Thursday 19 September 2013

Better MtG Online

(Massively updated on 19 October 2013. Bonus: better looking version on Medium.)

Suggestion for redesign of online Magic: the Gathering to better fit the digital medium and widen the audience

The aim is to replace Duels of the Planeswalkers and Magic Online with a unified system superior to both. The DotP programming base is assumed.

#Core Proposals
#Reasoning and Additional Notes
#Conversion of Excess Cards
#Mechanical Nuances
#Random Awards Mechanics
#Grid of Mechanics
#Half-Constructed for Newbies
#MtG on Mobile
#Culture of MtG and Sustainability of Digital
#What's in This for Wizards of the Coast?


I'm convinced Magic: the Gathering can benefit from changing its approach to digital distribution. It has the potential to attract a lot more players, become more welcoming to newcomers, unify "entry" digital Magic (DotP) and "classic" digital Magic in one system, and prepare a platform for expansion into mobile space, while increasing profits. I've composed a holistic proposal of how all systems of online Magic could be changed to do this. Here it is. I tried to think through every detail, so there's a lot of information crammed in here. I apologize for the volume, and I ask to give it some thought when reading. It's a lot to take it, but it makes sense when you understand the whole picture. Please absorb sequentially.

#Core Proposals

(#Corollaries are equally important, see below)
  • Philosophical shift from mimicking paper Magic to making full use of digital possibilities.
  • Players can purchase the game once and unlock the rest of Magic by playing. Repeat payment is required to gain access to new blocks or sets. The old is never taken away.
  • Cards are randomly awarded for game matches. See #random and #conversion.
  • Players are offered to purchase boosters and pre-constructed decks to speed up collection.
  • All features of MtG are available in this game, including constructed deck building and old formats. The breadth of supported sets and formats mirrors MO.
  • Players can log in with an MO account, get to keep their MO collection. It might be one-time sync, or voluntarily repeatable sync.


(See #Reasoning below)
  • Because of the constant influx of free cards, card exchange and trading has to be removed. Otherwise Wizards' merchandise would be continuously devalued, eventually into obsolescence.
  • Due to the above, because players can no longer shed excessive cards, cards become capped at 4 per account (basic lands unlimited). It's a shift from tradable collectibles to unlockables.
  • Excess cards in existing MO collections are converted into random replacements. Mechanics of #conversion see down below.
  • Players can unlock cards in non-Standard sets, both as rewards for games, and by buying boosters and decks. Limited is also supported for those sets.
  • Limited is changed to no longer require boosters. The game itself provides random card pools. Conversely, players don't get to keep those cards. The format is available for all supported blocks.

#Reasoning and Additional Notes

1. Why remove card exchange.

The constant influx of "free" cards into the market would cause continuous lowering of prices, leading to a fast decline in purchases of boosters and decks. Eventually, Wizards' merchandise would be devalued into obsolescence. To keep microtransactions flowing, card value needs to be stable, and this will only be possible if they never change hands. Also, doing away with card exchange confers additional benefits:
  • It makes acquisition almost completely random (see #mechanical and #random), slowing down abuse of potent cards and copycatting of popular decks. Helps variety and prolongs the metagame.
  • It leads us (see next point) to opening up card acquisition in non-Standard sets, both through rewards for games, and through boosters and decks from Wizards. Players can now pay us for decades' worth of Magic.

2. Why include old formats.
  • As a consequence of removal of card exchange, we have to open up acquisition of cards in old sets. Out-of-print cards can no longer change hands, but we still want players to be able to get them. Because cards now only come from Wizards, and because we're relaxing print mimicking, the natural solution is everything for sale.
  • The entire history of Magic, this humongous stockpile of mechanics and flavor, is too rich a resource to be relegated to hardcore veterans or people willing to scour the market for select old cards. With these changes, the playerbase able to experience old content will be growing rather than dwindling. The promise to never reprint reserved cards doesn't apply fully, because acquisition is now different.
  • Thanks to this and to the changes to Limited (see below), those unwilling or incapable of heavy preliminary investment of time or money will also have a way to play with old content, to cast a wide look at it.
  • This gives room for growth to overachievers who could otherwise cap out in Standard and have nowhere to expand to.

