Loot boxes and their predecessors have been a hot topic lately with the release of Shadow of War and Star Wars: Battlefront II. Loot boxes have been in video games since at least Team Fortress 2 and were likely inspired by collectable card games like Magic: The Gathering, so why all the fuss now? Shadow of War and Battlefront II are two highly anticipated games that are perceived to be particularly egregious or abusive implementations of this system which has resulted in a more general backlash against the practice. In some ways I think this is a useful discussion to have, as it makes us reflect on how games keep our attention and whether there are unintended consequences. However, I also think this discussion is driven by the fact that game developers are getting better at capturing some of the surplus that players enjoyed, and so misses the most interesting points to be made. In this, the first of two planned posts, I would like to talk about what loot boxes are, and what role (if any) they have in gaming.
What are loot boxes?
While not a loot box itself, the best introduction to the idea would be to consider a collectable card game (CCG) like Magic: The Gathering. Magic‘s innovation was to take the existing concept of trading cards (like baseball cards), and put them into the framework of a game. Decks could be augmented with cards that were randomly distributed in booster packs, retaining the collectable nature of trading cards, while also conferring benefits in the game itself. This feature is desirable for a publisher, since the sale of booster packs ensures a steady revenue stream from existing players. The value of this insight is easy to underestimate, and so it is worth reflecting on the fact that the instability of cash flows have brought down major game publishers (such as SPI and Avalon Hill) in the past, making Magic one of these strokes of genius that solves a legitimate business problem while spawning a wildly new popular genre.
Did consumers benefit? Yes. Magic didn’t exist before and so at worst consumers are indifferent to the offering. What about the booster pack element? Since Magic didn’t exist before, we don’t have the counterfactual of ‘Magic without boosters’, but the existence of Fantasy Flight Games’ Living Card Games (similar offerings that provide all the cards of a given release in one pack, removing the random element of deck building) and digital versions of Magic without booster packs have not diminished the popularity of the original card game. The randomness was a feature brought in from trading cards, and at the very least did not present an impediment to widespread adoption of the game.
17 years after the release of Magic, Valve released the MANN-Conomy update for Team Fortress 2 (TF2), implementing a loot crate system analogous to the purchase of booster packs. Crates are randomly granted while playing the game, and a key may be purchased in order to unlock the crate and receive the loot. The loot is randomly generated at the time of opening. When the MANN-Conomy update was released TF2 was just shy of 3 years old and still popular, no doubt due to the fairly regular free updates being pushed to the title. The problem with this business model is that the only revenues the game generates come from new copies of the game sold. Valve’s decision to continue updating the game gave existing players a reason to keep playing or pick the game back up, providing a reason for new players to still buy the game, but eventually there is a point where the additional units sold cease to justify the cost of adding additional content.
Adding loot crates allowed for additional revenues to be generated from existing users who opted to pay to open the crates. In this sense the introduction of loot crates was almost certainly a success as in less than a year TF2 became free-to-play with the Über update. There is a certain logic behind this decision as a multiplayer-only game like TF2 benefits from a robust player base, and so going free-to-play is self-reinforcing in that players enjoy the game with a larger player base and, consequently, the highest potential for revenue is achieved with more people generating random crates that some will later go on to buy. The success of this model need not be inferred from Valve’s decision to go free-to-play but through the simple observation that TF2 is 10 years old as of this month, continues to finance its updates, and has enough players to keep it roughly in the top 5 of Steam’s concurrent players (in fact, it is the only game in the top 10 not released this decade).
From this background we get an idea of what loot boxes are and why they may appeal to developers. In its simplest form a loot box is a microtransaction that confers some benefit inside the game. The exact nature of the benefit is randomly determined, and, depending on the game, can range from purely cosmetic to some mechanical advantage. The value of a loot box to the player is the probability weighted value of its contents. The value to a business is not just the revenue, but specifically a stream of revenue over time that had previously only been available to games with subscription models (i.e. MMOs).
At its core, loot boxes represent a pricing decision on the part of developers in terms of what content they release. An older and simpler version of this decision could be considered as how much of the game to release as a demo to incentivize a purchase. A more modern example might be to decide what features to leave to an expansion pack/DLC, or what content should be left to owners of a ‘deluxe edition’ of the game. Almost every one of these decisions is controversial among some gamers, but they really seem to be slightly more granular instances of a necessary conversation at the beginning of a project: How much game can we afford to build? How much can we charge for it?
