Dota 2 AutoChess Teardown - 3/5/2019

**The following post is an example of a design analysis and teardown written by me and does not reflect the opinions of my employer. It was written just after the release of the game.


• WHY WE CHOSE: Like the original DoTA mode in Warcraft III and the DayZ mode in ARMA 2, Auto Chess is an independently-made mode for Dota 2 that has been experiencing extreme popularity since its release. Its hype makes it primed to become a brand-new strategy game genre that will be copied and iterated on like MOBAs and Battle Royale games in recent years.

• GAMEPLAY: Auto Chess is a competitive strategy game that pits 8 players against each other in the same level in synchronous-asynchronous auto-combat. The game consists of rounds in which each player strategically selects and places characters on their own chess-like board. As the game progresses through rounds, player’s choices are tested against varying enemies, rotating between AI controlled creeps and other players’ character setups. The last player standing wins, with the top three players receiving rewards.


o Over 5.5 million downloads

o Over 100k CCU (20% of Dota 2 playerbase at a given hour)

o Dota 2 peak players Jan. 2019: 8.28% gain

o Dota 2 peak players Feb. 2019: 18.74% gain

o 5 Star Steam rating with 2.5 million ratings

• PUBLISHER: Valve: Steam Workshop – Makers: Drodo Studio (China)

• RELEASE DATE: 01/03/2019




• WORLDWIDE AND CROSS-GENRE APPEAL: Dota 2 is popular in both the US and Asia, giving this game mode a built-in player base in those regions. The strategic and competitive nature of the game appeals broadly to Dota 2’s existing audience, as well as those who are not dedicated Dota 2 players, but are interested in card games like Hearthstone. Recently, popular Hearthstone streamers have also begun streaming Auto Chess.

• SLOT MACHINE PSYCHOLOGY: The character randomization component of the game makes it feel like there is an even playing field and you always have a chance to win (and if you failed, you can blame the game). This component drives the habitual nature of the game, making it compelling to return to repeatedly.

• STRATEGIC DEPTH WITH TACTICAL DECISION MAKING: Like its namesake, Chess, and collectible card games (Magic, Hearthstone), there are deep learnable strategies that require responsive decision-making dependent on your opponent’s moves, providing a varied game experience from session to session.

• MATCH PACING: Like MOBA’s and BR games, the player can experience the full power scale of their characters and team in a single session. Being eliminated early leaves the player feeling like the task of upgrading was unfinished, driving the desire to play another match in an attempt to complete the progression experience.

• HUGELY EXPANDABLE: Currently, the pool of characters a player is presented is limited to a subset of Dota 2 heroes. There is vast room to expand this pool within the existing Dota 2 characters and by creating new characters and items, creating even more depth and shaking up the metagame, essentially mimicking the cadence and effects of card game releases.


• USABILITY: The implementation of the mode makes full use of everything Dota 2 has to offer, which is equivalent to a standalone gameplay experience, but is constrained by Dota 2’s controls implementation. Unless you are familiar with the controls of Dota 2, the player must depend on internet searches to figure out even the basics of play. Specific usability choices had to be made to work within Dota 2’s limitations, making certain actions require 1, 2, or sometimes 3 more clicks than necessary at importune moments, like when a player is working within a strict time limit. This experience can be notably frustrating. These choices are seemingly not intentional, but purely due to the fact the game is an adaption of Dota 2.

• ACCESSIBILITY: With depth comes complication, the difference between mainstream success and niche success is introduction of complication over time, Hearthstone is a fantastic example of this. Auto Chess dumps the player in the deep end with no meaningful on-boarding or explanation of the game’s core mechanics and strategies. Figuring out the basics of success are dependent on either learning through failure or an internet guide.

• SESSION LENGTH: The match session length ranges from 30 minutes (if a player loses immediately) to 50 minutes (if the player is in the final 3). While not vastly different from MOBAs, the match length for BR games tends to be shorter, especially for the players who lose early. This session length could become a barrier to frequent long-term play.


• This game mode is primed to have extreme success in the worldwide mobile market. The first company that releases and adequately promotes a version of Auto Chess on mobile will reap huge profits. All the game’s mechanics are proven mobile-friendly allowing for a quick turnaround mobile port.

• A PC version could be released to compete with the Dota 2 mode. A PC release that is not limited by the constraints of being a mode for another game could steal and then expand the market, but would require a unique spin and careful decision making to avoid being blasted by the vocal and troll-like Dota 2 community.


Like slot machines in Vegas combined with collectible card game packs, the mode is a masterclass in dopamine manipulation. The emotional pacing of each round is a compelling up and down of urgent, frantic decision making and action, followed by short downtime where the player gets immediate cause and effect feedback. Each round starts with pulling the slot machine handle (or opening a card pack), then deciding, then seeing the results of those decisions. This pacing makes players feel as if they will always have a chance to win if they get lucky and compelled to return and try.


• TACTICAL SLOT MACHINE: Each character a player can place is randomly chosen from a shared pool of characters. This randomization lets the player “get lucky” draws that set them up for success as the rounds progress. But because this pool is shared among all the players in the match, there is the unique strategy of choice manipulation to get the characters the player needs and deny opponents theirs. This mechanic is used in board games, but rarely in digital strategy games.

• SPEED CHESS: Paired with the randomization is a timed component. Players must make all their decisions and place their characters within 60 seconds. This has the two-fold benefit of spiking the stress level of the player and making the game more interesting to watch.

• SYNCHRONOUS-ASYNCHRONOUS COMPETITION: While the matches are 8 players, there is not player directed head-to-head competition. When two players face off against each other, a copy of the opponents’ team is populated on each players’ board with AI controlling the actions of both teams’ characters. In most cases, players aren’t even aware that their team is off facing another team, since your focus is on your own board at most times. Because of this lack of direct competition, toxicity is minimal and the competitive aspect, which arguably is the biggest barrier to multiplayer strategy game accessibility, is less intimidating.


As a game mode for Dota 2, Auto Chess cannot effectively monetize.

What they have implemented is a simple currency for appearance items in the store. Each match, the top three players earn a currency (candy) which can be used to unlock skins for the player’s courier (representation of the player character). Players can earn up to 10 candies a day.

Their obvious intention is to sell candy, but at this point there is no way to purchase more. Early on, the Auto Chess developers were selling the currency on EBay, but that practice was discontinued.


• THE RACE IS ON: There is no doubt that gaming companies have noticed the success of Auto Chess and many will start copying the concept. A mobile version was already released in China, but was removed soon after due to intellectual property theft concerns.

o CLASH OF AUTO CHESS?: Supercell (Tencent) and the Clash IP are the most obviously suited to capitalize on a mobile version this game mode. Making a version of Clash Royale with the rules and mechanics based on Auto Chess would be a nearly trivial endeavor. They also have the expertise to retain the depth of Auto Chess while effectively solving the accessibility and monetization issues. Cross promotion within in their own games would ensure visibility and downloads.

o WAR FOR PC: If the trend-chasing behaviors for MOBAs and Battle Royale games hold true with Auto Chess it is very likely many versions, at multiple levels of fidelity will release, explode in popularity, then fade out of the gamer mindshare… Until a major studio releases a true AAA version that is suitable for E-Sports or extremely mass-market.

▪ VALVE: It is rumored that Valve is either creating their own version of Auto Chess and/or is in talks to purchase/employ Drodo Studios. It seems most likely that Valve will make the first big move to capitalize on the game mode.

▪ TENCENT: Due to the cross-regional popularity of Auto Chess, it seems feasible that Tencent will either make a play to purchase and staff the Auto Chess developers or copy the game through one of their existing companies for a PC release in Asia and in the West.

▪ HI-REZ STUDIOS: Hi-Rez Studios, makers of SMITE, Paladins, and Realm Royale, have a clear history of early trend chasing and have pivoted to chase trends in the past. Additionally, their experience with making character-centric games would be a benefit in their speed to market.

▪ RIOT: While Riot and League of Legends would make an obvious contender to steal this market, Riot does not have a track record of agility with game releases or a propensity to follow trends. Its likely they could introduce an event mode, like their release of the popular custom game mode ARAM, but a full standalone game investment does not seem likely.

o AN UNLIKELY FORAY INTO CONSOLE: Few strategy games or card games see major success on consoles. It seems unlikely that Auto Chess will buck that trend in its current format. A unique take on the game that transforms the controls and view to something that is suited to controllers and TVs could see some success. Any entrants into the console market will most likely be indie studios with small budgets using Unreal 4 or Unity for development.

• E-SPORTS: There have already been a few Auto Chess tournaments, the latest sponsored by the creators. There is an obvious desire by Drodo Studios for Auto Chess to flourish as an e-sport that leverages the built-in Dota 2 audience. As the game is copied and iterated, I believe Auto Chess is the first game that has a real chance at becoming a globally popular mobile e-sport. Clash Royale had a brief moment of streaming and talk of e-sports, but if a company doubles down on the direction to make Auto Chess as a for-mobile e-sport instead of a cash grab mobile game, it could see competitive success.

• CANDY SHOP SHUTDOWN: In its current form, the Auto Chess store is shady at best. The creators do not own the skins they are providing and the potential for fraud and lost profits stemming from sales of candy on a third-party market, is high for Valve. Without support from Valve in some official fashion, Drodo Studio faces an uphill battle trying to turn a profit of any kind from this game mode.



• There will be a chase by game developers to copy and capitalize on the heat created by Auto Chess. Many will fall into game design traps by trying to monetize the characters or make the drawing of characters more predictable, effectively breaking the core mechanic of the game. Those that iterate within the mechanics of the design could see huge success.

• For Auto Chess to retain its following and grow in popularity, it must evolve quickly by broadening its depth. Already there is a developing meta that can make the game predictable which, if allowed to stagnate, would spell the end of its growth.


• Easy to learn and hard to master is an axiom for a reason. Give players a glimpse into all they have to learn without overwhelming them and they will return to play over and over again.

• Properly used, randomization can create new gameplay opportunities that on paper may sound unfun or unintuitive but are satisfying and habit forming in practice.

• Applying pressure through a timer at critical decision-making points is a simple way to amp up the overall experience.

• There must not be a single “right way” to play. The more viable options the player has the more compelling it is to return and try them all.