In my last article I considered the Value Design of Atari's arcade game Pong and showed how Tokenisation, Smart Contracts, and Play-To-Earn functionality were not new concepts, but could all have been implemented by Atari when the game originally launched in 1972.
In this article, I'm going to consider tokenisation within Pong in more detail, and explore how varying the game's Value Design to introduce tokenomics might have been supported through changes to its Play, Game, Product, and Experience Designs.
At its most basic, Atari could have created an "Atari Token" and 'paid' whoever purchased a Pong game 1 Atari Token. At this point the value proposition for Pong would have been altered slightly from its original "Pay 25 cents for one game of Pong" to "Pay 25 cents for one game of Pong and 1 Atari Token". This seems like a trivial change, but only because our assumption is that 1 Atari Token would be essentially worthless. So let's begin by playing with that variable to visualise the ramifications of changing it.
Again, Minmaxing is the first tool to apply in order to establish whether there are any practical, real-world limits defining a token's range of values. Minimising the token's value to zero renders it entirely worthless, at which point the game's value proposition simply reverts to its original "Pay 25 cents for one game of Pong", since the "1 Atari Token" would no longer have any value.
At the other end of the spectrum however, things start to break really fast. By maximising the token's value it becomes clear that issuing an Atari Token would make it possible to instantly turn a highly profitable product such as "a game of Pong" into a highly loss making product.
Why is that? Well, it's because each Atari Token issued by Atari creates a future claim on the company's assets equal to the value of the token, and the total size of that claim would be the sum of all the tokens Atari issued. If an Atari Token was simply issued to a player each time they purchased a game, the total number of tokens would quickly rise, and without a defined rule to limit how many tokens could be issued, they could create a total liability that would render the business insolvent.
Clearly then, issuing any sort of token that has a defined monetary value attached to it (even a trivial one) has to be considered carefully because the potential effect on a business is so great if it should be misjudged.
The variables to consider here are as follows:
1) The value of a token;
2) The total number of tokens;
3) The conditions of token issuance, including rate of issuance;
4) The conditions of token redemption, including the rate of redemption;
These days the term 'Tokenomics' is generally used to describe the collective set of all variables within a tokenised economic system, such as a cryptocurrency. Balancing these variables to incentivise user engagement without placing too great a financial burden on the business is critical to making such a system work.
From the list above, it’s clear the biggest immediate risk for any games business intending to use a token system as an engagement tool within one of its games would be the value per token, total number of tokens to be issued, and the conditions of token issuance. These have the potential to create a catastrophic effect on the business if misjudged in a way that created a liability beyond the ability of the game or business to sustain. At that point it would be necessary for any tokenised game to draw on revenue sources outside of the game itself, such as capital reserves, investment revenues, or profits from other operations within the wider business.
But that's getting ahead of ourselves. Before considering the possibilities of extending a tokenisation model beyond the scope of an individual game, I first want to be sure I have a proper appreciation for all the factors that impact a simplified tokenisation model. Specifically, I want to be sure it is possible for token usage to be designed in such a way as to be sustainable entirely within the scope of an individual game.
To ensure this, another critical variable would need to be known in advance, and that is the total revenue generated by the game over any given time period. Only by knowing that, would it be possible to be sure the conditions for token issuance, value per token, and total token limit did not inadvertently create an unsustainable burden on the business.
On this basis, it would seem safest to determine the tokenomics for any game, only after the game has been fully developed, released, and is generating actual revenue from players within the marketplace. Doing so while a game was still in development and unreleased means all of its key performance indicators, such as its acquisition cost per customer, retention rate, average revenue per player, and total revenue per day/week/month/etc. would be entirely speculative, being based on assumptions that could well be proven wrong.
This is useful to consider, since it makes it obvious that the value chosen for the token has some very hard practical limits imposed on it. It should definitely not be arrived at lightly or arbitrarily, because the practical limits are ultimately governed by the amount of profit made by Atari from each game played. Therefore, assuming sustainability is the stated aim, then the first rule to check in any tokenomics design would be whether the value of the token (or tokens) issued to a player has the potential to be greater than the value of the revenue the player creates for the business within the transaction period.
Providing the value of the token(s) issued was less than the profit per game, Atari would still be operating a profitable business, but as soon as it exceeded the profit per game threshold then each game of Pong played would be costing Atari money. There are businesses that build this in as part of their business model, but these are beyond the scope of my exploration for the moment. For the purposes of my own research, I'm going to assume that Atari's business model required them to remain profitable purely as measured by selling game sessions to players.
Fortunately for our purposes, Pong's revenue per game is already explicitly defined at 25 cents. That means that the range of values available for the token are between zero and Atari's profit per game, which, for ease of consideration, we'll say is equal to the price per game at 25 cents. In reality it would be less than this.
With this foundational step understood and the total value limit for the token or tokens issued per game defined to have a limit equal to the revenue per game of 25 cents, the next problem becomes how best to issue the minimum value of tokens to create the maximum engagement boost for players.
I think it's worth taking a moment here to acknowledge that it would be possible to reframe this task to focus on 'incentivising player behaviours that maximise revenues for the business', rather than as maximising fun for the players. I actually tend to find that sort of clinical thinking useful for analysing Value Design myself, and may even use that sort of terminology when required sometimes. Ultimately, however, it is a means to an end for me as my focus is always from a perspective of creating better entertainment for players.
This highlights the perpetual tension between the two complementary, yet competitive elements at the heart of every commercial game - the decision about when to prioritise player needs versus when to prioritise the business needs. Both matter, since no commercial videogame can exist for long without adequately solving both the player and business needs. As I see it, keeping these often opposing forces in balance is the ultimate artistry of commercial videogame design.
However, there is no doubt that for every indie game purest that just wants to make games for the love of making games that players love, there is an equal and opposite number of predatory businesses that just want to extract the maximum revenue from their players for the pure profit of it, and who will happily deploy the sorts of addictive behavioural or psychological techniques most often associated with gambling in order to achieve their aim.
This phenomena is not new to the advent of Play-To-Earn however, and, to be clear, my interest is not in either of these extremes; though I acknowledge both exist and that predatory businesses often find more places to hide within the relatively complex business models of Free-To-Play and Play-To-Earn. My own interest is only in identifying how these business models can be used to balance creativity and commerce in healthy ways that generate the best entertainment experience possible in a commercially sustainable way. At the same time I will not move immediately to condemn new business models without due consideration because of entirely opportunistic or predatory behaviours of some market participants either.
With that said, back to the problem at hand: how best to create a Value Design for a tokenised version of Pong that uses the minimum value of tokens to create the maximum fun for Pong's players.
In the previous article we already identified the most obvious way to generate tokens within the game, which was winning. Within this system a token would be issued to the player upon winning their game, and in theory these tokens could be given zero monetary value since they would still have a reputational value associated with them due to only being issued for a victory.
At this point tokens essentially begin to act as an unofficial high-score table, since without any monetary value or other mechanism of exchange to remove tokens from the system the number of tokens simply records the number of matches won by individual players. The higher the number, the higher the number of matches won, and whoever has the most tokens has therefore won the most matches.
Introducing any sort of exchange mechanism would immediately change that however. In the first instance let's rule out giving the tokens an absolute monetary exchange value and instead give them a value that is only redeemable within the game. Then modify Pong's Product Design to make it possible for players to use 3 different bats - the standard White bat, a Silver bat, and a Gold bat that all function identically; the only difference is the colour. Lastly, make the Silver and Gold bats available to players who redeem 100 and 500 Atari Tokens respectively.
Even without providing a monetary exchange value for each Atari Token, the simple act of making it possible for players to access Silver and Gold bats with a defined amount of Atari Tokens immediately infers a real monetary value upon them. That's because players will perform a personal economic calculation to determine the value they would personally place on the Silver and Gold bats as soon as they become aware of them.
Then, since a token is only awarded upon winning a game it would be necessary to play 100 games and win every single one of them to earn enough tokens for the Silver bat. So the minimum cost of a Silver bat is immediately established as 100 x 25 cents, or $25. Likewise a Gold bat would require a minimum spend of 500 x 25 cents, or $125 to earn, and that is assuming a player wins every game. Realistically even the best players would be unlikely to win every game, and so it could be assumed that the total spend required to earn a Silver or Gold bat would be at least 20% higher than the minimum theoretical cost, so $30 and $150 respectively.
This, in turn, then implies a per-token market value of 30 cents or more. In other words, if a player wanted to use a Silver bat in their games of Pong and they weren't prepared to earn it by playing and winning games themselves, then their alternative would be to purchase the required number of Atari Tokens from a person or persons who had. The price they were willing to sell their Atari Tokens for would set the market price of Atari Tokens, which would fluctuate continually depending on various factors, such as how willing players were to part with their Atari Tokens.
Atari could increase demand for Atari Tokens further by introducing more in-game purchasable items than just the Silver and Gold bats. The more desireable items, the more demand for Atari Tokens, which incentivises more Atari Token production (through the playing of more games), and an increase in token price, particularly if the rate of token production was unable to keep pace with the demand for Atari Tokens.
That illustrates how the most basic tokenised economy could be created within a simple arcade game, such as Pong. The act of 'minting' an Atari Token in exchange for every game won and letting players exchange these tokens for the right to use a Silver or Gold bat in their game creates a market for Atari Tokens, which also establishes a real-world dollar value for the tokens.
It’s important to note that Free-To-Play games do most of this already. Many have elaborate shops that allow players to purchase all sorts of items with tokens they receive from playing the game. Some go further by establishing two or more different tokens with different issuance rules - specifically one that is issued by playing the game and another that is only issued in exchange for fiat currency. They then make different items only purchasable with certain tokens to ensure players have to spend fiat currency for certain items, although sometimes that can be avoided, if the game's business model uses incentivised advertising for example.
The only part Free-To-Play games usually avoid is the ability for players to transfer the tokens they earn between each other, since that adds a layer of additional complexity to the market, as shown above. In direct contrast, the Play-To-Earn model actively embraces that additional complexity and uses it to offer a different value proposition that is likely to appeal to a different demographic.
The last thing to note here is that, at first viewing, the price of the Atari Tokens may appear to contradict the earlier assertion that the value of tokens generated per game could not exceed the revenue generated per game. Well, it's worth digging into this in more detail at this point because, this applies if and only if the liability created by the token is redeemable via a 3rd party. This is what creates the danger for the business, since granting a token that is redeemable for 30 cents would certainly be foolhardy, because this would have to be met from capital reserves within the business if redeemed by a player. Likewise, granting a token that is redeemable for any asset the company doesn’t control, such as a hat, or a meal, would be equally dangerous, since the company has no control over the cost of that asset. Should the asset supplier unilaterally decide to double the cost of that asset, then the company’s liability immediately doubles along with it. However, by making the tokens redeemable only for the company's own digital in-game assets it makes it possible for the monetary value of the tokens to float freely on the open market without any risk to the business, since the redemption cost to the business remains independent of the monetary value determined by the market.
In this way, even if each Atari Token reached the value of $10.00 on the open market, whether due to players bidding up the price to become first to acquire a Silver or Gold bat, or speculators hording tokens, the actual redemption cost for Atari would remain exactly the same as at the time the Atari Token was issued. The additional value in that example is created entirely by external market demand, and so the additional liability cost is also borne by the market participants rather than the token issuer.
Therefore, as I see it, the simplest tokenisation model for Pong would be the issuing of 1 Atari Token in exchange for each win, coupled to the ability to redeem the tokens in exchange for the right to use some sort of visible status symbol within the game, and the ability for players and the public to own and trade the Atari Tokens between themselves.
While a Silver or Gold bat achieves this conceptually, it obviously wouldn’t be much use in practice due to the monochrome display used by Pong, but I’m happy to leave that issue unresolved for the purposes of this thought experiment. In closing then, and putting colour/monochrome VDU challenges to one side, I can see why having a single token redeemable for only a single object in a game might not be considered "Proper Tokenomics" by some people, due to its over-simplicity; however, that's a deliberate decision I’ve taken at this stage, since it allows me to double-check I understand each step of a tokenisation system from first principles, starting with in-game redemption options, to see what the implications are and where the limitations lie.
Having identified the fundamental source of real-world value within an over-simplified tokenised system, my next question would be, "is it possible to turn a single token, redeemable for a single object in a game, into a “Proper Token” within a “Proper Tokenomics” system through some type of operation, however simple or complex?"
If so, what is that operation, and at what point does “Proper Tokenomics” emerge? If not, then what are the fundamental differences that stop “Proper Tokenomics” emerging from such a simple tokenised system as the one I’ve described here?
Regardless of the outcome, I suspect it will uncover a useful tool for exploring the potential of tokenisation in all its forms. I think it would be hard to predict the effects of tokenisation on an economy at the macro level without truly understanding what's happening at the micro level. To me, understanding where the Value Design of tokenisation meets the Product Design of a game is a critical prerequisite to understanding the full potential of tokenisation within a Play-To-Earn business model.
And that's what this exercise is ultimately about for me: checking that I understand what's going on at the micro level first, so I can be sure any assertions I make about potential effects at a macro level later on are based on fundamentals that are well understood, rather than intuited.