3. Why change Limited.

Despite its origin as a marriage between gameplay and card acquisition, born of limitations of paper Magic (boosters as the only vehicle for set-wide randomness), and its role as a stimulator for booster purchases, Limited has become its own format, unlike anything else in Magic. It provides a different experience: 1) forced deck building on-site from a random pool of cards; 2) different rarity distribution from Constructed; 3) different power level distribution and card variety from Constructed. This is important for further points.
  • In paper Magic, usage of boosters is what enables Limited and makes it fundamentally different, by providing a pool of cards randomized from the whole set. That's why boosters are required. In digital Magic, there is no such technical necessity: the game itself can provide random cards. We can choose to drop this requirement when we're not mimicking printed Magic.
  • Due to the relaxed availability of cards in the rest of the game, booster-based Limited would become a fringe format, ignored by most of the players we're aiming to attract. This would be wasted potential, wasted gameplay. It should open up to provide unique and different experience and thus increase player engagement.
  • If Limited, or some version of it, is accessible only through payment, this would gate some unique content behind a paywall. This is against the core changes, which are aimed at opening up access to full Magic through a limited, pre-set payment (one-off or subscription). Furthermore, it might be seen by players as a transparent ploy, producing distaste.
Taking the open route results in increased player engagement and player retention, and allows to elegantly solve some additional problems:
  • Problem #1: Exploration of old content. We want players to have an easy way to explore old Magic content. In line with the general design, there needs to be a "free" way with an option to pay to quicken the process. Players get some starter decks for every block, and they can earn more cards by playing. We also offer to buy pre-constructed decks and boosters in every set. The problem here is that there's SO many sets and cards, that given the intentionally slow acquisition rate, exploration would be too sluggish and players might lose interest. Increasing rate of card acquisition in old sets is also dangerous because this could distract many players out of Standard by enticing them with bigger rewards in the old sets, and we want players to be focused on Standard. So we have a problem here, and Limited can provide a good solution. By playing Limited in old blocks, players can quickly get exposed to huge breadth of cards, art, mechanics, that they would otherwise never experience due to unrealistic time requirements. Of course, brief possession of cards is not the same feel as having them in your collection, so to give players a longer lasting sense of those cards, we should include in the UI a special section for cards that the player had used in Limited games. Display them dimmed.
  • Problem #2: Fairer environment for newer players. The free Limited format becomes a platform for players to face off in an environment where skill and luck matter more than achievement. Lack of cost allows new players to participate without fear of being crushed by advanced decks, making the matchmaking more relaxed.
As a result of these changes, it becomes unnecessary to keep the "old" Limited format around. Most people would opt to avoid microtransactions and play free Limited, making this format unpopular. Playing a free Limited match and buying a bunch of boosters would provide the same number of cards. The old Limited format used to serve roughly the same function (random or half-random card acquisition as a result of a match) that is now core to the game, making it somewhat redundant overall.

4. Additional notes.

This set of changes allows us to do something new: errata more than just rules text on cards, but things like mana cost or other numbers. It couldn't be done before, but it's possible now. Could become a very last resort instead of banning the card. Reasons why possible now:
  • Digital medium and server-centric approach. Couldn't be done in paper.
  • A shift away from mimicking printed Magic, a necessary change in philosophy.
  • Financial reasons. Players invest in cards based on the assumption they stay as-is. This targeted investment prevented serious errata. However, with the shift to mostly random acquisition, players no longer invest in particular cards, and thus, won't get an investment devalued if a card they previously got randomly is nerfed.
  • This should be a very rarely used measure, the last resort, and only be considered after deciding to otherwise ban the card.
Bonus idea regarding old sets and multiple card editions: make all editions of each card available not outright when unlocking that card, but be unlocked on per-set basis by playing that particular set, with previews available. Make cards remember art choice, give it a sense of permanence.

When purchasing the game, players automatically receive some starter decks in every block (set?) they have access to. This includes old formats, which allows people to play block Constructed in old blocks and spurs the exploration of ancient content. For each block, the players receive several decks that nevertheless don't cover the whole color wheel. This limitation allows them to discover something new about each block on their own, and may somewhat entice them to purchase pre-constructed decks from Wizards.

#Conversion of Excess Cards

Conversion occurs when player has more than 4 copies of a card. This primarily happens when transferring Magic Online possessions. It also safeguards against possible #random awards bugs.

Replacement comes from the same block as original. We may give players a choice to prioritize format over block. Conversion takes into account card color, converted mana cost, and what the card does. Each card is assigned a position on the mechanics #grid (see below). Replacement must come from the same position.

  • Format
  • Card position on the color / class #grid
  • Rarity
  • Keywords or other mechanics
  • CMC
Rarity can be overridden. In this case, it's still respected and more cards are given: 1r = 2u = 4c, 1mr = 1r + 1u.

#Mechanical Nuances

  • When MO possessions are transferred to new accounts, excess cards (over 4) are #converted into random cards.
  • From the start, players are given some randomly chosen starter decks in every supported block or set.
  • Booster packs and pre-constructed decks are available for all sets, not only those in Standard.
  • Pre-constructed decks are the only source of non-random cards, and each can only be acquired once.
  • Pre-constructed decks come with additional bonuses besides the cards inside. This helps make them more meaningful than just being a bundle of card unlocks. The bonuses aren't power related, and mustn't be too stackable. Perhaps card acquisition improvements like a chance for bonus card when unlocking cards in this block.
  • We may consider an alternative method of rewards: instead of cards, award game credits that can be used to purchase booster packs and decks. Upside: simpler to size the rewards. Downside: feels less like things and more like cash.
  • Players can choose art, frame, and flavor text for their cards from between all editions ever printed. The rules text always matches the Oracle.

#Random Awards Mechanics

Per #core proposals, cards are randomly awarded for playing games. Cards from booster packs are governed by the same mechanics.
  • No basic lands (infinite supply baseline).
  • Moderate rewards against AI, slightly better rewards in PvP.
  • No less than 1 card per game.
  • Rarity is always random, but number of awarded cards may vary.
  • Rewards scale linearly with number of participants in multiplayer.
  • No difference in rewards between win and loss. No exceptions sans tournaments.
  • Increased rewards in official tournaments.
  • No reward when leaving game prematurely.
  • To avoid annoyance with stalling decks, games are limited by time; whoever has the highest health or the biggest library, wins. Turns are also timed.
  • Random awards never attempt to exceed the 4 card limit. Whenever the system detects an attempt to, it replaces the card, following the rules of #conversion.
  • Cards with an ability to exceed 4 copies in deck (Relentless Rats) are also limited to 4 by default, but give a special prompt offering the player to unrestrict them, also increasing their likelihood of unlocking to help make them relevant.
  • Cards are awarded in the format being played (Standard by default), or in the block when playing block Constructed. The player has a choice to lower this priority, enabling replacements from other formats or blocks when relevance of available replacements is too low.
  • Cards are acquired quicker in "eternal" formats due to volume of content. Accordingly, the merchandise prices there are lower.
  • Players have two ways to control random awards: based on theme decks (see #Half-Constructed for Newbies), and based on preferences that tilt randomness towards color and card class they prefer. For card classes, see the #grid proposal below.
  • Random awards attempt to provide a healthy mix of card classes. The engine dynamically adjusts the probabilities to increase the chance of getting card classes that have been awarded in lesser amounts. Card class selection based on preferences or theme decks increases the chances for the dominant class or classes, but not too strongly. This is necessary so that players don't have to switch preferences around to "mine" for cards of the missing classes.

#Grid of Mechanics

Mechanism for estimating relevancy of #conversion card replacements and #random awards. Imagine a 2D grid of colors and classes.
  • Colors:
    • Colorless
    • White
    • Green
    • Red
    • Black
    • Blue
    • Land
  • Classes:
    • Aggro
    • Control
    • Engine
    • Mana
    • Cards
(Cards = tutoring and advantage. Land is only used in conversion, not available for player preference.)

