Design Dive #1: How VALORANT Map Design Differs From Counter-Strike

One of the most fascinating things VALORANT is changing in the tactical shooter genre is map design. Part of these changes are from the fact that agents have unique abilities, but it goes much deeper than that. Riot is making fundamental changes to a decades old formula. And they're doing it with iconic Counter-Strike map makers like Volcano, the creator of the map Cache.

I've played thousands of hours of competitive Counter-Strike, worked on competitive multiplayer game design and I've been playing VALORANT since Closed Beta.

Let's take a design dive on how VALORANT's map design challenges Counter-Strike's decades of shared philosophy.

The Fantastic Changes

To kick things off, VALORANT added some great fantastic changes to the formula.

You can shoot through protective cover

Counter-Strike 1.6 had this, but has since become a lost art. In Counter-Strike: Global Offensive most of the popular maps have 'headshot-only' spots where you are only revealing your head, and the material of the cover is not penetrable by bullets. This raises the skill floor for fighting on the maps (knowledge burden to know the spots + very small targets). It also can be frustrating when people annoyingly duck in and out of cover. Making these covered spots less advantageous appears it is a design tenet in VALORANT maps. I love this change, as it makes it much easier to balance the advantage of defending vs attacking.

The Areas to Plant Are Larger and More Strategic

In Counter-Strike, the bombsites are typically very specific areas only a few meters wide. In Valorant, the bombsites can be even 5x larger and sprawl most of the area. This increases the strategy and freshness of where you plant the spike and how you hold against retakes. While this is similar to the CS map Inferno, VALORANT takes it further with how well it resonates with agent abilities and bombsite designs to create strategic choices.

My favorite example for strategic planting is Haven's B site. It's a very hard site to protect with 3 big entries in the middle of the map. As of now, most players are planting in the default safe-to-plant-but-not-to-hold spots (D on the diagram), which is why most players hate trying to execute B on Haven.

In reality, you don't actually have to even protect the site. You can plant near middle (1 on diagram), and just watch the bomb from the middle area. You can plant near the C or A rotations and have it fully in view and only hold one side.  I predict the typical plant spots will change a lot in higher ranks as new agents and strategies form.

Each Map Has a Unique Mechanic

Bind has one way teleporters. Haven has a third site to plant at. Split has vertical zip wires (and man are they better than ladders). Ascent has buttons that close metal doors.

This solves the commodity maps problem Counter-Strike has. There are so many maps are created with the same mechanics that it is difficult to know when to care about a new map. This leads to pains where great level designers like FMPONE can work for years on a map like Subzero and ultimately not garner enough playtime to stay in rotation. Unique mechanics per map on VALORANT gives players a unique reason to hate or love it, and allows the map to give it's fresh angle to the game for the test of time like champions in League of Legends do.

The Not So Perfect Changes

Despite VALORANT being in R&D for a long time, there a lot of great room for VALORANT map design to grow positively. This is just the beginning of live service. Here are a couple problems I've identified.

Overly Constrained & Condensed Map Formula

Three main things are applied everywhere on the maps: less maximum distance between players, choke point widths are standardized to a fault, and higher density of angles and corners to check. These drastically limit how maps flow and feel. All of these philosophies are very Inferno-like. Let's quickly break these down:

Less Maximum Distance Between Players
The longest sight paths in VALORANT maps, where you fight players far away, are around 2/3rds what you would see in the CS Map Dust2, and around the same as what you would get in the CS Map Inferno.

Choke Point Widths Are Standardized to a Fault
There are standard widths applied everywhere in VALORANT maps. Any choke point is always easily blocked with a single smoke.  This is common in Counter-Strike, but not categorically true (and also more difficult to pull off). In Dust 2, attacking Long A regularly warrants using 2 or 3 smokes. To me, this enriches the variety a lot and I do not know why it isn't present in VALORANT, especially when a team of agents have a lot of capacity to block vision with smokes and walls.

Higher Density of Angles and Corners to Check
Every rotation path is littered with places where someone could be hiding around a 90 degree angle, behind a box, or on top of some vertical element.

As a result, maps are much smaller and their effective sizes are even smaller. Below, we compare Counter-Strike map sizes vs VALORANT's using a standard box for a world-scale rough unit of measurement.

The effective map sizes for VALORANT are actually about 25% smaller than these results because the maps have large spawn areas which end up being mostly negated by the Buy Phase. VALORANT maps all feel smaller than Inferno, and no map comes close to the ever popular Dust2 in terms of scale.

These constraints create 3 great advantages:

1. When you engage an enemy, your abilities are in range to be useful. The lanes are small enough that one smoke is always meaningful.
2. By having shorter distances to render, the game runs better on lower end computers. More people can play VALORANT. Plus, no one likes being killed by a pixel half way across an island (or maybe you do as long as you get to go to the Gulag after 😅).
3. By having dense maps, you hypothetically reduce travel times and increase the overall action. These aren't the only variables determining player pacing though.

