Insert Coin to Continue: An Argument Against the “All Games Will Inevitably Go Mobile” Predictions
This post was originally published at Gamasutra on October 25th, 2013. At the recommendations of a friend, I have decided to re-post it over here in my own page.
There have been a few articles recently predicting the doom of the game industry because of statistical analyses. Of these, the more commonly known pieces are ‘Game Over’ and ‘Third to a Billion’, both published at ASYMCO by Horace Dediu. Through his analyses, he concludes that the videogame industry is dying. This is, of course, not a novel or revolutionary claim to make. Whether the culprits are poor design choices, rehashed title lineups, or a flood of boring games coupled with gamers getting old, many commentators [some more eloquent and well versed than others] seem to agree that, in one way or another, “games are dying”. Andrew Groen at Penny Arcade argues that the industry is not dying, that it’s simply changing. In his piece, he talks about how slower growth does not mean death, and I agree. He also makes the argument that gaming is shifting to a more mobile-friendly market and refining itself to do away with high-priced AA titles and turn them instead into AAA experiences. Others still hope that the industry will shift from a console-centric market to a PC-centric one through Steam and their peripherals. All of these arguments have merit and they address Dediu’s charts presented in his ‘Third to a Billion’ piece on how android devices are dominating sales charts over traditional consoles. This is of no surprise, really. Videogame consoles are released on an 8 to 10 year cycle and are fairly task-specific. A Nintendo Wii, for example, only plays Wii games, and the term “Nintendo Wii” refers to only one piece of hardware: the Nintendo Wii. An “Android Device”, however, is far more than just “one thing”. My phone, which I update every five or so years, is an android device. My sister’s phone, which she updates every other month, it seems, is an Android device. The HTC One, the LG Optimus, the Motorola Droid, and the Samsung Galaxy are all “Android Devices”. Furthermore, the Galaxy Tab, the Barns and Noble Nook, the Kindle Fire, the Google Nexus, and the Asus Transformer are all “Android devices”. Heck, even some car PCs, such as the BMW E46 Android CarPC, are “android devices”. And so, it’s no surprise that when Dediu juxtaposes game console adoption rates to the adoption rates of “Android devices”, more commonly known as ‘smartphones and tablets’, that the artifact that is considered a necessity in our culture – one that many people buy a new one every year – would have the superior numbers. Smartphones and tablets are, after all, multipurpose devices that double as gaming platforms and not gaming platforms into themselves. Still, deceptive as his implications may be, his analysis and overall assertion that “Android devices” sell more than game consoles is undeniable, and I have no contention with it.
What I do have a problem with is his assertion in the ‘Game Over’ piece that the videogame industry will die unless it moves entirely to the smartphone market. In his concluding comments he makes this assertion clear:
“The implications are that Nintendo, Sony and Microsoft are beyond the point of no return in this industry. Gaming, as a business, cannot be sustained as a platform independent of a general purpose computer. Like other “applications” that used to have systems built around them conforming to their needs the dedicated-purpose solutions came to be absorbed into the general-purpose platforms. And the modern general purpose computer is the smartphone.”
That seems like a reasonable enough assertion to make going by his numbers. Smartphones sell more than consoles – therefore, Nintendo should only make games for smartphones. He further shows a steady decline in the sales of consoles over time. In his charts, he shows that hardware sales of all companies – both portable and static – have a certain number of consoles sold at release quarter. The number of units sold per quarter steadily increases up to a climax, and then it steadily begins to drop. Every quarter from the climax point on will find a lower number of units sold per quarter than the previous. This trend, he notes, is true of the Wii U, the Wii, the Nintendo DS and 3DS families, the Gamecube, and Sony and Microsoft consoles. The numbers presented are, as far as I can tell, accurate. However, they do not spell the doom of gaming, and the reason is a simple one.
