Sold, rejected but not forgotten: Tracking 2nd level and 3rd Level investing mistakes

Some 15 months ago, I had a post about stocks I had either sold or not bought after analyzing them.

This time I want to update that list plus provide some “theoretical” background why I think this is an important part of any investment process.

Types of investment mistakes

Of course there are many mistakes to be made in investing. Nevertheless, for this exercise I would categorize “investment mistakes” into 3 general categories along the typical process of most “stock pickers”. The process normally looks something like this

A) Stock screening & quick analysis
B) Deeper Analysis
C) Buy decision (or not buy)
D) Sell at some point in time

At all stages, mistakes are easy to be made. Nevertheless I would argue that most analysis goes into what I would call Level I mistakes:

Level 1 mistake: Buying a stock which performs badly.

This is quite easy to identify, because if one looks at most investment reports, one will see the performance of the current portfolio with the bad performers “jumping out of the report”. I guess most of the efforts in many investment firms goes into finding out the reasons for those underperformers and then trying to improve.

A lot less effort usually goes in what I would call “level II” mistakes:

Level 2 mistake: Selling a stock which outperforms strongly after selling.

This is a little bit more tricky. There is a lot less literature of “when to sell” compared to “when to buy”. Once you have bought a stock, there should be already a fair amount of time and effort been made to analyse the stock. So in my opinion it makes a lot of sense to keep stocks on one’s radar screen even after selling. Nevertheless it is of course much more fun to look at new stocks and forgetting about the old stuff. Especially if sold stocks systematically outperform, one should check if one is not a prisoner to some well known investment biases

In my opinion, it also makes a lot of sense to systematically track the performance of sold stocks in order to find out if one could (and should) improve the investment process

Finally, there is a third systematic family of mistakes:

Level 3 mistake: Stocks analysed intensively but not bought

This is one of the advantages of blogging: Whenever I write a longer post, I already have invested quite some time on the specific stock. So it is quite easy for me to track those posts where I did for any reason not buy such a stock.

Again, I think one should look at this closely in order to identify potential biases etc. in one’s investment process.

From theory to practice: The last 17 months

So let’s look first at all the stocks I sold since the March 2012 post:

Stock Reason sold /not bought Date Perf Perf BM Delta
KPN Special situation expired 10.05.13 34.2% 3.3% -31.0%
IVG Conv ESUG 21.05.13 16.1% 1.1% -15.0%
Bouygues Portf. Mgt. 28.06.13 21.7% 7.1% -14.6%
Total Prod dissapointing CI 08.03.13 21.3% 7.1% -14.2%
Mapfre Autumn cleaning 31.10.12 30.1% 19.9% -10.2%
Dart Too expensive 18.07.13 7.1% 2.7% -4.3%
Total Prod dissapointing CI 10.04.13 12.1% 9.0% -3.2%
OMV Autumn cleaning 30.10.12 22.6% 19.5% -3.0%
Dart Too expensive 26.07.13 4.5% 2.8% -1.7%
Buzzi business model problems 23.05.13 2.8% 2.7% -0.1%
WMF too expensive 11.04.13 5.0% 8.1% 3.1%
Fortum Autumn cleaning 30.10.12 6.4% 19.5% 13.2%
Piquadro Sold because of business problems 08.08.12 2.7% 24.5% 21.8%
Nestle Sold because of Pfizer acquisition 23.04.12 11.8% 35.0% 23.3%
EVN Autumn cleaning 31.10.12 -9.9% 19.9% 29.8%
Praktiker Sold because of Anchorage 04.07.12 -81.3% 30.6% 111.9%
        avg 6.6%

Explanation: a negative number means that the stock has outperformed my BM since I sold it, a positive number means the Benchmark outperformed vs. the stock.

On (unweighted) average, the stocks I sold underperformed the benchmark, so this looks OK. This is clearly driven by the Praktiker bonds, where I am very happy that I sold them. On the other hand, I missed out some nice gains as well. With KPN for instance, I think I was a little bit too quick with the trigger finger. My “autumn cleaning” exercise was on average also positive. So this is a good encouragement to follow-up on this exercise.

