Tarkov: Breaking the Rules

Escape from Tarkov works much better than it should by conventional standards of “good game design”. As a matter of fact, it even deliberately weakens some of them and turns the resulting “issues” into strengths by applying a unique kind of core loop.

The first-person shooter from Russian developer Battlestate Games has already sold millions of copies despite entry hurdles such as immense complexity, a reliance on players’ self-motivation in learning the game and a premium pricing model. This is an attempt at an explanation from a systems-design point of view.

Tarkov’s core game is a realism-driven multiplayer first-person shooter based in a near-future scenario around private military companies. Depending on the size of the area, 5–14 players are randomly distributed across the map in closed instances (“raids”). Their most fundamental goal: reach one of the exit points (“exfil”) that is available to them.

To do so, players will usually have to traverse the map extensively, often encountering other players as well as AI-controlled enemies, so-called scavengers (“scavs”). Each player’s specific route depends on various factors: Do they avoid conflict or seek it out? How much do they want to stay in cover? Do they need to visit a specific location to complete a task? What’s their health status? How much ammunition, food and water do they have left?

“Shoreline” is one of the maps in Tarkov. Players are randomly distributed across the starting positions and must reach one of the exits available to them. (illustrative representation)

The gameplay is intense, the atmosphere dense. The levels are designed to offer myriads of options for cover, firing angles and tactical maneuvers. So far, so potentially exciting, but ultimately just a well-made shooter? No. The true Tarkov experience is fueled by its metagame and the structure connecting all the firefights.

While the individual raids are self-contained 10- to 40-minute sessions, the equipment and goals of each player stem from the overarching metagame. On the one hand, players fulfill tasks for the NPC traders in order to gain their trust and unlock access to an increasingly more powerful arsenal of weapons, ammunition and equipment. On the other hand, players also upgrade their hideout with certain building materials to gain access to higher-quality crafting recipes. Or, at a much later stage, to outright generate money via their own Bitcoin farm.

Inventory management is part of everyday Tarkov life. Unsurprisingly, one of the most popular features from the recent past is the “sorting table”, which temporarily provides unlimited space to rearrange items.

These two axes of progress lead to specific goals for each raid. For example, players sometimes have to visit a specific location to complete a task; or they need to defeat a certain number of scavs or players; or they may have to find specific items to initiate their next hideout upgrade. More generally, earning as much money as possible, i.e. hunting for particularly rare items to sell, is also a common motivation. Depending on the task a player is pursuing, they may completely change their equipment “loadout”.

That is where the next puzzle piece of the Tarkov formula takes the spotlight: Everything you bring into a raid is at stake. When you die, your corpse simply becomes a container for other players to loot. Your equipment for the next raid has to be provided from your stash or by purchasing it from the afore-mentioned traders. The meta progress — especially concerning items and money supplies— is not a one-way street, but a full-fledged part of the game that generates interesting decisions and risk-versus-reward considerations.

Since your equipment can be permanently lost, the gameplay often feels quite similar to that of a roguelike. Putting yourself in an unfavorable position or moving too loudly, so that — potentially better equipped — other players could deduce information about where you are located, can at any moment have fatal consequences. Every step matters.

One of the roguelike genre’s central questions: Persistent progress (and thereby long-term meaning of the time you invest) or not (thereby fully focusing on skill and learning)? Tarkov says: “Why not both?”

Roguelikes often use their metagame to make life easier for players over time, giving them a sense of steady progress. In Rogue Legacy for example, there are countless small upgrades you can unlock that permanently increase your HP or damage. This usually causes resentment among the more “hardcore” parts of the audience. How can your skill be measured in such a system? Am I actually learning something myself or is it just my pixel self getting stronger? Am I prevented from playing the “real” game until I invested enough time?

The other extreme is completely doing away with persistent progress. In the genre-defining Rogue or the modern classic Spelunky each run is self-contained and has no connection to the next one — beyond actual skills and knowledge the player acquired of course. Since nothing remains in-game after a run ends though, many players get a feeling they wasted their time.

Therefore an intermediate variant of progression has recently become more and more popular: The player unlocks features and variants, but no power advantages. In The Binding of Isaac there are characters and items to unlock, in FTL there are spaceships, in Into the Breach there are new mech squads, and in Monster Train additional cards. A neat side effect to this way of doing things: More complex gameplay elements can be pushed towards later runs when players are already a bit more experienced with the game.

Tarkov does have persistent progress. However, it quickly pales in comparison to the depth of the other parts of the metagame.

Tarkov does all of the above:

  • Your avatar increases basic skills like stamina or strength over time. Experience points and hideout upgrades are also permanent.
  • Your equipment (and your money supply) is at stake and can be lost forever. Exception: Items placed in the “secure container” during a raid are preserved. (There is a mechanism to save you from truly going bankrupt by the way: You can periodically do “scav runs” allowing you to take the role of a random NPC scavenger and transfer any loot you find after escaping to your own stash. Probably Tarkov’s “friendliest” feature).
  • Completing tasks for the NPC traders unlocks permanent access to higher-quality (and more expensive) equipment.

Much more importantly though, the metagame fuels the game’s depth. It enables players to form their own infinitely nuanced objectives and thereby introduces a just as nuanced definition of “victory”. Formally, surviving a raid with a successful extraction could be considered a “win”. However, it might be of little value if you did not complete a task, did not find useful crafting or building materials, or did not obtain valuable new equipment.

Vice versa, a raid that ends in death could be considered a success if, for example, you finally completed a particularly tricky quest or managed to put a valuable item into your secure container before dying.

This softening of the traditional meaning of a “victory condition” is not just a side note in Tarkov’s design, but fundamentally required. Tarkov is not a fair game at all. Rich, high-level players with pimped rifles, the highest armor class and super effective painkillers are thrown into raids with newbies. Lone-wolves can run into duos or trios. The only thing that matters for the game’s matchmaking is that these players selected a certain map at the same time.

To expand the breadth of possible situations and thus its replayability, Tarkov deliberately goes against what you would consider good design in the sense of “flow”. The level of challenge does not cater to or even care about the player. Instead, evaluating when to avoid and when to take on conflict is itself part of the game. Beyond your current task, equipment and the remaining space in your backpack, you will also have to adapt your style of play to your current environment. Figuring out what equipment other players are running — for example based on weapon sounds — and deciding how aggressive or how stealthily to move along is just another part of the Tarkov skill package.

The basics, explained by one of Tarkov’s most prolific players.

Now, thanks to the afore-mentioned highly individual definition of success, this actually works extremely well. Remember, this game is not a battle royale, and the maps are not primarily combat arenas. Avoiding other players or fleeing from them is a perfectly valid strategy. Sometimes it is also a bad one though. The possible combinations of raid compositions and each player’s current metagame state are practically infinite and constantly generate new situations. This adds a strong element of planning to the mix that goes way beyond classic “deathmatches”.

However, when the paths of two players do cross after all, the complexity of their moment-to-moment decisions is also significantly enhanced by the metagame. The most fundamental consideration being: What do I have on my body or in my backpack (i.e. what do I have to lose)?

And it goes on: Where can I find cover nearby? How far is the next extraction point? Could other players know where I am? How quite do I have to be (movement speed and thus noise can be regulated in multiple steps)? Which angle would they expect me to take? Am I up against a group? Did I hear or see other players during the raid who might try to “third-party” and profit from a conflict? How likely is it that they themselves just tried to complete a particular quest and have already made a run for it by now?

In short, the metagame connects the short-term raid arcs with a larger whole in such a way that they themselves gain an enormous amount of depth and generate remarkable emergent stories quite regularly.

“The perfect raid is the kind of raid that you are so thrilled about that you want to post it, on reddit for example […] like you are so eager to post it and tell the story about it.” (Nikita Buyanov — Game Director on Escape from Tarkov)

The above quote sums up the unique core of the Tarkov vision. It is relatively safe to assume that the resulting gameplay formula will pop up in other places in the coming years. The Cycle: Frontier, a more accessible sci-fi variant, is already in the making. Lost Light is coming to mobile. Battlefield 2042 has “Hazard Zone”, its own Tarkov-like mode. But there is also great potential beyond the first-person-shooter world. After all, the raid gameplay can be replaced with almost anything without damaging the cornerstones of the concept. Even single-player variants are conceivable.

Now, is Tarkov itself worth your time? Yes, but be ready to dive in deep. Just getting to know the often quite extensive maps takes weeks (there is no minimap in the game, you get a compass and do traditional map reading on a second screen). The value of the countless items, ammunition types and weapons can only be fully assessed after hundreds of hours. And you will have to consult the wiki when trying to find where exactly to go to complete quests.

As for the shooter part itself, you can of course partially transfer your skills from other FPS games. But even that part requires an unusually methodical approach with its “as realistic as playable” motto, which makes it feel much more like an ARMA rather than quick arcade action. The reward for all that effort? From a player’s point of view a completely unique experience and a basically infinite variety of exciting situations; and for game designers a title that will question your cherished “design wisdom” and provide you with fresh food for thought for a long time.

On that note…

Game designer chasing the elusive beauty of systems. https://twitter.com/Ludokultur