For any given product, there are three types of customers: the early adopter, the mid-level adopter, and the late adopter. Whenever a product is release, a fraction of the total potential customer base will buy said product. During the next few years, more and more people will adopt said product. During the product’s later stages, late adopters and those seeking to replace a broken product will make a purchase. The most common trend is that the mid-range adopters outnumber the early adopters as well as the late adopters. In essence, the principle here proposed posits that on year one, there will be X number of sales. For any given constant period, in comparison to any other similar amount of time, the number of artifacts sold will be a number higher than X, or X = X+Y. After public interest in the item begins to fade, sales will drop. This will be reflected as X = X – Z. This will continue until X = 0. The mathematical function for this is as follows:
This formula is by no means new. It is a concept known as the Gaussian Standard Distribution – a commonly occurring continuous probability distribution that tells the probability of a number falling between any two real numbers. Its applications are broad, and it can be used to analyze grade frequency in education, to figure out which card is more likely to be drawn from a deck, or, I claim, to analyze sales data. If my suggestion that sales data for any given product, due to consumer behavior, adheres to this mathematical principal, then it means that every product should conform to some form of a bell curve, some with more deviation than others, depending on consumer interest. In essence, a sales chart for any given product will look like some version of this:
The figure above represents hypothetical sales data by millions spread throughout several years. The figure to the left represents a standard curve with no deviation, the chart on the right represents a curve with some deviation.
A hypothetical narrative example of this principle would be as follows:
- A corporation announces a product. Public interest rises.
- Over the lifetime of the product, 2,000,000 people will buy the product.
- During the first year, 200,000 units are sold. These are the early adopters. They acquire the product because of an interesting feature or due to brand loyalty. Those who don’t buy the product wait to see if there are any major complaints or problems with the product.
- During the third and fourth year, 300,000 and 400,000 units are sold respectively. People have decided to “jump on the bandwagon”. The product has obtained good reviews and there is good word of mouth, so the mid-adopters buy the product.
- During the fifth and sixth year, 300,000 and 200,000 units are sold. Interest in the product wanes – not because it’s a bad product, but because many of those who would want the product already have one.
- During the seventh, eighth, ninth, and tenth year, 150,000, 100,000, 100,000, and 50,000 units are sold respectively. These late adopters buy the product for a number of reasons. Perhaps the product was updated to include a new service, or the price of the product dropped to a price that these customers was worth the price of admission.
- The remaining 200,000 units are sold during the following five years at a rate of 40,000 per year. These are largely people who “missed out” or people looking to replace a broken unit.
Regarding the cumulative sales of a product, graphs should show a sudden increase in the number of total units sold during the first part of the life of the product, then a slow in the increase of total sale of units, leading to a point where sales are no longer recorded. The mathematical principle is that X being total of units sold and Y being units sold in the current period, then at the end of every period X = X + Y. The hypothetical narrative example of this principle is as follows:
- On the first year we sold 20 units.
- On the second year we sold 35 units, for a total of 55 units sold.
- On the third year we sold 50 units, for a total of 105 units sold.
- On the fourth year we sold 95 units, for a total of 95 units sold.
- On the fifth year we sold 20 units, for a total of 220 units sold.
- On the sixth year, we sold 2 units, for a total of 222 units sold.
- On the consecutive years we did not sell any units. Our unit sales total is 222.
Under this principle, any given cumulative sales chart should reflect a version of the following:
I would argue that these projections hold true to any non-essential non-renewable product. Food, for example, is an essential need and is, therefore, mostly exempt from this principle. However, everything from phone models to car models to, yes, videogame consoles, I would confidently state, follow some version of the principles explained above.
Certainly, it may be wonderful to suggest, as Dediu does, that “Android devices” are in an ever increasing trend, but if one looks at the individual products (the Galaxy Note II as a separate product from the Galaxy 4, for example) and not at the broadest possible labels of “Android devices” or “Smartphones”, the patterns above will no doubt reveal themselves.
By now, the astute reader has no doubt noticed that this is mostly speculation and that the examples are hypothetical. Below I have compiled a short collection of sales charts from various perishable products where patterns similar to those here explained here are proven true.
On the chart above, we see the increase in unit sales of the iPod going through the predicted growth and leveling off.
Above we see sales data as presented in Macworld for the iPhone. As predicted, there is an increase, a peak, and a decrease in the number of units sold. Sales trend by year also show similar patterns – at the beginning of a cycle, new units are sold. In subsequent quarters, less units are progressively sold, until a new cycle begins. The same is true of consoles. The only difference is that the life cycle of phones is shorter (12 – 18 months) than that of consoles (8 – 10 years), and that phones are marketed as essential, while consoles are marketed as entertainment.
