Virtual Worlds and Real Metrics:

Virtual Worlds and Real Metrics:

Tangible Statistics about Imaginary Goods

In recent years, virtual currencies, virtual economies, and virtual goods have become a common part of the internet landscape. However, while there is now widespread consensus that virtual economies are an increasingly valid and fruitful monetization strategy for internet applications, there are very few management tools for virtual economies, and, aside from a few obvious bromides ("Avoid hyperinflation!"), no agreed-upon consensus as to the key metrics to use in doing so.

This talk will begin with some history (covering some early ideas in measuring virtual economies) and then explore some of the metrics that make sense for a virtual economy. We will cover both how to increase profit in a virtual world and how to detect interesting user behaviors and profiles through economic measures. Throughout the talk, screenshots of Twofish's analytics platform will be used to illustrate key points.

Well, Kind Of

Outline

• Who Am I• Virtual Worlds: A Primer• Virtual Economies and Real Economies • Making it Concrete: The Velocity of Money• Moore’s Law Kicks Ass

Outline

• Who Am I• Virtual Worlds: A Primer• Virtual Economies and Real Economies • Making it Concrete: The Velocity of Money• Moore’s Law Kicks Ass

Who Am I?

Bill leads the Twofish Elements product roadmap, platform development and engineering organizations. Bill's love of economics started as an undergrad, where he taught econometrics and was the recipient of a URECA grant in economic modeling and continued throughout his grad school career in mathematics.

An active and vocal participant in the software development community, Bill has published 2 books and more than 20 journal articles, and is a long-time board member of SDForum. Bill has held various chief architect and executive engineering roles at companies including Engage (acquired by Spark Networks), Echopass, and Hipbone (acquired by Kana).

This is not a Twofish Product Talk

• More about illustrating the problems than talking about the (Twofish) solution

• Goal is to take a step back and talk bigger picture / deeper than is usually the case

• Goal is also to be somewhat provocative– “I would rather be vaguely correct than precisely

wrong” – John Maynard Keynes

My Employer is Not To Blame

• Twofish is a great company– We make software to support virtual economies

(SaaS backend play)– I’m the CTO

• You ain’t the customer, I ain’t here to sell you nothing, and this is mostly my opinion– “And besides, the wench is dead.” – Christopher

Marlowe.

Ask Questions

• This is going to be an idiosyncratic slice of – The last 30 years in gaming– Economic theory– Software design– Other stuff

• “The footnotes are not entirely without merit” – G. Santayana

Outline

• Who Am I• Virtual Worlds: A Primer• Virtual Economies and Real Economies • Making it Concrete: The Velocity of Money• Moore’s Law Kicks Ass

It All Began in 1985 Or So

Habitat Really Did Pioneer Things

• Primitive Graphics• But an economy– You could buy items

• And people were people– Communication between people mattered– Communities formed– Ethical debates over killing other people– Accumulating items for status– ….

… Avatars are hatched with 2000 Tokens in their bank account, and each day that they login the receive another 100T. Avatars may acquire additional funds by engaging in business, winning contests, finding buried treasure, and so on. They can spend their Tokens on, among other things, various items that are for sale in vending machines called Vendroids.

There are also Pawn Machines, which will buy objects back (at a discount, of course).In order to make this automated economy a little more interesting, each Vendroid had its own prices for the items in it. This was so that we could have local price variation (i.e., a widget would cost a little less if you bought it at Jack's Place instead of The Emporium). It turned out that in two Vendroids across town from each other were two items for sale whose prices we had inadvertently set lower than what a Pawn Machine would buy them back for: Dolls (for sale at 75T, hock for 100T) and Crystal Balls (for sale at 18,000T, hock at 30,000T!). Naturally, a couple of people discovered this. One night they took all their money, walked to the Doll Vendroid, bought as many Dolls as they could, then took them across town and pawned them. By shuttling back and forth between the Doll Vendroid and the Pawn Shop for hours, they amassed sufficient funds to buy a Crystal Ball , whereupon they continued the process with Crystal Balls and a couple orders of magnitude higher cash flow. The final result was at least three Avatars with hundreds of thousands of Tokens each. We only discovered this the next morning when our daily database status report said that the money supply had quintupled overnight.

In the years since then …

• Vast expansion in graphical capability• Vast reduction in barriers to building• Major success stories:

– World of Warcraft– Linden Labs– Habbo Hotel

• Better gameplay, more sophisticated digital asset systems, the emergence of gold farming, the emergence of secondary market systems (player to player commerce)

• But not a lot of change in the fundamentals of things

Key Terms

• Virtual Currency• Digital Good• Item• Primary Market• Secondary Market

Outline

• Who Am I• Virtual Worlds: A Primer• Virtual Economies and Real Economies • Making it Concrete: The Velocity of Money• Moore’s Law Kicks Ass

Very Seductive Analogy

• At the “artifact level”, Virtual Economies and “Real Economies” share some common elements– Currency – medium of exchange, unit of account,

store of value• Real and virtual currencies are all of these

– Items and instances that get bought and sold• Whether the asset is purely digital or purely physical,

there are still things being bought and sold.

• At a deeper level, the analogy is misleading

Difference: Circulation

• Real world: Money doesn’t escape; it circulates

• Virtual world: “Cash in”, “cash out”, and the idea of “sources” and “sinks”

Virtual World Economic Circulation

CashIn

Vectors Transactions(some player to player, mostly

player to world)

CashOut

Vectors

Other Sinks

(simple case)

Virtual World Economic Circulation

CashIn

Vectors Transactions(some player to player, mostly

player to world)

CashOut

Vectors

Other Sinks

Gray Market

Secondary Exchanges

Currency Exchanges

Difference: Types of Currencies

• Real world: currencies mostly interchangeable, via foreign exchange

• Virtual world: “loyalty currencies”, “earned currencies”, “reward currencies” and “paid currencies”

• Foreign Exchange behaves very very differently

Difference: Control of Purchase Points and Distribution

• Hard for people to trade virtual currencies and items outside of virtual world– “Gray markets” do exist, but lots of friction

• Means that you can actually have different foreign exchange rates – Based on bulk volumes– Based on user demographics

Difference: Tangibility

• In real economies, most goods are subject to laws of physics– They’re in a place, cannot be transported easily.

• Up until very recently, most money was subject to the laws of physics

• In virtual economies ….

Difference: Pricing Discrimination

• Real world:– Location-based differentiation, coupon-based

differentiation, …– But the cost of money is pretty similar (except rich

people can borrow at better rates)• Virtual world:– Arbitary criteria on all pricing

Difference: Scarcity

Scarcity (also called paucity) is the problem of infinite human needs and wants, in a world of finite resources. Society has insufficient productive resources to fulfill those wants and needs. Alternatively, scarcity implies that not all of society's goals can be pursued at the same time; trade-offs are made of one good against others. In an influential 1932 essay, Lionel Robbins defined economics as "the science which studies human behaviour as a relationship between ends and scarce means which have alternative uses."

Difference: Opportunity Cost

Opportunity cost or economic opportunity loss is the value of the next best alternative foregone as the result of making a decision.[1] Opportunity cost analysis is an important part of a company's decision-making processes but is not treated as an actual cost in any financial statement.[2] The next best thing that a person can engage in is referred to as the opportunity cost of doing the best thing and ignoring the next best thing to be done.

Difference: Comparative Advantage

In economics, comparative advantage refers to the ability of a person or a country to produce a particular good at a lower marginal cost and opportunity cost than another person or country. It is the ability to produce a product most efficiently given all the other products that could be produced.[1][2] It can be contrasted with absolute advantage which refers to the ability of a person or a country to produce a particular good at a lower absolute cost than another.

Famous Economists of the 1800’s

• Adam Smith: invisible hand, analogy of the pin factory

• David Ricardo: comparative advantage, the evolution of trading patterns and the international economic order

• Karl Marx: the workers control the means of production!

Big Important Question

• Virtual and Real Economies have the same artifacts– Means that the measurements you can make for

real economies can be made for virtual economies– And the people are the same

• But Virtual and Real Economies have very different fundamental underlying structures

• So what are the right metrics for Virtual Economies?

Outline

• Who Am I• Virtual Worlds: A Primer• Virtual Economies and Real Economies • Making it Concrete: The Velocity of Money• Moore’s Law Kicks Ass

The Fundamental Question

• How do I measure the health of an economy?• One answer is “an economy is healthy if people

are buying stuff”• If the world is closed and all money is recycled,

then you can talk about “how many times a dollar is used in one year (on average)”– Note this assumes transactions are mostly

commensurable and involve physical money• How do you calculate that?

I have never met with any attempt to determine in any country the average rapidity of circulation, nor have I been able to think of any means whatever of approaching the investigation of the question, except in the inverse way. If we knew the amount of exchanges effected and the quantity of currency used, we might get by division the average numbers of times the currency is turned over; but the data, as already stated, are quite wanting.

Jevons, 1875

Measuring Velocity of Money

• Tons of effort expended on this• And lots and lots of recent newspaper articles– The velocity of money slowed late last year– The “multiplier” effect is a related statistic.

• Take GDP• Divide by Money Supply

Key Point

• This is one of the most analyzed macro-economic statistics out there

• The original idea: try to figure out if people are buying stuff, was good

• The execution was definitely colored by the environment:– How do you do that in 1875?– Even now, how do you do that?

Outline

• Who Am I• Virtual Worlds: A Primer• Virtual Economies and Real Economies • Making it Concrete: The Velocity of Money• Moore’s Law Kicks Ass

Reminder: I’m not Unbiased

• I work for a company that does this for a living

Starting Points

Twofish Architecture

Twofish ElementsR900OLTP

R900OLAP

ETL defined using Kettle

Mondrian(open source OLAP)

Highly CustomizedVersion of OpenI

The Point

• Perfect Data– Twofish Elements is, essentially, a highly scalable accounting

system that tracks millions of accounts simultaneously– We also store user demographic information and have a virtual

title system (“items”)• WOW Hardware

– 128G of Ram with 4 cores is now around $10K• The BI Revolution

– Pentaho is a couple million lines of code we didn’t have to write• (and, of course, some clever people writing some very

clever code)

Velocity of Money Revisited

• “Single Account Coin Velocity”– If I model a user’s account as a FIFO queue, how long does it

take before they start to spend money they deposit today?• I can compute this for each and every user in my world– For each and every currency they have

• And then I can aggregate by demographics (or across the world)?– What’s the velocity of money for women, aged 30 to 39, in

California?

Sam Walton’s Dream

• Item Sales, broken down by – User demographic– Time– Type of Item– …..

Ben Bernanke’s Dream

• Accurate macroeconomic statistics

(and then we can drill down by type of sink/source etcetera)

Dave McClure’s Dream

• Automatically run tests; adjust prices and exchange rates to maximize revenue

????

(not available in this quarter)