Grid's effect on conversion: when converting an excess card, the algorithm strongly attempts to respect card's position on the grid and pick a replacement of the same kind.

Grid's effect on player preference (see random awards mechanics). Player chooses one color and class. Random rewards respect this choice. When no more relevant rewards are available in the current format, a choice is given between a different preference and a relevant substitute from another format.

I also suggest displaying player's biggest preference near their name, similarly to how DotP shows player's preferred color. It could be a title, or something else flavorful. It's the preference under which they've been awarded the biggest number of cards.

#Half-Constructed for Newbies

Newbs need guidance. For them, it's better to be restricted in what they can randomly unlock. Otherwise the game becomes too overwhelming. What can we do?

An apparent solution is pre-constructed decks with fixed expansions. Each of them gets a sidelist of relevant cards, and by default, cards are randomly awarded from that sidelist for the currently selected deck. There would be a prominent option to switch between this mode, and mode of awards based on preferences (see #random and #grid). However, this would be inelegant, and not friendly towards players who want to expand their deck with cards not on the list, without having to strain their brain about card preferences. Instead of this, I propose what I'll call theme decks.

Instead of using pre-constructed sidelists and locking pre-constructed decks within a specific list of cards, we use the same engine we built for relevance-based random awards (based on the #grid of mechanics and other conditions) in the reverse way, to automatically determine a "theme" of each given deck, no matter if pre-constructed or not. In this context, a theme is a low number of highly manifested profiles shared by many cards in the deck. For example, if a deck is dominated by green and red, and by aggro and removal, that's its theme.

Then we allow the player to easily toggle between random rewards based on deck-fitting cards or the grid of preferences. Imagine a big switch: when on, preference selection is dimmed out, and you get cards based on the "auto-preference" corresponding to your deck's theme. When off, custom preference takes priority. This selection interface should always be easily visible.

This seems like a reasonable solution that helps blur the line between theme decks and constructed decks, and encourages custom deck building instead of discouraging it. The engine for determining deck themes and card relevance is already a major part of the proposal, just used in a reverse way.

One last note here is that every deck needs a variety of cards, and not just more of the same. Sure, control decks need control cards, but they also need some brawn to beat the opponent's face in. Aggro decks need more aggro, but they also need removal. This problem needs to be alleviated by making the random awards engine consistently provide a variety of card classes, dynamically adjusting the probabilities, and giving the preferred classes some but not too much advantage. Otherwise, all players would be "mining" control for creature removal before moving on to other classes, and would have to constantly switch their preferences around. See the last point in #random awards mechanics.

#MtG on Mobile

The last years have brought proliferation of ultra-portable computers. Most people have a smartphone around, many have tablets, and digitization of life and entertainment is astride. This is fertile ground for Magic. E-books on portable computers have superseded paper books and greatly increased reading engagement among the ever-busy people. Similarly, moving Magic onto handhelds can greatly increase the playerbase, raise the portability of the game, and somewhat futureproof it against the digital revolution.

What I suggest would be unthinkable to implement in the confines of the existing MO and DotP systems, but the overhauled system I propose above might be better suited for this.

The way I imagine it:
  • Wizards create an MtG app for iOS and Android.
  • The app has the same capabilities as the desktop version and players can play across desktop and mobile.
  • It requires a minimum screen size of 7'' to display the battlefield. Sorry, smartphone-only owners.
  • Unlike the desktop version which is limited to online AI and multiplayer matches, the mobile version is LAN-capable and offline-capable.
  • For a LAN match, several local devices are linked into a group by using Wi-Fi Direct or Bluetooth (e.g. like Open Garden does). One of the devices is designated as battlefield, and has a minimum required screen size (~7''). Other devices can be smaller, and are used for players' hands, plus also allow to browse the rest of the battlefield data.
  • Offline capabilities. The player's profile, their card collection, gameplay statistics, reward preferences, and decks, are stored on the servers. Online connection is only required for online matchmaking, microtransactions, and receiving card rewards. When offline, all data is cached on the local device, encrypted and protected in whatever way necessary to prevent cheating. Offline, players can play against AI or engage in LAN multiplayer. They can also create and change decks. Successfully completed games entitle them to rewards. Upon next connection to the servers, deck changes are synced to the server, and players receive the card rewards for which they've been made eligible through offline gameplay. The same mechanism is used when playing against AI or LAN when online, with the difference that rewards are doled out right away.
  • The app receives optional broadcast functionality. It might periodically announce itself to the nearby devices (by using Wi-Fi or Bluetooth), and at the same time scan the area for these broadcasts from others. If nearby MtG-enabled devices are found, it produces a notification. To reduce battery consumption, seek-and-broadcast is performed only once every few minutes, synced to identical intervals though global clock (say, every 5 minutes, 5 seconds before and after).
It would be an engineering challenge, it might be costly, but trading card games in mobile space have never been done by an established TCG company. Being the first and doing it right could grant MtG a crucial advantage.

#Culture of MtG and Sustainability of Digital

Magic stays in business because, by its nature, it requires and fosters human connection. The only way to play "real" MtG is by physically getting together. In order to play, you need to bring in friends. Showing and explaining it face-to-face allows you to draw them in, overcoming the initial complexity that would otherwise repel most people. Additionally, having fun together associates that emotion with the game. In other words, the nature of MtG makes it very difficult to market without human interaction, but makes crowd marketing viable by making real human interaction central to the game.

Now let's talk digital. Suppose we made an online multiplayer-oriented MtG game. You play and like it. What's the likelihood you invite friends? Low when compared to paper Magic, because friends who also play are not required for having fun. There's no shortage of opponents because the system matches random people automatically. You don't need to bring friends to play. Furthermore, even if you did, it would make no difference. Contact with other players is brief and fleeting, and they're more like AIs to you, so playing with friends wouldn't be much different than playing with random strangers. So just as with single player games, the rates of acquisition and retention should be much lower than with paper MtG, because the digital game doesn't require and foster human connection as strongly as the real game.

In other words, the paper game is probably the primary player acquisition driver. A digital-first MtG game risks being unsustainable unless it overcomes the dangers described above, and reproduces the crowd marketing effect of paper MtG. Part of what needs to be done is to avoid making it convenient to play alone. An apparent, even though brute, solution is to remove the automatic random matchmaking system, replacing it with ingame manual matching through chat rooms. This would force players to actively seek out opponents, promoting better activity on community forums, stimulating players to bring friends who're more regularly available than strangers, and to acquire friends inside the game. This might to some extent replicate the social culture that kept Magic afloat throughout decades, and make online-first Magic viable, enabling the redesign I'm suggesting.

#What's in This for Wizards of the Coast?

There are loads of people out there who want a great TCG like Magic, but would only play digitally. Because Magic Online follows paper MtG so closely, the acquisition model is alien to them, and there is no clear way to "catch up" and experience the old sets. There is huge potential with this audience, and it's being slowly conquered by games like Hearthstone.

Giving content away for free seems like giving up revenue, but it's not. Historically, MtG players always had ways to acquire cards for free or extremely cheaply. Longtime players sometimes relinquish huge collections of not-so-valuable cards, giving them to newer players cheaply or even at no cost. Wizards don't make money from these exchanges, but it helps bring people into the game. This occurs in paper MtG and Magic Online. Free card acquisition in the absence of trading fills the exact same role.

Revenue comes from two sources. One is expansions. Per the #core proposals, we require players to periodically purchase access to newly released blocks. They're motivated by the new content and by the shifting Standard format, which reduces the available pool of cards by rotating blocks out. The second source is boosters and pre-constructed decks. Players will be buying them to speed up their collections.

This design has the potential to greatly expand the playerbase. Lowering the barrier for entry, giving ways to explore Magic without frequent reinvestment, and using the graphically modern DotP engine, may explode the popularity, and with more people comes more revenue. Historically, online games reported significant profit boosts after switching from subscription to free-to-play.