I have an unpopular view on this topic that I’ll present up front: The people who make and sell games should be allowed to charge what they want for them. As a consumer of games I’d them to cost as little as possible, and I have been very vocal in instances where I feel the demands on the consumer have outweighed the benefits provided. Despite this, I do not think the developer’s own priorities should be ignored in this conversation. For instance, I am not crazy about the fact that the ARK developers released paid DLC while they were still in Early Access, but they were within their rights to do so. I also remember refreshing the Steam page when I saw the price increase before launch because I couldn’t believe it. Long time viewers can recall my incessant complaining whenever Paradox releases Crusader Kings II DLC in multiple parts (though, to their credit, the Stellaris releases have been exemplary). The point here is that, while I know my preferences as a consumer, I would not wish to compel these developers to price their games a certain way.
Games are not life saving medication. They are one of the least essential things in my, or anyone else’s, life. Every single one of us has looked at a game and passed over it because the asking price hasn’t matched the value we’ve assigned to it. In fact, we do this so often the act is unremarkable and we often don’t even register that we’ve assigned some value at all (and, as the likely pile of unplayed games in your library indicates, we are sophisticated enough to place a value on the option of playing a certain game at an arbitrary point in the future). The disappointment or injustice we feel when there is a price we don’t want to match is simply the recognition that there is some benefit we would get from playing a given game, but not enough to justify the expense (even if that is for essential reasons like choosing between Wolfenstein II or eating until next payday).
In a static sense, a change (increase) in the pricing strategy is a negative for the consumer: less content is available at the same price. Dynamically we can view it as a necessary evil to ensure that games continue to be developed and supported to the level we’ve become accustomed to.
Unlike direct price changes, loot boxes can shape the experience of a game. While we are accustomed to thinking about loot boxes and similar mechanics in a negative light, they can be a tool to affect the pace of a game and provide more value for a player. Consider a game like Rock Band. Most people seemed to dislike this feature, but the career mode required you to unlock some of your favourite songs, rather than just letting you jump straight into the hits. Jumping to the hits seems to be the default response to a game like Rock Band, but consider how you would feel about the product if you could actually do this. You’d play a handful of songs, ignore the ones you hadn’t heard about, and then wonder if you’d really gotten your money’s worth (especially if you started buying more hits off the store). I can’t say the songs in Rock Band changed my life, but I had a good time with the lesser known songs, and made an effort to perfect them. The design of the game made sure I didn’t spoil my dinner by loading up on dessert.
Here’s a thought experiment. Imagine the malicious intentions that would be ascribed to the developers of Hotline Miami if the game included microtransactions that would allow you to unlock a level without beating the preceding one (exact same game, just with the added microtransaction). “Yeah, what’s up with Assault man?” “They only included that level to sell more keys…” I think if Hotline Miami started with all the levels unlocked, most players would have finished the game, but have one or two levels they never completed. Some kind of paid unlock (either straight up unlocking a level or a power up) adds gradients between ‘let players choose and know they’ll short circuit the game’ and ‘you must master every punishing obstacle we put in your way no matter how long it takes.’ You can get a shortcut, but you have to pay up, and for some they will value their time less than their money. You can see this mechanism as a financial deterrent to ruining the full experience, but also enabling people who may not have a surplus of time to experience the whole game. The latter is an oft-cited rationalization for these kinds of microtransactions in games like Hearthstone where it is theoretically possible to grind for all the cards and solo adventures but impractical.
Ironically, loot boxes also better align a consumer’s interests with the developer. If the purpose of these microtransactions is to establish a long tail of revenues, then it means that the financial success of the product is directly tied to the sustained enjoyment of the game and the value of the additional content bought while inside the game. In English, the game has to be good and you have to actually want the stuff they’re selling you. While quality has always been a factor in the financial success of a game, there is a known problem of game endings being rushed or generally unsatisfying due to the fact it’s the content players are least likely to experience. With a 2 hour window of playtime for returns on Steam, there is also an incentive to front load a game to keep people engaged for at least that long. This article began with the story of TF2 and the idea of financing fresh content through loot boxes which is a case study of Valve’s interests and players aligning through this mechanism.
