The Blog Experiment

In my last post I said I would go through the blog recommendations of a select number of blogs since January 2012 and monitor performance of a mechanical portfolio that follows their buys and sells.

The rules are (slightly updated since last post):

  • Starting in January 2012 with $100,000 cash
  • Each position size will be 3 or 5%, depending on cash balance and authors position size
  • Buy price will be the price 1 day after the post was public
  • Only enter position if the author personally has a position
  • Exit if any author does
  • If there is no cash in the portfolio and new position is needed then the position with the highest gains will be sold to accommodate it
  • Long shares only, no options, shorts, arb plays or bonds
  • No over the counter stocks
  • Currency conversion charges assumed to be 1%, dealing fees $15

The blogs I began following were:

(I removed OTC adventures as it only really covered over the counter stocks).

The results

Firstly I’ll start by saying I completely underestimated what a mammoth task this was. It has taken around 8 hours of solid work just to do 2012. Getting historical quotes on companies, along with exchange rates and making a virtual portfolio which I could get historical values of and including dividends was a nightmare. It got even more complicated when the portfolio ran out of cash and I realized none of the holdings had gone up by much making selling decisions difficult.

The full list of transactions so far can be found here.

Blog experiment 2012

H1 2012 performance

At the 6 month mark the portfolio was fully invested with 20 stocks at 5% position sizes. The value was $104,125 representing a 4.1% increase. This compares with the S&P tracker gain of 7.0%. That isn’t too bad considering it obviously takes a while for the cash to get invested in the market, and ideas don’t usually pay off so quickly.

Change of strategy

At the 6 month mark it became clear that the 5 blogs were generating far too many ideas for my portfolio to handle. Already fully invested it would have required selling 2-3 positions every month after mere 5-10% gains. I therefore removed two of the blogs, Barel Karsan and Value & opportunity.

This decision was based partly on performance of the picks to that date, but also how well the other writers kept readers up to date on their portfolio, which made copying them easier.

So from July 2012 I followed only 3 blogs

I also reduced the default position size to 3% to make room for more ideas.

H2 2012 performance

In the last 6 months of 2012 the portfolio returned 10.9% versus the S&P 500 tracker return of 4.5%. So in its first full 6 months of being fully invested it did well.

Individual results were mixed however, the full spreadsheet has individual realised gains/losses, the largest losses were on Barel Karsan positions that I liquidated when cash was needed and no holdings were close to target. It probably wasn’t the best time to sell but most have not performed that well since either.

More results

I will keep going up to 30 June 2014 and post the final results hopefully next week. I was hoping to do it all in one post but had underestimated the sheer scale of the task. I do plan to keep this going long term, just to see how it does, and will probably add another couple of blogs to the list from 2014 that produce good ideas but in a manageable quantity.

Founder of Investing Sidekick. Works as a research analyst and is an avid value investor, always searching for undervalued shares.

3 thoughts on “The Blog Experiment”

  1. That’s a lot of work. Thank you for posting it!
    The interesting conclusions, of course, will come with several years of results. Looking forward to see how the portfolio performs up to June 2014.
    Thanks again for doing this!

  2. Great idea, thanks for sharing!

    Because currencies and commissions are individually depends on your country, broker and hedges i think it is less valuable.
    Also i would not take out stocks because running out of money but stick to the blogs recommendations and just use their dates & prices of buying and selling + dividends (and if no selling have been made i would use the current price.)
    A comparison between all blogs as for all stock ideas exactly as they did it would have been better in my point of you.

    Thanks for putting time and effort!

Leave a Reply

Your email address will not be published. Required fields are marked *