Magic Formula on the ASX

[systematic]

Guys, please don't take this as, "this is my thread, you can't post to it"...but this thread, like the NCAV thread I started...I'd like it to stay on the topic it was set up for.

Can we please move any "charts vs FA vs TA or whatever" to another thread?

 

[Boggo]

Fair enough.

The stock is ACO.

That chart represents the last months of the company's history as a tradable stock. It was suspended at all time highs. And now it's in administration and I doubt shareholders will see a cent.

The main point, it's in a up-trend, and if you were in the trade, and it's pretty hard to deny that some discretionary or technical systems traders would have found entry signals, you'd have lost the lot. I don't see many obvious exit signals.
.....

I've had one of those too with TA, AED, back in around 2007 I think (don't even want to look), and it took me to the cleaners
Lesson there was having the level you are willing sell at being set to much lower than the stop trigger otherwise your sell price can get "jumped" over and you are still holding (hope that makes sense, current example pic below).

Thanks for the reply, Boggo.

...Been around too long to have any sensitivity about that. But I do think your amended comment reads better. I don't like when opinion or experience is stated as fact for everyone.

Fair comment.

Stop Trigger example

 

[lauriejay]

Magic Formula for ASX

After reading Greenblatt's book "the little book that beats the market' ,i'm contemplating using his methods on the ASX.
However his website is very handy but only works for the US Markets.
Does anyone recommend a good Stock Screener website or software that is free for a beginner.
I would pay for one in the long run but as i'm new i don't want to spend hundreds of dollars on software that won't help me.

Also i'm thinking of joining NAB Trade as they have the option of trading in the US aswell as Australian markets.
Has anyone had any negative comments about using them?
I was going to go Bell Direct or CMC but they don't have an overseas option for trading.

Any comments would be gratefully appreciated.

 

[luutzu]

Re: Magic Formula for ASX

After reading Greenblatt's book "the little book that beats the market' ,i'm contemplating using his methods on the ASX.
However his website is very handy but only works for the US Markets.
Does anyone recommend a good Stock Screener website or software that is free for a beginner.
I would pay for one in the long run but as i'm new i don't want to spend hundreds of dollars on software that won't help me.

Also i'm thinking of joining NAB Trade as they have the option of trading in the US aswell as Australian markets.
Has anyone had any negative comments about using them?
I was going to go Bell Direct or CMC but they don't have an overseas option for trading.

Any comments would be gratefully appreciated.

To screen for stocks so that you can narrow down to the select few you can understand and are currently looking real interesting... that you can practically use any free research database - on web or free with your trading account.

To get deep into the business and properly analyse its financials so you could be confident about its operations... you'd need more than those basic figures.

Either work that out from the various texts, learn accounting then set about with excel or pen and paper...

Or try my, ahem, awesome, and free, web-based app: www.danginvestor.com

 

[Boggo]

Just a suggestion.
After back testing various sectors etc I found the best results with just the stocks that make up the All Ords.

That is best for my system but will be either be too big or too small for others.

Possibly a starting point maybe rather than dealing with around 2000 stocks ?

 

[systematic]

Just for the exercise, we'll sell losers on Friday close tomorrow, and winners on next Friday's close.

 

[systematic]

Second year update.

This time we did an 'All Stocks' and also an ASX200 version (the 'All Stocks' would relate closest to the previous year).

In the All Stock portfolio: 7 out of 10 trades were profitable. The average gain of the winners was 29.5% (there were no runaways, max was JIN at 77.5%). The 3 losers lost 9,12 and 20%

The overall price return was 16.6%

Dividend payments (interestingly, 9 out of the 10 stocks paid a div, with 8 of those 9 at 100% franking) resulted in an additional 4.77%

Total return for the 12 months:
21.41%


In the ASX200 portfolio: 1 stock (SAI) was bought out half-way through the year (was not replaced). 7 out of the 10 trades were profitable. The average gain of the winners was 26.8% (again no runaways, max was MND at 65.5%). The 3 losers lost 4.3% (which was lost in the last week - wasn't sold the week prior with the other losers as it was sitting on a win) along with 8 and 26.2%

The overall price return was 14.9%

Dividend payments (paid by all 10 companies with 9 out of 10 being fully franked - not as surprising, as this is the ASX200) resulted in an additional 4.43%

Total return for the 12 months:
19.38%


The 'market' (this time I'll just use ASX300 / ASX300 accumulation) did around 2.5% and 7%


Comments for fun:
- I was surprised / amused that the 2 portfolios (All Stocks and ASX200) produced virtually identical results.
- The previous year had 2 takeovers, this year had 1 (only in the ASX200)
- FLT (Flight Centre) was common to both All Stocks and ASX200 portfolio and gained 33%
- PRT and MND were on the first year's list...both had lost about 17%. PRT remained in the All Stock portfolio and gained 30% this year; MND was in the ASX200 portfolio and gained 65% this year.
- Dividends are important; this year accounting for almost a quarter of the return, same as last year

 

[Alan5056]

Hi Systematic, I was wanting to do some Magic Formula investing but I' struggling to ind good data to do the analysis. I signed up with the trial for Lincoln Indicators but it seemed a bit excessive to pay $1600 a year just to use their data. Is there a more accessible source of data that you know of?