These are the positive cases for how loot box style systems can help the design of a game. If we generalize even further we can say that elements of this mechanic have existed in RPGs ever since people started randomizing loot (Diablo is essentially paying up front for an infinite supply of monster shaped loot boxes). As with any mechanic in a game, it can be done well or it can be done poorly. Games that implement these mechanics can break the flow of an experience by constantly asking for another payout. Nobody wants to play a Hobbit game where Gandalf shows up and says “We’re going to go on an adventure! But first, all I need is your credit card number…” Creating an imbalance due to the presence of paid content is also a potential problem, crystallized in the dreaded “Pay to win!” epithet.
I have dedicated more time to the potential benefits of this system simply because I think the negative case is so well represented that it is essentially taken for granted. When considering loot boxes and microtransactions in general I find it more edifying to consider them as a tool first (i.e. not to assign a value judgement to them) and consider their fit as a solution to a problem with the attendant costs and benefits.
Paying for games
In the simplest and bluntest terms, people do not want to pay more for games. More precisely, people do not want to feel like they are paying more for games. In truth, games have never been less expensive from a consumer’s perspective. Here is a chart adapted from an Ars Technica article:
The nominal price is the price you would see reported in a catalogue from that year, but inflation can make a direct comparison difficult as the purchasing power of a dollar in 1977 is not the same as a dollar in 2017. This chart presents the prices for games in real terms, that is, adjusted for inflation. The blue line reports the upper bound of prices while the red line reports the lower bound of prices (as reported by Ars Technica). There are some complaints with this graph. I grouped cartridges in with discs which partially explains a decrease in prices around the year 2000, and Ars Technica simply reported the high and low values in 2013 as $59.99, while previous years would include a game like EyePet along side Call of Duty: Black Ops. Reporting ranges of prices have only become more difficult over time, as there are legitimate questions as to whether or not it is fair to include a game like Hellblade: Senua’s Sacrifice beside this year’s iteration of Battlefield Creed: Modern Warfare. Furthermore, as the EyePet example illustrates, we’re only taking into account someone’s reported prices, not any sale prices or the composition of how many games at a given price people bought (so, if 1/10 of gamers bought games like EyePet and the remaining 9/10 bought games like Call of Duty: Black Ops, the upper bound makes more sense to use). Finally, the product itself has changed over time. We simply expect more from a AAA game today, than we would expect from a game like Super Mario Bros. 3 (estimated nominal price in 1990: $50). We could try to compare a game like Shovel Knight, which is heavily inspired by NES games, and surmise that the cost has gone down substantially, but this would fail to acknowledge that Shovel Knight has the benefit of decades of game development and fewer hardware restrictions that NES developers would face.
These objections aside, the price for a given game is, at worst, essentially flat since 1977, and most likely has declined in real terms before we account for quality improvements (better graphics, sensible mechanics, new genres, voice acting, online multiplayer). In fact in the Ars Technica data set, blockbuster games seem to hit $59.99 (US) in 2006 and stay there. While there is a consensus that the cost of making games is going up, I was less successful at finding representative data for the cost of making games, but I have two rough and somewhat intuitive measures. The ESA in Canada publishes an annual essential facts report which appears to include average cost every 2 years across several platforms and genres (I reached out to the US equivalent but the e-mail address provided for information and historical reports did not work). The earliest available date for Canada is 2011 and lists the average cost of console development as $10,083,000, while the average cost for the same platform in the 2017 report is $12,536,957, an increase of about 24%. Now there are lower and higher average costs presented over the years, and Canadian game development may not be representative of broader industry trends (for instance, the highest cost in the 2011 report was apparently $30m while this list has three games with $50m production budgets for the same year). If we want to consider the upper bound, the only game since then which has cost less (in nominal terms) was The Witcher 3: Wild Hunt. It’s sensible for there to be a lot of variance between projects which are going to be at different stages over time, but the overall trend certainly seems to be increasing. We could also consider an important input cost into games: wages. The same reports list average salary as $62,000 in 2011 and $77,300 in 2017 an increase of about 25% from the 2011 level (these are not adjusted for inflation). Now again, these are mostly Canadian numbers, they are not directly measuring the costs for the same products as our measure of prices, and they are only doing so for half of the periods observed. Still, these numbers suggest that our intuitions about the costs of making games are correct, games are becoming more expensive to make and you’re either hiring more people to make them (though this is harder to support with what we see here) or that you are paying up for the know-how that brings your game to market (likely true given what we see in salaries), and that this has been the case for at least half the time the price stopped increasing.