Those are great things! The pain is that the present result is an overly homogenized experience. There are great unique mechanics and themes for each map, but without more freedoms in the way maps are fundamentally designed to flow, VALORANT maps do all feel like Inferno. Even though Inferno is my favorite Counter-Strike: Global Offensive competitive map, I still crave more variety in the map setups.

Buy Phase Barriers: Brilliant Idea But Promotes Staleness

VALORANT's Buy Phase barriers allow teams to position on the key areas of the map before the round starts. It seems like a live-service dream: you can tweak the timings of engagements on maps without having to edit the maps themselves, which is costly. It also lets the team better coordinate since they telegraph where they are going to attack or defend before the round starts. So what's wrong?

The problem with Buy Phase Barriers is that they remove very important round variation in the genre. In Counter-Strike, you are unable to move during buy phase, and you spawn in a semi-randomized spot in your team's starting area. On attack, you might spawn closer to B site, where the player who is defending B site spawns further away. You don't know for sure. It's all varying around 3 seconds or so. When you run in, the players will not engage at a predicable time or place.

VALORANT on the other hand, has no variations here. When I'm defending, my buy phase barrier allows me to literally set up just aiming at the choke point I'm protecting. If an attacker is pressed against their buy phase barrier and rushing a choke point, there is an exact timing for that. I just pre-fire my gun at that time, and I often get kills. This is painfully simple in VALORANT due to the buy phase barriers being static.

This makes things feel unfortunately similar each round. Yes, players can inject their own tempo into the game by delaying or using abilities, but rushing is thrilling. It is very important it is that it is fun, fresh, and fair again and again. Plus, this isn't fun for the defender's either. It is not engaging to pre-aim and pre-fire every round, but players will optimize the fun out of the game if it's advantageous for them to do so.

Taking VALORANT Map Design to the Next Level

1. Expand the Formula For Paths Between Bombsites and Spawns

Let's look at what is working: VALORANT nails the variety of bombsites despite having a design playbook for them as well. Each bombsite has a high ground (usually called 'heaven'), a raised platform area, places to boost up to with abilities, one or two crevices that you can hide in. They are combined into a satisfying dance of angles that play well with abilities. Despite the formula, the designers are able to express great and novel setups.

The problem is the paths between are very transactional and repetitive most rounds. You start near a Buy Phase barrier, and then run into a tunnel. Most conventions are good ideas but ultimately lose power when followed in every case.

Instead, VALORANT could be pushing the limits in a way that makes sense only for this game. For example, I would love a map where you could dual smoke off a wide lane, run up half way and boost your team on a wall to get onto less accessible optional high ground path to a bombsite. Setups with more tactical variation through larger size and options create deeper meta-gaming and strategies.

2. Introduce Per Round Variance for Freshness

VALORANT can design better variations in the gun duels when challenging territory. It could be as simple as having the Attacker side Buy Phase Barriers randomize their positions slightly. This way, there is nothing defaulting to static about the path from round start to first engagement. I would really enjoy that as a player!

3. Release Maps More Frequently

Right now in VALORANT you have 4 maps and you can't choose what one you get. This might work for a hero shooter like Overwatch, does not work well for the tactical shooter because the tactics are in the map itself.

One might find no issue with this, because a minority of 5-10 Counter-Strike maps get the most playtime. However they are a by-product of player choice, from a gauntlet of thousands of maps over decades. The small amount of highly played maps is also partially because players want to compete on maps that they know well. You get to pick in Counter-Strike, and there are many more maps, modes, custom servers for the players who want more. It's opt-in player burden.

They state in their Birth of Ascent article that they keep their maps in the Greybox phase (early stage gameplay testing) for several months to several years. I worry about the amazingly high quality bar Riot has set for maps in VALORANT. Not only do they have unique mechanics and a lot of design care put into them, they have to be extensively play tested with many skill levels, and they have a lot of art requirements. These maps are bursting with beautiful art hidden everywhere, even outside of the map!

VALORANT long map pipeline places a harsh burden on managing their release cadence. This could weaken the game's success during live operations if they have not already stockpiled maps and talent on their maps team. I'm painfully ready for more maps!

In closing

VALORANT has exceeded my expectations on their approach to cover, how they design bombsites, their philosophy of adding uniqueness to each map, and the incredible distraction-free art that they Tetris-pack the maps with.

My hope is that the future of VALORANT includes:

1. A wider variety of path designs connecting the maps together.
2. Buy Phase Barriers randomly shift slightly on the attacking side so that rushing immediately doesn't create a predictable timing.
3. A higher cadence of map releases and eventually more player choice in what maps to queue for.

Overall I'm very optimistic that VALORANT is on a great path and the team just needs more time to expand their map roster. I have no doubt the team at Riot is capable and aware of these ideas in some form. This is not meant to be a harsh criticism or a claim that I know best, but simply me sharing my design reflections on VALORANT and Counter-Strike map design.

Thanks for reading! If you would like to receive new posts via email, sign up here: