without the additional stats then you could find one or two close to what you seek.

This would essentially a DIY trust with expenses - although better focused than many (exceptions exist which focus e.g. Fundsmith). Depending on how involved you want to be, this may tilt the balance towards funds.

What Works on Wall Street portfolio

Unlock the rest of this article with a 14 day trial

About the Author

JamesS

13 comments

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

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