Price increases are not the only way to deal with rising costs. Games are relatively inexpensive to reproduce and selling more can help cover the added expense. There are certainly more people buying games, and this is likely due in part to the investments into making them more visually appealing and easy to play. But it’s not at all clear to me that the net increase in players is enough to keep pace with the rising costs of production, especially as long term gamers have lower cost options tailored towards a more ‘hardcore’ audience. This returns us to our original topic: stagnant prices for the base game are being augmented through the partitioning of the full product into the base game and supplemental components ranging from expansion-style DLC to microtransactions like loot boxes.
This kind of behaviour is not limited to gaming. I’m sure everyone has experienced the phenomena of buying something at a store and noticing the size is shrinking over time. You may have a favourite chocolate bar and, had you kept the wrappers (please say you don’t) you might notice it shrinking by a few grams every once in a while even though the price usually stays the same. Soft drinks are similar in that bottles will shrink in size, and in both cases the price per gram or millilitre will have gone up by a small amount. Nobody likes higher prices, but most readers will likely find these examples rather unremarkable. The truth is that while we value transparency, as consumers we tend to punish companies that transparently increase prices and so have given them every reason to play this little game where things are repackaged and the old size is reintroduced after enough time has passed. It is not surprising to me that we now see similar practices in gaming seeing as it does not seem to take much to make a very vocal group of gamers feel slighted and start posting manifestos in all capital letters in reviews.
I don’t think it’s wrong for gamers to advocate for their interests as consumers, but I think this conversation is most productive when the demands of consumers acknowledge the costs. Here’s a demand without accounting for costs: “I want the graphics to be like Battlefront II, it should have a single player campaign with the scope of Witcher 3, multiplayer should be populous and regularly updated and balanced, and if there is DLC it should transform the game to a level of XCOM: Enemy Within or War of the Chosen. Also, it should be free to play.” There are two ways this request fails to acknowledge costs. First, the scope for a project like this is immense, and requests like a sprawling single player campaign potentially contradicts the request for an active multiplayer community. But conceivably some brilliant design army could concoct a satisfying product that appeals to enough people to balance the demands and delivers a truly awe inspiring product. What is likely meant by ‘should be free to play’ is some form of monetization that involves cosmetics instead of denying the player access to meaningful portions of the advertised experience. This is an easier case to dismiss because it seems so unreasonable on the face of it, and yet this is often the kind of demand we make when we talk about a game’s monetization strategy. The second way it fails to account for cost is how the developer is likely to monetize the game. If we take the request as mandatory, then the only option left to a project like this is to use every trick in the book to extract as much money as possible from the player base through paywalls, loot boxes, experience boosters, telepathic data wizards, and hired goons. The cost here comes in the form of how the game will be shaped to extract the payment we do not want to give up in the first place.
A segmented market
The thing that struck me the most about the complaints about Battlefront II were that the consensus seemed to be that the game was very good and that the complaint was that the loot box system was the line in the sand because it would open the floodgates. It’s hard to see this complaint as any more than “I really want this game but I don’t want to pay for it.” What I think this marks is a clearer split in terms of the types of games that people buy and play.
It occurred to me that I selected out of big new AAA releases some time ago. This is not because I don’t like them, but rather than I have so many opportunities to play other things that it is not a good use of my money to buy a $79.99 (CDN) game when I will receive comparable enjoyment from an older game in my library or a lower cost independent game. I am quite fond of indie games of all shapes and sizes and their price is usually less than half of what the premium (AAA) offerings are. The quality of indies is increasing all the time, and so I think it is worthwhile for players to ask “knowing that I’ll probably wind up buying a few loot boxes, do I think this game is worth it?” and I think a number of people can honestly answer no and still be perfectly happy with the games they play.
