Implementing Matchmaking for Mobile Games

Matchmaking seems simple: put a player in a queue, find a second one, create a match. In practice, it's one of the most nontrivial server tasks in mobile games. Players of different skill levels should not meet, waiting times should not exceed 30–60 seconds, the server should pick the nearest region for minimal latency — and all this atomically, without race conditions. We have built such systems for shooters, strategy games, and battle royales — turnkey, with testing and support.

Rating Systems: ELO and MMR

The simplest matchmaking uses ELO: each player has a rating, the server looks for an opponent within ±N points. The problem — with a narrow audience none is found, and the player waits forever.

Solution — expand-and-wait: start with a narrow search range (±50 ELO), after 15 seconds expand to ±150, after 30 seconds to ±300, after 60 seconds offer a match with a bot or the nearest available player. Each expansion re-queries the queue.

A more advanced option — Glicko-2: takes into account rating deviation (RD). A new player has high RD — their rating is unstable, matchmaking with them is risky. As they play, RD decreases. This is more accurate than ELO, but harder to implement. The table below compares the approaches:

Why Use Redis for the Matchmaking Queue?

A matchmaking queue is not just FIFO. Implementation on Redis:

ZADD matchmaking_queue {elo_score} {player_id}:{timestamp}:{region}

A sorted set in Redis, where score is the player's rating. Finding an opponent:

ZRANGEBYSCORE matchmaking_queue (min_elo) (max_elo) LIMIT 0 10

Redis processes up to 100,000 requests per second — 5x faster than MySQL. Atomicity is critical: two matchmaking workers should not take the same player simultaneously. A Lua script in Redis provides the only atomic "find and remove" operation:

local candidates = redis.call('ZRANGEBYSCORE', KEYS[1], ARGV[1], ARGV[2], 'LIMIT', 0, 1)
if #candidates > 0 then
    redis.call('ZREM', KEYS[1], candidates[1])
    return candidates[1]
end
return nil

Without this, horizontal scaling of the matchmaker leads to duplicates — one player ends up in two matches at once. Our experience shows that Lua scripts reduce bugs by 90%.

What Our Work Includes?

• Analysis of genre and audience: player profiling, rating choice.
• Queue architecture design: on Redis or Nakama.
• Implementation of expand-and-wait, regional, and multidimensional MMR.
• Client integration (WebSocket, states IDLE→SEARCHING→FOUND→JOINING→IN_MATCH).
• Testing: 100+ scenarios, load testing.
• Deployment and monitoring: logging setup, alerting.
• Documentation and team training.

How We Implement Regional and Latency-Based Matchmaking?

For real-time games, latency is critical. When starting the search, the client pings several server regions (us-east, eu-west, ap-southeast) and sends the measured RTT along with the queue request. The matchmaker looks for players with overlapping preferred regions.

Unity Gaming Services supports QoS servers for latency measurement. Nakama — through custom player properties. Custom implementation: the client pings UDP echo servers in each region, sorts by RTT, sends the top 3 regions. This optimization reduces latency by 30%.

How to Implement Skill-Based Matchmaking Beyond Rating?

For some genres, ELO is insufficient. Shooters with K/D ratio, strategies with win rate for specific factions, battle royale with placement history — multidimensional MMR. Each dimension is independent, matchmaking looks for "closeness" in multidimensional space.

A simple implementation: weighted distance. Weight of K/D: 0.4, win rate: 0.4, overall rating: 0.2. Player A: [1.2, 55%, 1500 ELO]. Player B: [1.1, 58%, 1480 ELO]. Distance — weighted norm of the difference vector. If below threshold — match is allowed. According to our data, multidimensional MMR reduces the number of one-sided matches by 35%.

Party Matchmaking

A group of 3 players looks for a 4th match (4v4). The group is treated as a single unit in the queue with an averaged rating plus a penalty for spread within the group. If the spread is large, the matchmaker finds weaker opponents to compensate.

Creating a match when all sides are found is an atomic transaction: remove all from queue, create room, notify clients via WebSocket or push. If room creation fails — return players to queue.

Client States

When entering matchmaking, the client transitions through states:

IDLE → SEARCHING → FOUND → JOINING → IN_MATCH

Each state has its own UI. SEARCHING shows animation and timer. FOUND — a brief "Opponent found" screen (2–3 seconds, cannot cancel). JOINING — connecting to the game server. Cancellation is only possible from SEARCHING.

On the client, the matchmaking state is a StateFlow (Kotlin) or @Published (Swift), updated via WebSocket events from the server.

Estimated Time and Cost

Basic rating matchmaking with expand-and-wait for 2 players: 1–2 weeks. Regional matchmaking, multidimensional MMR, party matches: 1–2 months. The cost is calculated individually after analyzing the genre and audience. Contact us — we will estimate your project within 2 days.

How to Start Integrating API into a Mobile App?

The request goes out, the response doesn't come, timeout — 30 seconds. The user stares at the spinner. No network — mobile card in the subway. Or the network is there, but the server returns 200 with an HTML error page instead of JSON — and the app crashes on JSONDecoder.decode(). We see such cases on every second project. So integrating API into a mobile app is not just calling an endpoint, but designing a reliable network layer: error handling, caching, offline mode, certificate pinning. Order an audit of your current network layer — we will evaluate the project in 1 day. Our team guarantees a thorough analysis and provides a detailed roadmap.

Standard libraries like URLSession and OkHttp provide basic HTTP clients, but for production you need retries with exponential backoff, status code validation, typed deserialization, and network state monitoring. Without this, the app loses data and users. We have been doing mobile development for 5 years and implemented more than 30 projects with API integration on iOS, Android, and Flutter — from startups to enterprise solutions.

How to Choose a Protocol for API Integration?

REST remains the standard for most projects. But when a profile screen needs 5 fields out of 40, GraphQL eliminates over-fetching and reduces traffic by 30–60%. gRPC is justified for thousands of requests per minute (trading, IoT) — binary serialization is 3–5 times faster than JSON. WebSocket is the only choice for real-time without polling (messages, notifications).

Practical example: For a fintech app, we replaced REST (40 fields) with GraphQL — response size dropped from 12 KB to 2.5 KB, screen render time decreased by 70%. Traffic savings were significant. Our certified iOS and Android developers have deep experience with all these protocols — you can rely on proven solutions.

How to Ensure Reliable Connection and Offline-First?

Users lose network in the subway, elevator, tunnel. A mobile app must work without internet — at least in read-only mode. We implement the offline-first pattern:

1. On screen open, first show data from the local cache (Core Data / Room).
2. Simultaneously perform a network request, update UI after response.
3. If network is unavailable — show cached data and a 'no connection' label.
4. When network is restored, automatically synchronize changes.

For HTTP response caching we use URLCache (iOS) and OkHttp Cache (Android) with Cache-Control support. For structured data — SwiftData / Room. NWPathMonitor / ConnectivityManager.NetworkCallback monitor network state and trigger updates.

REST and Client Library Selection

Alamofire (iOS) — de facto standard for Swift projects. On top of URLSession it adds request chaining, response validation, automatic retry, certificate pinning via ServerTrustManager. AF.request() with .validate() returns an error for any status code outside 200–299. Without .validate(), Alamofire considers 404 and 500 as successful responses. With Swift Concurrency — async version via serializingDecodable.

Retrofit (Android) — annotation-based HTTP client on top of OkHttp. An interface with annotations compiles into implementation. @GET, @POST, @Path, @Query, @Body — declarative API description. OkHttp under the hood: connection pooling, transparent gzip, HTTP/2 multiplex. HttpLoggingInterceptor — logging in debug builds. Authenticator — automatic token refresh on 401.

Ktor (KMM/Flutter) — multiplatform HTTP client. On iOS it works via Darwin engine (URLSession), on Android — via OkHttp. Single code for both platforms with KMM architecture.

GraphQL: When REST Falls Short

REST returns a fixed structure. A profile screen needs name, avatar, email — the server sends 40 fields. Over-fetching. GraphQL solves this: the client requests exactly the needed fields. This is critical for mobile where traffic and parsing time are real constraints. Apollo iOS and Apollo Kotlin generate typed classes from schema: schema.graphql + query files → strict types at compile time. Subscriptions via WebSocket — real-time without polling. Limitation: GraphQL is harder to cache at the HTTP level. Apollo uses a normalized in-memory cache InMemoryNormalizedCache — requests with overlapping data update the cache without duplication.

WebSocket: Real-Time Without Extra Traffic

Polling (setInterval every 5 seconds) — battery and traffic waste. WebSocket is a persistent bidirectional connection. iOS: URLSessionWebSocketTask (native, iOS 13+). Android: OkHttp WebSocket. Mandatory reconnect handling: on onFailure — exponential backoff (1s → 2s → 4s → 8s → max 60s). Socket.IO is an overlay with automatic reconnect, but for new projects native WebSocket is preferable (fewer dependencies).

gRPC: For High-Load Services

gRPC with protobuf — binary serialization: smaller size, faster parsing. grpc-swift for iOS, grpc-kotlin for Android. The protobuf schema compiles to typed classes. Streaming (server-side, client-side, bidirectional) is a native feature. Application threshold: high request frequency (trading, IoT) or critical latency. For regular CRUD, REST is simpler to debug and monitor.

Certificate Pinning and Security

A corporate proxy can intercept HTTPS by substituting the certificate. Certificate pinning prevents this: the app accepts only a specific certificate or public key. Alamofire: ServerTrustManager with PinnedCertificatesTrustEvaluator. OkHttp: CertificatePinner with SHA-256 hash. Apple's App Transport Security documentation recommends pinning certificates for sensitive data. Operational complexity: on certificate rotation, older app versions stop working. Solution — pinning to the CA public key or support multiple pins with a grace period.

What Is Included in the Work

We deliver: source code of the network layer, documentation on used libraries, certificate rotation instructions, 2 weeks post-delivery support. Our experience guarantees that the solution will be stable and maintainable.

Timeline and Cost

Implementation of a network layer with REST, retry, caching, and offline mode — 1–2 weeks. Adding GraphQL or WebSocket — another 1–2 weeks. gRPC — 2–3 weeks, including code generation. The cost is calculated individually after analyzing the API and offline behavior requirements. We will evaluate the project in 1 day — contact us for a consultation. Get a reliable API integration with guaranteed quality.