Let me know what you think about this design. Would it work? Leave your notes, or weigh in on the discussion on the WotC forums. Share with your MtG-loving friends and in MtG communities. Who knows, with enough voices, Wizards might just make it real.

Friday 30 November 2012

A "thank you" to Eidos Montreal for DE:HR

Republishing an email I wrote to Eidos Montreal as a gratitude for Deus Ex: Human Revolution.

I want to give a major kudos to Eidos Montreal for the amazing game you have delivered with Deus Ex: Human Revolution. It goes both to the development team, and to management who weren't afraid to greenlight a story-focused, single player, non-sequelized game.

In today's gaming industry, companies are increasingly bent on multiplayer-focused franchises with DLCs and microtransactions and subscriptions; bloated publishers with overblown marketing budgets say "a game needs to sell 5 million copies to make profit" (Dead Space 3); studios are forced to dumb down design and storytelling down to "appeal to a wide audience" (abysmal ME3 writing); consolization results in most games being poorly implemented on PC with lack of graphic and controls options; games are ripped into chunks before the release to sell the rest as "bonus content" (Saints Row 3); anything that doesn't relate to sports, "military" stuff, or orcs and elves, is just not expected to sell.

In this environment, Eidos Montreal has provided an astonishingly outstanding and refreshing game.

The quality of storytelling and presentation is amazing and light years ahead of 99.9% of contemporary games, no matter how seriously they take themselves. The writing of dialogies in DE:HR was mind blowing. It is absolutely unique to hear in-game dialogue with so much sense to it! It's really intelligent and deeply touches on the issues presented in the game. It would be not out of place at mouths of world-class experts discussing same issues IRL twenty years from now. In today's gaming, when dialogues in most games are skin deep, when we are expected to feel threatened at "My minions will destroy you!" (D3) or cognitively pleased by the "truth" behind the Reapers (I made them kill you so you don't kill yourselves!), seeing real literatural quality behind the game was incredibly emotionally refreshing and hope-inspiring. The writer(s) have put their minds and souls into this, and please, let it be know, that this does not go unappreciated. And then this is combined with the incredible effort put into making the world live, and make it feel real. The detailed world, the attention to small things, the tiny details that make up the environment. The dialogues between people, the emails you read along the way. The ways the world changes as you progress through the game. The psychological nuances behind every dialogue in the game. The amazing attention and effort put into technological and scientific background for the game, and the myriad ways it shows through. It goes a long way towards making the DE:HR world feel, like none other, connected to reality, like a continuation of our everyday life and the progress we witness. The feel it invokes is like none other.

Technologically, it was just as refreshing a break from bleak staples of the industry, as story wise. The detailed inventory system, and the highly customizable controls make it feel like a game truly made for PCs, with no regard to console control limitations contemporary games suffer from. The custom engine provides the means for the unique gameplay design Deus Ex is worthy of, giving a healthy break from FPS-derivative-saturated market. The sound work is outstanding. The designers were unafraid to unleash an old-school complexity of interface, inventory and items systems to make the experience feel literal and connected, further driving home the "reality" feel of the near-future game world. The pacing of the story was near perfect, with awesome shakeups in the form of the boss fights (which are great, no matter what the whiners were telling you!). The free-world exploration, coupled with the high detailedness of the game world (environments, items, people, dialogues, quests, things to do, places to explore), along with the lengthy narrative tightly packed with the amazing narrative, well timed plot twists, and touching side quests, provided a lengthy experience to truly become immersed.

Special kudos to the music team who delivered one of the most amazing gaming sountracks ever. I haven't seriously appreciated it on my first playthrough, but the music for DE:HR is one of the most astonishingly powerful works of art gaming sountracks have ever seen. It has become a theme for some parts of my life, and hearing the tracks always invokes the immersion with the universe EM has crafted.

I regard Deus Ex: Human Revolution as one of the most amazing and important releases in the computer game industry. It is an outstanding achievement, something that is not afraid to stray out of the pack, and truly does the gamers the justice we deserve while proving the industry wrong, delivering one of the most engaging, thoughtful and memorable gaming experiences ever.

Thank you.