Even Dediu’s graphs show similar trends to what is here described. Yet, If we are to believe Dediu, videogames are dying because they are selling less consoles as time progresses. Certainly, the Wii is dying. The Playstation 3 is dying. The X Box 360 is dying. This is because we have reached the end of a console cycle, and this is only the natural process of consoles. The public shows an initial interest, the interest will increase, then, as the public adopts the technology, less units will be sold per quarter.
Let us consider the chart below, originally published in the Journal of Technology Management and Innovation, and prepared by Paulo A. Zawislak, Fabiano Larentis, Cássio B. Machado, and Alexandre M. Andrade.
This chart shows the life of various consoles. In all of them we see similar trends – there is an initial consumer base, more people adopt the system in the mid years, and in the later years console sales per year decrease. Consider the Super Nintendo system. Its yearly sales peaked in 1992, and since then yearly sales declined. As more people adopted the console, there were less people who owned the console and, therefore, less people to buy it, as they already owned one. On 1996, the Nintendo 64 came out with the purpose of shifting users’ attentions from the SNES to the new Nintendo console. However, it didn’t work, as many of the SNES loyalists decided to opt for the Sony Playstation. Now consider the Sony Playstation. Its yearly sales began to drop on 1997. On the year 2000, Sony released a new product, which shifted the public’s interest to their new console. The continuation to this graph can be seen in Dediu’s ‘Game Over’ article.
This is the life cycle of a console – consoles are released, people become interested in them and buy them. When everyone who is going to buy a console has one, companies release a new console.
In his piece ‘Game Over’, Dediu makes the argument that the 3DS and the Wii U kicked off the 8th generation, and that their release failed to re-ignite growth. I would blame this on Nintendo’s poor marketing and branding strategies – most non gamers I know think that the Wii U is nothing more than a U Draw tablet and that the 3DS is just a bigger version of the DS. But that’s something we can discuss in another post. The fact is that the clearly not Playstation 3, the Playstation 4, is due soon, as is the X Box One. The PS 4 already has preorders reaching the 1.5 million dollar mark in the U.S. alone.
Dediu’s prediction is that the consoles will fail and that ultimately all gaming will be done through smartphones. My prediction is that the PS4 and the X-1 will sell well on release and on the following two to three years, peaking at 3 – 4 million units, after which they will sell progressively less consoles. What do I base my prediction on? Sales numbers and trends that I have observed in every single console generation since the NES. This doesn’t mean that I am entirely in the right. Perhaps Dediu is correct. Perhaps, what seems to me as someone who has a stake in the mobile market projecting his wishful thinking of having Nintendo games be exclusive to smartphones into his predictions is, in reality, a visionary who has successfully unlocked the mysteries of consumerist whim and has thrown in his lot with the winning side. It is true, after all, that smartphones are here for the long run, and if I am to be honest, in the long run, I do see all aspects of computing – perhaps even design – being done in a single device (although I see this all-purpose computing device as something closer to a Microsoft Surface-like device with phone connectivity rather than the traditional smartphone as we know it). However, I don’t think that major videogame companies will shift their models from a console-based market to a mobile-based one. And the reason for this is the way that the market for mobile gaming currently resembles something closer to the state of gaming thirty years ago just before the Great Crash than it does a stable market.
I have no doubt that, at least for two more console generations, Nintendo, Sony, and Microsoft will venture into the mobile market – sure. A few major companies such as Sega and Square Enix have already done as much. However, I don’t see a complete shift, as mobile games, regardless of quality or merit, are seen as inferior. Despite the higher resolution, higher quality, and better writing found in the iOS version of Final Fantasy Tactics, for example, the game was considered by many commentators as being expensive, surprisingly expensive, inflated, or downright criminal. The same authors remark on Square’s pricing scheme for iOS exclusive games such as Final Fantasy Dimensions. Likely because of this pricing scheme, none of their games are currently in the top 200 grossing games lists. However, customers feel comfortable paying 60$ for a new game of reputable quality. This, I think, is not something that the gaming industry is likely to give up.
Will the gaming industry diversity? Absolutely. There are currently mobile versions of several games developed for consoles, and many companies famous for their console games are creating mobile-only games. Sonic Dash comes to mind. Will the industry give up consoles? Given the type of games best suited to the audience of mobile games, the current trends in pricing, and the diversity between mobile and console markets, I find that unlikely.
Sources Not Cited in Text: