# Was the BTC Pizza deal really such a bad deal? Happy BTC Pizza day!

--

Also, visit our webpage Forum discussion for similar analysis and musings:

https://www.quantonomy.fund/forum/market-mis-behavior/was-the-btc-pizza-deal-a-good-deal/edit

On May 22 2010 Laszlo Hanyecz, from Jacksonville, FL, agreed to pay 10,000 Bitcoins for two delivered Papa John’s pizzas. He sent the BTC to another online BTC early adopter and this person used USD to buy the pizzas and asked the pizza parlor to deliver them to Laszlo. History was made. It was one of the earliest recorded transactions in BTC and maybe the first physical object bought using cryptocurrency (even if in an indirect manner). It is often used as a way to show how much value BTC gained over the years. At a point this year BTC reached a high of $63,000 so, the two pizzas were worth $630 million when BTC reached its all-time high. While this way of converting the value of BTC in dollars has some usefulness, showing that BTC is a strongly anti-inflationary asset (most goods need more dollars to be bought as time pass by) it is also a not perfectly fair way to estimate how good of a deal the Pizza Acquisition was. We need to find a way to estimate the value of BTC at that time to understand if the transaction was a decent deal or not, given the fair market value of BTC then and not now. The first BTC exchange created was MtGox on July 18, 2010. These transactions were recorded and relatively easy to find online. There were few registered BTC transactions earlier than this date that converted BTC into fiat currencies. It is not easy to find them but I plotted 2 of these early transactions on the graph above. The first one happened on October 12, 2009, where New Liberty Standard buys a total of 5050 BTC from an online user called Sirius for an amount of $ 5.02 through PayPal. Here is a good reference about the early history of BTC transactions:

https://academy.bit2me.com/en/history-exchanges-bitcoin-trading/

The graph above is a power-law BTC price model I have worked on for quite some time now. I have been one of the first people over the years pointing out that BTC has a predictable, mathematically interesting price pattern that is not found in other assets. This was much before the Stock to Flow (S2F) price model became popular. You can find two of my early posts here (username Econophysicist1 on reddit):

and also here with a better model given I had more data:

https://www.reddit.com/r/Bitcoin/comments/9cqi0k/bitcoin_power_law_over_10_year_period_all_the_way/

also, you can watch this video, where I explain more about the model:

I will elaborate more on BTC mathematical behavior in subsequent forum posts. But to summarize the model plots the price of BTC on a log-log plot (log on the x-axis, and log on the y-axis). This is very unusual. It is the case that assets that have grown by several orders of magnitude in a given period of time are plotted in what is called a semi-log graph, which is a log of the y axis (usually price, or market cap) only. I never saw any asset price history plotted in this way before. The reason to do this is that it can reveal if the asset has a power-law growth. Power laws are very common in nature in particular when there are multiplicative forces in action (instead of linear processes) and this is typical of systems that are the result of many agents interacting in nonlinear ways. Power law manifest often in natural and human phenomena like rivers and mountains formation, cities expansion and growth, distribution of wealth in a population, and so on. A good book on this topic is Scale by physicist J. West.

The model was constructed using the earliest consistent BTC transactions available, data from the first BTC exchanges. These data points are the black line in the graph. I then fitted a straight line using a regression model through the log of the price and the log of the time, this is the red line in the graph. The idea is that if a phenomenon shows power-law properties it will look like a straight line in a log-log graph. The fit is very good and the R², a measure of goodness of fit, is very high, at about 0.94 (that means 94 % of the behavior of BTC can be explained by this model). We will go into the math of that in another post. I also reorganized the data point to be represented as days from the Genesis Block that is the day the first BTC was created, on January 3rd, 2009.

I also added, as explained above some of the first recorded transactions that didn’t happen on a real exchange with purple dots. The Pizza Deal is the third purple dot as indicated in the graph. What is interesting to notice is that the other 2 dots are quite close to the general trend line (the red curve). In other words, these transactions, even if they happened early on, were relatively fair. The 2 pizzas were bought for about $25 dollars giving a value to BTC at about $0.0025. If we use the trend line as a better reference point the value of BTC would have been $0.051 or 20 times more. Laszlo could have bought 40 pizzas instead of 2 if he asked for a more fair deal. Being Italian I approve of the choice of using pizza as the physical mean of exchange but it was still a very bad deal. Of course, the value of this transaction is more historical and a proof of concept given the two pizzas were some of the first items ever bought with BTC or any cryptocurrency at all, so for sure an event worth celebrating and memorializing forever.

Happy Pizza Day to all crypto enthusiasts!

Rember to visit us at:

for more interesting discussions about the (Mis)behavior of Markets.