Everything on Statlocker
Everything on Statlocker, in one place. A depth and breadth of features far beyond what most stat sites offer, and we are nowhere near finished.
We built Statlocker because we believe game stats can go so much further. Too many trackers show you a match history, a few win rates and a leaderboard, and stop there. A small, passionate team has worked on Statlocker almost every day since April 2025, taking that idea toward its logical conclusion. We love Deadlock and have played many thousands of hours of it between us, which is why the questions this site tries to answer are the ones players actually ask.
Match history, hero and item win rates, a tier list, leaderboards and builds are all here. Those are the floor rather than the product. What sits on top is the part that took the work: the public Deadlock API is the shared floor every tracker builds on, and a site built only on it can show what that API already contains and nothing more. Statlocker ingests well beyond it, through replay files decoded into per-second timelines and a native desktop companion that contributes matches and client-side detail the API never returns.
A great deal of it happens behind the scenes. Multiple data scientists develop the models behind our player ratings, MVP awards and match analysis, and people work purely on the interface. We know this much depth can be intimidating, so real care goes into keeping Statlocker beautiful and easy to find your way around, however far you want to take it. The questions worth answering are the hard ones. How well did you actually play? Where did you make the biggest difference? What could you have done better? None of those are visible in a KDA.
Standing on the shoulders of giants
Statlocker is partnered with Dotabuff, who spent well over a decade building the definitive stat site for Dota 2. For a new generation of Valve MOBA stat tracking there may be no team on earth better qualified to tell us how to execute it, and we lean on that experience constantly. A great deal of what we are building is shaped by lessons they learned the hard way, years before Deadlock existed.
Player profiles
A profile is six separate views of a player rather than one match list. Every tab below is a live example from a top-ranked account, so you can see the real thing rather than a description of it.
Matches
/profile/{account}
Full match history with per-match detail, filterable by hero, mode, result and patch. Opening any match gives the complete breakdown for all twelve players.
Performance
/profile/{account}/performance
Rating and rank over time, including Performance Points progression, and how each metric compares against the wider population rather than in isolation.
Heroes
/profile/{account}/heroes
Per-hero performance for everything the player has played: win rate, rating on that hero specifically, and how their hero pool is distributed.
Laning
/profile/{account}/laning
Laning phase performance separated from the rest of the match, graded against what the matchup would normally produce rather than against a flat average.
Players
/profile/{account}/players
Who this player queues with and against most, and how results change with each of them. Team games are not solo games, and this is the part most trackers omit entirely.
Awards
/profile/{account}/awards
MVP awards and accolades earned across a career, derived from our own match analysis rather than from a single scoreboard column.
Hero statistics
Ranked win rates
/heroes/ranked-winrates
Every hero ordered by win rate in ranked matches only, scoped to the current patch.
Standard win rates
/heroes/standard-winrates
The same ordering across every match mode. The two lists disagree more often than people expect.
Brawl win rates
/heroes/brawl-winrates
Win rates scoped to the Brawl mode, which has a meta of its own.
Tier list
/heroes/tier-list
Tier placement derived from win rate rather than assigned by opinion, so it moves when the game does. A Brawl tier list is maintained separately.
Win rate by rank
/heroes/winrate-by-rank
How each hero's win rate changes across skill brackets. A hero that is strong in expert hands and weak everywhere else looks average when those are pooled, and this separates them.
Patch movers
/heroes/patch-win-rates
Which heroes actually moved across the latest patch, measured rather than asserted from the notes.
Team composition win rates
/heroes/team-win-rates
How heroes perform together. Composition effects are invisible to a per-hero table by construction.
Laning matchups
/heroes/laning
Lane matchup outcomes across the roster, showing which lanes are actually favoured rather than which are assumed to be.
Hero mastery
/heroes/mastery
How much performance on a hero improves with experience on it, which varies enormously between heroes.
Community difficulty
/heroes/community-difficulty
How hard players themselves rate each hero to play, collected from the community rather than assigned by us.
