Today I will expose and analyze some of the data collected during my nine months of portfolio management. Without further wait, here's what happened during these nine months.

The first (last week's) article can be found here: Insider: Nine Months of Portfolio Management–Introduction

Overall Result

First things first, the overall result.

+45.81% was my overall profit in tix after these nine months--pretty much what I excepted at the beginning of the adventure (between 30% and 80%). After the April result, I had expected to finish with a 50% or more profit. However, being "forced" to sell piles of junk in a very limited time in order to close the portfolio dragged the overall percentage down a little bit by the end of April.

Here is the final spread sheet with all the selling prices. The month by month reports are below.

Let's take a closer look at some key figures with some charts and pies.

Performances Distribution

The distribution of all the positions performances, expressed in %:

I highlighted some examples in red. The average performance of these 152 positions is +64.3%, and you see that most ended in a zone between -20% and +80%.

Why this number is different from the +45.81% overall performance? Since all the positions didn't weight the same, i.e. I didn't invest the same amount of Tix in each position, the average of the positions's performance is different from the overall Tix performance. In other words, performances of "small" positions (where I invested only few tix) matters less in the final result than "big" positions (where a lot more tix were invested).

For instance, Burning-Tree Shaman finished with +364.6%, but contributed a very minor part of the total Tix profits, with +5.51 Tix (+0.11% of the total profits). While Kiki-Jiki, Mirror Breaker increased by +46.8% for a +45.98 Tix in profits, about +0.9% of the total profits.

Winners and Losers

Another number straight from the results: the percentage of winning and losing positions, and the average % gain and loss within each of these winners and losers:

Overall, primary and secondary portfolios combined, 76.3% (104/152) of the cards I bet on yielded positive returns. From as low as +1.9% to +422.2%. The average performance of the winning positions being +93.2%.

On the other hand, 37 of the positions I invested in, 22.4%, ended up negative. From -4.5% to -87.5% with an average of -30.3%.

Two positions finished exactly where they started.

Another 80/20 rule coincidence? Interestingly or not, the top 20% of my winners are responsible for about 80% of my Tix profits.

Month by Month Progression

Here are the Excel spreadsheets of the portfolio progression month by month. The prices were recorded the first of each month. When I was still holding the cards, their value were estimated based on Goatbots or Mtgotraders prices and were recorded as Current Value. When cards were sold, their selling price defined their value. When only part of the cards of a position was sold, the Current Value was determined by adding the value of the sold cards plus the estimated value of the remaining cards.

• Initial(Primary Portfolio)
• September
• October
• November
• December
• January
• February
• March
• April
• Final

You can see now how the different positions evolved and when they were sold and bought for the secondary portfolio. This is a brief overview of what happened during each month.

• September, October: No action took place during these two first months. I only grew the Quick Flips part of the portfolio.
• November: After the PT Theros, I started to sell some cards, mainly Chandra, Pyromaster and Archangel of Thune. Quick Flips made a big jump.
• December: I reinvested part of the sale from the primary portfolio and opened the secondary portfolio. The Quick Flips had a little down, trying to quick flip Theros cards was not a good idea as they inevitably headed down.
• January, February: More cards from the primary portfolio got sold, and the Secondary Portfolio grew well.
• March: The PT and the B&R announcement boosted the Secondary Portfolio and I sold some of these positions.
• April: I sold positions whenever they had a profitable bump and I started selling the some losers that I had in big quantities.
• End of April: I massively sold everything that was left. Mostly spending hours to close all my losing positions that had not done anything so far.

Data Mining

Not surprisingly, not all the cards got the same fate. And not all the card categories (M14, Modern, Ravnica block or Innistrad block) behaved the same. Let see here if we can draw some conclusions and distinguished patterns based on these results.

Categories Distribution

First, the distribution of the different categories. This pie is slightly biased because not all the positions were held together all the time. For instance, some positions--mostly M14 cards--were sold as soon as October, which allowed me to buy Modern and Innistrad positions.

Anyway, if I take everything into account, here is what it looks like:

With 80 positions, Modern cards constituted the most space of this portfolio. Cards from the Return to Ravnica block represented about a quarter of the total number of positions. And finally, Innistrad block positions and M14 cards accounted for a little bit less than 11% each.

This pie represents the amount of Tix dedicated for each of those categories. Again, some Tix were used twice as the income from October sells were used to fund some Modern and Innistrad positions:

As you see now, the distribution is quiet different. Return to Ravnica and M14 positions got the most Tix, with 2470 Tix and 1937 Tix, respectively--more than 75% of my portfolio combined . Modern cards, which were the most diverse in terms of number of positions, were allocated only about 25% of my money. Innistrad cards only got less that 8% of my total Tix available.

Categories Performances

This brings us now to the performance of these four categories. Did they perform accordingly to their size in my portfolio? Not really.

M14

Despite having ten winning positions and seven losing positions, the M14 cards finished with an average % profit of +59.4% and with a benefit of +905 Tix.

Indeed, to the exception of Kalonian Hydra, the loss from the losing positions was minimal. On the other hand, M14 positions held several big Tix makers with Chandra, Pyromaster, Mutavault, Chandra's Phoenix and Ajani, Caller of the Pride.

Since none of my positions received the same initial Tix investment, the comparison is not quiet fair, but six M14 cards trust the top eight of my Tix makers. I'll discuss core set speculations in a future article, but, particularly in this instance, the core set seems to be one of the most interesting and lucrative types of investments on MTGO.

Return to Ravnica

At the opposite of M14, the Return to Ravnica positions were a mess.

They have about the same winners/losers ratio as M14, but Ravnica's losers hit hard. Several of them lost more that 50% of their initial value, including my Top losing trio Varolz, the Scar-Striped, Loxodon Smiter and Duskmantle Seer. The percentage performance is a little bit confusing, as it indicates a result of +5.5%. This is due to small positions ending up positive, such as Reap Intellect, but having very little Tix weight.

Overall, the 2470 Tix of Return to Ravnica positions made me lose 107 Tix. Reasons for this failure, and how to solve it, will be discussed in one or two specific articles.

Modern

As highly anticipated as it was, Modern positions didn't disappoint at all. Almost 90% of these finished on a positive note, and the biggest Modern loser, Figure of Destiny, ended up with "only" -37%. I'm also pretty certain that in one or two more months, all the losing Modern positions at this point in time (end of April) would have been be positive. The time frame of this 9 months portfolio made me sell these cards even though they were losing--a detrimental rule of the game.

Innistrad

Similarly to Modern positions, the Innistrad block positions also performed pretty well. Few losers, a great percentage performance (+75.3%) and 240 Tix gained. A mix of redemption and Modern playability effects, combined with the cheap buy prices in October/November, contributed to the success of these positions. Here again, most of the Innistrad block cards will keep rising in the coming months. Sets rotating out of Standard are definitely worthwhile, if not safe, investments to consider.

Performances According to Cards Cost

There's probably many different ways to breakdown and interpret the data of this experience. Here is a graph comparing the buying prices of my positions and the performance of the corresponding cards:

I kind of like this representation because it makes the point that percentage performance is clearly not equal to Tix performance. It also tells us that depending on your time and the size of your portfolio, you should not necessarily consider all the opportunities available all the time.

• Here, cards that costed me less than 1 Tix to buy increased by 107% in average, a pretty decent number. With 1965 of these "penny" cards, I netted 501 Tix in profit.
• Of the cards that were in the 1-5 Tix range when I bought them, 1502 finished with a +46% increase in average. They made me 459 Tix in return. You see that with a % performance that is half of the "penny" cards category, and with roughly 25% less cards involved, I gained slightly less Tix (459).
• Now it's getting interesting. The 232 cards that had a value of 5 Tix or more when I got them were sold with a 58% profit in average. In terms of Tix, they made me 1043 Tix richer.

Time

To try to make these numbers more practical and to put them in perspective,, let's assume one thing: the time it takes to buy and sell a playset of cards. Let's consider you know when and what card you are going to buy and sell, you have made your research and now you are looking for the best buying prices and--later on--the best selling price.

To estimate that time, I'll take into account a brief research among the different bots (Mtgotraders, Goatboats, MtgoLibrary + classifieds for humans), then the actual transaction, then recording your transaction in your favorite Excel sheet. And finally you have to do the same thing whenever you sell.

For the whole buying and selling process, I estimate that time of about two minutes per playset. Sure you can be faster, but there's also times where it's going to be longer. Prices doesn't match with the price list, you have to go through two or three bots to get your playset, MTGO is lagging, etc... I think two minutes is fair.

With the figures in the chart above, it will take you ~982 minutes, or 16 hours and 20 minutes, to to make 501 Tix with the "penny" cards category. About 30 Tix per hour.

With the same maths, with the "1-5 Tix cards" category you would make about 38 Tix/hour.

Finally, dealing with the ">5 Tix cards" category, you would make more than 500 Tix/hour!

Depending on the size of your bankroll and the time you spend online, these numbers maybe more or less true and accurate. With a small bankroll you can't really deal with high price tag cards, but dealing with cheap cards can grow your bankroll fast as the % return is high. At the opposite, if your bankroll is bigger, dealing with cheap cards for a higher % return is a waste of time, your hourly Tix rate remaining low. You may want to target expensive cards that can yield more Tix in no time, still with a % return lower.

Conclusion

This was a summary of some data extracted from my experiment. Feel free to dig into the raw data and analyse them from new angles.

Choices, mistakes and lessons I made and took from this portfolio will be discussed in articles to come. They will be dedicated to more specific topics and/or cards categories.

Thank you for reading.

Sylvain Lehoux