Next come all the stocks I have analysed but not bought in the same format:

Stock Reason sold /not bought Date Perf Perf BM Delta
Reply Cashflow red flag 31.08.2012 140.6% 24.5% -116.1%
Banknordik forgot to follow up 26.11.2012 70.1% 19.5% -50.5%
Curanum not really interested 05.09.2012 66.7% 25.0% -41.7%
Severfield Too expensive stand alone 21.03.2013 47.7% 7.9% -39.8%
Walgreen M&A 04.07.2012 62.7% 30.6% -32.1%
Osram Target of 23 EUR not hit 08.07.2013 35.7% 6.1% -29.7%
M6 only short analysis, issue with CI 26.11.2012 43.0% 19.5% -23.5%
Cairo undecided 27.06.2012 58.1% 38.9% -19.2%
Halfords Negative momentum 06.06.2012 55.1% 40.5% -14.6%
Hankook could not buy privately 29.10.2012 31.0% 21.0% -10.0%
Astaldi too much debt 23.07.2013 11.2% 3.0% -8.2%
Porsche still don’t like them 29.11.2012 24.3% 17.7% -6.6%
CIR no margin of safety 17.07.2013 7.3% 4.0% -3.3%
EAC Watch only 29.07.2013 2.8% 2.6% -0.2%
Canal+ no real upside 19.09.2012 19.1% 19.6% 0.5%
Rallye leverage 25.01.2013 8.4% 9.3% 0.9%
Bongrain Doesn’t earn coc 26.11.2012 17.0% 19.5% 2.5%
Greek GDP linker   10.06.2013 -2.6% 3.7% 6.3%
Accell low FCF, insider selling 26.10.2012 12.3% 20.5% 8.1%
Solvac not cheap enough 13.12.2012 4.4% 14.9% 10.4%
Mr. Bricolage Too much debt 13.09.2012 9.4% 20.9% 11.5%
Viel Underlying busienss 18.12.2012 2.4% 14.0% 11.6%
Morgan Sindall no mean reversion potential 23.10.2012 1.0% 21.6% 20.6%
WSU because of US problems, not cheap 19.04.2012 10.6% 31.9% 21.2%
Maisons France Cycle 29.01.2013 -12.5% 9.4% 21.9%
Energiedienst Business model 04.02.2013 -12.3% 12.5% 24.8%
TNT Express Too expensive stand alone 21.11.2012 -5.2% 20.8% 26.0%
KHD insiders 30.07.2012 -0.7% 27.7% 28.4%
Fabasoft track record 25.06.2012 -2.7% 40.3% 43.0%
        avg -5.4%

Here unfortunately, the average doesn’t look so good. On average, the stocks I analysed but did not buy outperformed the BM as well. The most obvious miss is Reply SpA. However here, I still think that in the long run it pays to avoid companies with questionable accounting. In this case, clearly at least for now I was wrong to discard it.

A little bit more bothers me that 2 of my potential special situations, Osram and Severfield outperformed. Both were pretty clear-cut cases (Osram, classical spin-off, Severfield classical rights issue), but somehow I was lacking conviction to follow through on the idea. I think I have to be more careful to separate my careful market view and focus on quality from the special situation area.


Looking at sold stocks and stocks rejected lat e in the investment process makes a lot of sense. In my case, I think selling looks OK, whereas I will have to work on my “special situation” investments.


  • Pingback: Looking Backward and Forward into Your Investment Decisions | Student of Value

  • Congratulation! Interesting and thoughtful considerations!
    I also have a watch list of my “sold” positions of the last five years just to look time by time. Several positions by far outnumbered me, but I didn’t feel comfortable with their risk level, so I dont regret having sold them.

    I have the “luck” that I didn’t know your blogy entry about Reply when I bought it this January, so tis got my best performer so far this year. 🙂
    It is really impressing: With some french shares I followed your blog discussions and feel good with it. But even with other shares (Reply, Fabasoft) thar I bought independently I now realized that you already discussed them. At least you seem to have a similar (but much sharper) focus for potentially interesting shares.

    By the way, Fabasoft may be worth a second look. In the last years the company invested quite a lot to get a standing in cloud computing. As far as I understand that quite changed their business modell. This year they distributed a HUGE dividend (0,37€, close to 10%) because they do no more need all the money they earn (cashflow). That trend should continue, otherwise you dont start that huge dividends. And now they seem to have a good strategical position with chances for Take-over-offers, but a take over would have to pay a big premium. Their clients, pubic authorities, should be less dependable of market cycles that private clients.

    By the way, what about “crazy shares”? Normally soccer shares are a totally no-go for me, same as airlines. The branch likes to habit quite irrational and can hire-and-fire players and trainers in really expensive ways. But actually I even bought Borussia Dortmund – low depts, solid management and trainer, really low valuation.

    • thanks for the comment. Footballshare: Might be a good trade but usually they are bad investment because they are not run for hareholders.

    • @Al Sting
      I like Fabasoft too and see big potential (They also initiated an option modell @Euro 3,50, so right now a good entry point in this regard

      HOWEVER!! You should not expect a takeover. Majority is in founders hand and he shifted this majority into a charity foundation (Stiftung). So Fabasoft is (most likely) not for sale.

      • Good argument, thank you!

        But I don´t think, that this austrian “Fallmann & Bauernfeind Privatstiftung” is comparable to “charity foundations” known elsewhere. I think, it is more like a modell to save taxes or to delay them to later times.

        Look at
        […] Als wesentlicher Unterschied zu anderen Europäischen Stiftungsrechten kann eine österreichische Privatstiftung einen rein privatrechtlichen Zweck aufweisen. […] Wird Vermögen auf eine privatnĂĽtzige Privatstiftung ĂĽbertragen, kommt ein ermäßigter Steuersatz zur Anwendung (Stiftungseingangssteuer). Bei der AusschĂĽttung von Mitteln an die BegĂĽnstigten muss die Privatstiftung auf den ausgeschĂĽtteten Betrag Kapitalertragsteuer fĂĽr Rechnung des BegĂĽnstigten an das Finanzamt abfĂĽhren. Im Ergebnis wird also die Ăśbertragung auf die Stiftung gering besteuert und es kommt zu einer nachgelagerten Besteuerung bei der AusschĂĽttung. […].

        You can also quite simple dissolute an austrian “Privatstiftung”:
        “6. Auflösung – Löschung
        Eine Privatstiftung wird aufgelöst, wenn:
        – der Zeitablauf laut Stiftungserklärung eingetreten ist,
        – wenn Zahlungsunfähigkeit eingetreten ist (durch Gerichtsbeschluss),
        – der Vorstand die Auflösung einstimmig beschlossen hat,[!!!]
        – der Stiftungszweck erreicht oder nicht mehr erreichbar ist,
        – oder der Stifter zulässigerweise die Stiftung widerrufen hat.”

        So the foundation can be dissolved quite easily, if the owners decide to sell the company for a nice price and to enjoy the rewards.

    • @Al Sting

      The dividend of EUR 0,38 was paid for FY 2012/13 to 5 Mio shares, which makes a EUR 1.9 Mio payment.

      Did you check Fabasofts business report for this period?

      Earnings had been worse than last year and cash increased by EUR 0.5 Mio without significant change of other numbers like reduced liabilities. They did not earn the dividend paid.

      Maybe the foundation needed the money and Fabasoft will perform better in future. But according report outlook, clients are still price sensitive and future keeps challenging.

      It maybe a good company and the financial structure looks still robust, but from todays point of view, future dividend payments are questionable.

  • very good thoughts. I have done extensive Analysis on Deutz and Axel Springer in early spring this year and didn’t buy them. In each case I let myself influence by other People. I learnt my lesson. Have a look at Bilfinger & Berger, it is really cheap on free cash flow and with a Little improvement of the Eurozone, they should be able to Show good results.

  • Thanks! Always appreciate good ideas how to improve the investment process and decisions.

  • Despite the short timeframe I still think this can be an interesting starting point for discussion, since many of the “not bought”-stocks would pop up in a value-screener (I know of Reply, M6, CIR, Mr. Bricolage, Accell etc.). So these outperformed the benchmark by 5.4%. Your actual portfolio outperformed by a little bit more, I assume. But still two different portfolios chosen from a quantitative selection (except for the few special situations) outperformed. I find that quiet interesting.

  • I like your thoughtprocess, but I think the timeframe is way too short to distinguish between skill and luck.
    Furthermore if you include tax your level 2 misstake are even worse because by realising gains you have to actually pay hard cash for the latent tax.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s