 

[systematic]

Not sure if @Alan5056 is still about. After 2 years of posting between mid-2015 and mid-2017 (with a total return CAGR of 20.5% - beating the market and no doubt most traders and fund managers), I didn't bother posting selections for 2017/18 as I didn't feel there was any interest or demand - which was no problem, as I don't follow this system in my own plan. It was just for interest.

Anyway - for whatever reason I came up with the selections (as I see them) for 2018/19. Possibly due to the somewhat revived interest in individual stocks and the others running excellent trading threads etc.

This time no split between ASX200 and all cap ranges etc. Just the top 30 like Greenblatt mentioned in his book.

Will follow the Top 10 (to keep in line with previous 2 years) as well the as 30 as a group (31 actually - yes, I realised that!)

I might have to bother doing the selections for 12 months ago now, hmmm...not sure. Anyway, here's the current lot.

 

[12Percent]

Hi Systematic,

I am definitely interested in your ongoing findings, I have been reading several books on this topic and am in the middle of Tobias Carlisle's Deep Value book.

The findings that the ROIC inclusion in the formula cost returns and that the Earnings Yield alone produced better results when back tested has me intrigued, what are your thoughts?

They indicated that the heavily undervalued companies produced results regardless of the quality of the company.

I notice this site: http://www.shareranker.com/ produces the Greenblatt's formula on the ASX, currently there are a lot of miners there.

 

[systematic]

Hi Systematic,

I am definitely interested in your ongoing findings, I have been reading several books on this topic and am in the middle of Tobias Carlisle's Deep Value book.

The findings that the ROIC inclusion in the formula cost returns and that the Earnings Yield alone produced better results when back tested has me intrigued, what are your thoughts?

They indicated that the heavily undervalued companies produced results regardless of the quality of the company.

I notice this site: http://www.shareranker.com/ produces the Greenblatt's formula on the ASX, currently there are a lot of miners there.

Hey 12Percent, I look at a lot of things for fun. I've made a few comments on 'company quality' - it's a tricky area of so-called factor investing. I don't use it. But I might. My personal research on 'quality' as a factor has thus far led to my current conclusion that if I end up using it, it will most likely be on its own. In other words, I largely agree. Value works - on its own. I don't like growth at a reasonable price or quality at a reasonable price. Neither does the author of the book you mentioned. But that's not to say it's not a good strategy, and in fact it may be a better strategy profile for some investors compared to not using it (the ROIC bit).

My comment on your last sentence - there's always a lot of miners in an ASX screen

 

[InsvestoBoy]

I have recently been reading up on https://www.newconstructs.com/education/

Key points:
- If you are using the accounting numbers from annual report without adjustment from the footnotes, you are nearly always using bogus numbers.
- Value measures like EV/EBITDA and P/E and performance measures like ROE do a poor job of measuring real economic value.
- ROIC, WACC and PEBV (Price to Economic Book Value) from footnote-adjusted numbers are much better measures.

They provide a lot of useful and convincing examples of this.

The numbers they produce are the result of 30+ adjustments to the annual reports.

Accounting data is not designed for equity investors, but for debt investors. Accounting data must be translated into economic data to understand profitability and the valuation relevant to equity investors. Respected investors (e.g. Adam Smith, Warren Buffett and Ben Graham) have repeatedly emphasized that accounting results should not be used to value stocks. Economic Earnings are what matter because they are:

1. Based on the complete set of publicly-available financial information.
2. Comparable across all companies.
3. A more accurate representation of the underlying cash flows of the business.

Worth a thorough digging into their Education section. After you do you can see why Buffett doesn't use normal valuation metrics and get a closer feel for what he does use.

 

[12Percent]

Sorry, I didn't get to finish my post and did not mean for my final sentence to be a statement.

I notice in the previous postings of your top ten companies there were little to no miners, possibly just a result of cyclical factors however just curious if you were omitting them for personal reasons or rules you follow.

I have nothing against miners however, there are currently 5 in the top 10 so just considering the diversification issues, pushing it out to 20 takes the miners to 8 so a little better.

I'm going to run a trial where I look at the top 20 from the list however rank them purely on value using the Earnings Yield alone.

 

[luutzu]

I have recently been reading up on https://www.newconstructs.com/education/

Key points:
- If you are using the accounting numbers from annual report without adjustment from the footnotes, you are nearly always using bogus numbers.
- Value measures like EV/EBITDA and P/E and performance measures like ROE do a poor job of measuring real economic value.
- ROIC, WACC and PEBV (Price to Economic Book Value) from footnote-adjusted numbers are much better measures.

They provide a lot of useful and convincing examples of this.