Average performance stats
/heroes/performance-stats
Baseline per-hero statistics, which is what every percentile and comparison elsewhere on the site is measured against.
Damage breakdown
/heroes/damage-breakdown
Where each hero's damage actually comes from, split by source rather than reported as one total.
MVP statistics
/heroes/mvp-stats
Which heroes earn MVP most often, from our own MVP model rather than a kill count.
Hero library
/heroes/hero-library
Per-hero reference pages: abilities, itemisation, matchups and statistics together.
Items
Item win rates
/items/item-stats
Win rate and purchase frequency for every item, by hero and by rank.
Item meta model
/items/meta-model
Item effectiveness measured as win probability added rather than raw win rate. A naive item win rate mostly reflects who buys the item; this measures how much the probability of winning moved around the purchase, in comparable match states, with confidence intervals attached. Combinations that do not reach significance are not presented as findings.
Item library and shop
/items/items-library
The complete item database with stats, costs and upgrade paths.
Shop speedrun
/items/shop-speedrun
A timed game for learning the shop layout, which is a real source of lost seconds in game.
Builds
Build library
/builds/build-library
Browse and compare builds across every hero, with measured performance attached rather than a popularity ranking alone. Trending builds surface what is currently rising.
Eternus builds
/builds/eternus-builds
What the highest-ranked players are actually building, including the underrated choices that aggregate popularity hides.
Build lab
/builds/build-lab
Construct a build and see its computed stats, rather than guessing at how items combine.
Random build generator
/random-build
A constrained random build, for a challenge run rather than an optimal one. There is a random hero picker too.
Rank and rating
Statlocker computes its own rating, Performance Points, alongside tracking the game's visible rank. It rates how a player performs given the level they play at, rather than being trained purely on match outcomes.
Rank luck
/rank-luck
How much of a recent record came down to things outside your control, including the quality of the teammates you were given. Over a short sample a team-game record is substantially noise, and this quantifies how much of it was yours.
Rank distribution
/rank-distribution
How the playing population is distributed across ranks, which is the basis for every percentile shown elsewhere.
Leaderboards
Statlocker leaderboard
/statlocker-leaderboard
Ranked by Performance Points rather than by the game's ladder.
One-trick leaderboard
/one-trick-leaderboard
Players with the deepest specialisation on a single hero, which is a different question from who is highest rated.
Valve leaderboard
/valve-leaderboard
The official in-game leaderboard, tracked over time so you can see movement.
Match analysis
This is the part that does not come from a match summary. It is derived from replay files decoded into per-second timelines, and from models trained on them.
Anything marked Beta below is live and usable but still being developed, and its output may change as the model behind it improves.
Win chance
/vision/win-chance
Win probability modelled continuously across a match rather than reported at its end. The same model underpins the item meta analysis.
Playstyle retrospectiveBeta
/vision/playstyle-retrospective
A review of how you actually play, per match and across a career, rather than a scoreboard summary.
Deadlock puzzlesBeta
/vision/puzzles
Real match situations, reconstructed from replay data, presented as decision problems: given this position, what do you do. Scored against what actually happened, with a leaderboard.
Interactive map
/interactive-map
The Deadlock map rendered in three dimensions, used for positional review and for the puzzle system.
Match timeline
/match-timeline
A match played back over time rather than presented as end-state totals.
Recent and live matches
/recent-matches, /active-matches
Matches flowing through ingestion, and games currently in progress.
Favourites feed
/favourites-feed
Follow specific players and get a per-day summary of their games, rating movement and results rather than a raw match dump.
Competitive and community
Tournaments
/tournaments
Listings, brackets and results, with a calendar and a creator for running your own.
Guides
/guides
Written guides, including material for new players and for people coming from Dota 2 or from League of Legends.
Platform
Desktop companion
/deadlock-companion-app
A native desktop application rather than an Overwolf plugin, which means a much smaller footprint and no third-party runtime. It detects match completion locally and starts ingestion immediately instead of waiting for data to become publicly available, and reads the game's own local files so items and heroes resolve against the version you actually have installed.