Premium games almost seem to be taking an amusement park approach (or Costco or Amazon Prime). You pay an entry fee to come in and have a particular experience, but extras (cotton candy, popcorn, certain rides) come at a price. Some amusement parks, or Costco or Amazon Prime essentially make you pay to shop there, but the service and prices make this fee worthwhile. Sure, you can not have a snack or buy a souvenir t-shirt or any of these other things that seem to give people enjoyment, but you may be missing out on part of the experience and may resent the additional expense on top of the admission ticket. Whether it’s a game, an amusement park, or a store membership my advice would be the same: If you’re not getting the value, don’t pay for it!
It seems to me that games like Battlefront II and other ones that are supported through loot boxes are really intended to be the kind of game that the player mainlines for the majority of their play experience. I personally like a lot of variety and so have only a few games that I really go in deep with. Battlefront II does not seem to be one of those games, and so while I have tremendous respect for the achievement of bringing the Star Wars universe to life in such a vivid and exciting form, I have to say “this game just isn’t worth it for me.” I will, instead, support a game like Cultist Simulator because it looks interesting, it’s by a game designer whose work I have liked in the past, and the price point is very appealing. In fact, I’ve already Kickstarted it and so I have said “I value this” and have given the clearest possible signal that I would like to see more projects like this in the future. By selecting out of the top end of gaming, I also indicate that I am willing to accept limits on what the cutting edge looks like.
If gamers as a whole are willing to pay up for cutting edge experiences, then it is likely that teams and budgets will continue to expand in order to keep up with the arms race. If players are unwilling to pay directly, then part of that arms race will involve innovations in getting players to pay for these increased costs indirectly. If gamers are unwilling to pay at all (i.e. they select out as I have), then we will see budgets and the style of games made shift into something that is more sustainable. I think it’s unlikely that we will see big publishers like EA reverse course in terms of their premium products, but I also think that publishers like EA and Ubisoft are fully aware that there is a segment of their potential audience that wants smaller more impactful games with the associated reduction in price tags. Consider the EA Originals series, or games like Valiant Hearts (Ubisoft) which are high quality single player (mostly) experiences that are not intended to be something you put thousands of hours into. I happen to like this state of affairs, since it means that people who are looking for a game like Battlefront II can get it, and the profits go towards developing the kind of game that I see value in. What I can’t really stomach is the idea of looking at a premium game (which tends to imply a heavy degree of crunch or at the very least substantial effort on the part of the developers), wanting it, but turning around and saying to the creators “But I only think it’s worth 75% of what you put into it.”
The tent for gaming has gotten bigger, and this has meant that the ceiling for cost/quality has gone up along with the tools becoming more accessible and better for designers at the low and mid ranges. I cannot see how a greater variety of games being made available to us is a bad thing, even if it does mean that I am no longer consuming the bleeding edge of this market. It is hard for me to see people being priced beyond what they want to pay as being particularly outrageous given that games are so inessential and there is so much great stuff out there for very reasonable prices. If the complaint truly is about the way monetization strategies affect gameplay, then the decision seems even simpler. If the game has been reduced to ‘pay to win’ then the matchmaking will either be effective at balancing the match or it will fail. If the nagging to unlock content breaks your immersion in the narrative, this is a failure in the design of the game. If it fails, then it seems that this is no different from any other game with a defect, and it has never been easier to become informed about the quality of a game after release. In either case, we should make time and money available for the things we value, not the things we don’t.
What about gambling?
This post has focused more on the microtransaction aspect of loot boxes, since the majority of arguments surrounding addiction and gambling seem to be supporting the more fundamental position of “I don’t want to pay that much for this experience.” I think it’s important to understand why we’ve gotten to this point and why the people making games use these kinds of systems, but at some point we need to discuss whether or not we have any business implementing them and what the consequences are. I happen to think this is a much bigger issue than what most people want to talk about, and so, having discussed loot boxes as microtransactions, the next post will discuss the psychological aspects of loot boxes, whether it constitutes gambling, and whether such a system has any place in the design of games.
Note on affiliate links: I have an affiliate status with GOG.com for which I am given a portion of sales for traffic I drive to the site. The inclusion of a given title is for illustrative purposes first, but when it is available on GOG I will provide such a link. Naturally I encourage you to do your own price comparison or buy on your preferred platform. I include, on occasion, affiliate links from other broadcasters to support people who helped me in a given post or the cast in general. I have also received press copies of Stellaris and Crusader Kings II DLC.