The numbers they produce are the result of 30+ adjustments to the annual reports.



Worth a thorough digging into their Education section. After you do you can see why Buffett doesn't use normal valuation metrics and get a closer feel for what he does use.

While it's true that adjustments will need to be made, there's a balance of when and where to make it else it can be an overkill.

I mean, you could arguably readjust new constructs' adjusted figures. Depends on the company, its industry and the interactions at that moment in time you're doing the analysis.

So to "properly" adjust and make detail assessment for each and every year... while great if the investor has the time, might not be effective use of resources.

For what it's worth, I think that in general... the financial statements figures should be left as they are. That we ought to first assume management to be honest and interpret, assess the company's position and performance as professionally as they could.

Then with careful analysis/screening of those same, standardised figures... an investor can zero on in to obviously "interesting" [quality, potential bargain] businesses.

It is after the basic screen and filters, based on detailed set of metrics... giving selected few interesting opportunities to pursue further... that's where detail adjustments to the figures; calling around, looking up further detail of the business etc. should then be made.

e.g. management report rising net profit. It looks great. Are they being too optimistic of recovering/getting those reported profit? Are there cash flows backing up those great profit figures? Are inventory too low/high; receivables rising abnormally [i.e. they relax their credit policies to move items and record profit] etc. etc.

 

[systematic]

I notice in the previous postings of your top ten companies there were little to no miners, possibly just a result of cyclical factors however just curious if you were omitting them for personal reasons or rules you follow.

...It would have been cyclical. I was following along with the book and only eliminating financials and utilities.

I'm going to run a trial where I look at the top 20 from the list however rank them purely on value using the Earnings Yield alone.

...Why not just look at earnings yield alone then? I'm not sure why you'd use a 'magic formula' list if you're wanting to just look at the value component? But I may not have understood.

 

[12Percent]

I am yet to find the right screener I can use besides this list, I was just filtering the first few pages of the that list to find the top 20 earnings yields manually.

Were you doing the sums yourself or do you have a link to a screener with this metric built in?

 

[BBf]

12Percent, I've almost got it to work on https://www.tradingview.com/screener/

You could set it up like I have in the image just below, but it won't be the exact Magic Formula approach. I'll explain exactly what I did below, skip it if it's too basic for you.



Instead you could set no ROIC limit and export the data to CSV format to use in excel / google sheets or whatever spreadsheet software you use, which gave me columns B-F in the image below.

Second, I filtered out Financials and Utilities (column E) and any Market Cap under 50m (column F).

Third, I created the Rank P/E column (column G) - the formula I used in G2 was "=iferror(rank(C2,C$2:C,true),9999)" and autofill that for the whole column.

Fourth, the Rank ROIC column (column H) - starting with "=iferror(rank(D2,D$2),9999)" in H2

Fifth, the Sum column (Column I) - starting with "=sum(G2:H2)" in I2

Sixth, I inserted the Magic Formula Rank column (Column A) - starting with "=rank(I2,I$2:I,true)" in A2.

So I ended up with this on 19/7/19:


But, if you do all of this, you'll end up with a different list to shareranker.com, with different values for P/E and ROIC. So I don't know where that person is getting their data, and if I should go with shareranker.com or doing this myself.

The online lists are either free and don't have all of the values needed for Magic Formula (e.g. missing ROIC or EV or EBIT...) or I can't tell if they have them in their premium products.

Any help from anyone on this would be great!

 

[JacksonMSB]

I'm currently exploring the same scenario. I'm using a fantastic website called simplywallst.com. They offer a free trial. Mine's nearly over but I will be subscribing because the info is excellent. You do need to do some fact checking though. For example, they can be a bit slow to react to insider selling. Check the ASX report for yourself. As for magic formula stocks, I have decided I will use magicformulainvesting.com website for the USA picks and then review their profiles on Simply Wall St and look for comps in the ASX.

 

[JacksonMSB]

I'm currently exploring the same scenario. I'm using a fantastic website called simplywallst.com. They offer a free trial. Mine's nearly over but I will be subscribing because the info is excellent. You do need to do some fact checking though. For example, they can be a bit slow to react to insider selling. Check the ASX report for yourself. As for magic formula stocks, simplywallst website lets you filter for price on earnings and assets as well as setting market capitol mins/maxs. Once you select your top selections (I personally also look for low debt to equity) you can create a Watch List which will then prepare performance results also all able to be filtered ie: industry, performance, dividend, etc. Fantastic tool owned and operated from Sydney, Australia.

 

[JacksonMSB]

As a back testing exercise, I used the simplywallst website to filter NYSE companies (as per magic formula) based on PE+PB+$50M+ market cap. I then sorted the results by every conceivable variable: high to low / Low to high / market cap / PE / PB / value / Total score, etc. It didn't matter. I could not replicate the magic formula website top 30 companies. I could accept some variance but not one of MF's companies appeared in my search results. NONE! Consequently, I'm having major misgivings about using this strat.