The numbers behind every move
DataPlane.pro is a data-first reference portal for strategy game players, researchers, and coaches. Opening statistics, engine evaluations, win rates, and historical match records — all in one place, built for people who study the game as seriously as they play it.
About This Portal
Built for players who want the full picture
Most game sites tell you what happened. DataPlane tells you why — and what the numbers say about what should have happened instead.
-
Not opinion-driven
Every claim on DataPlane is backed by a number, a dataset, or an engine evaluation. If we can't quantify it, we don't publish it.
-
Not a news site
We don't cover drama, rankings gossip, or tournament recaps. DataPlane is a reference resource — structured, searchable, and built to be useful on your tenth visit as much as your first.
-
Not a training app
We don't teach you openings with flashcards or gamify your study sessions. We give you the raw data, the evaluation context, and the historical record — and trust you to draw your own conclusions.
-
Who it's for
Club players and serious amateurs. Coaches building lesson plans. Researchers studying game theory. Enthusiasts who simply want to understand what the engines actually think — and why.
Coverage
Four games, one data layer
DataPlane covers the four pillars of abstract strategy — each with its own statistical framework, historical record, and engine evaluation methodology.
-
Chess
The most analyzed game in human history — and still full of unanswered questions. DataPlane covers opening statistics by ECO code, win rates by color across skill brackets, engine evaluations from depth 20 to depth 40, and historical game records from 1850 to the present day.
-
Go
The game with more possible positions than atoms in the observable universe. We cover opening joseki frequencies, win rates by komi setting, AI model comparisons (AlphaGo, KataGo, Leela Zero), and territorial analysis across professional game archives.
-
Checkers
Solved in 2007 — but that doesn't mean there's nothing left to study. DataPlane documents the complete game tree summaries, variant rule sets, historical championship records, and the journey from human mastery to computational proof.
-
Abstract Strategy Games
Othello/Reversi, Hex, Amazons, Connect Four, Nim, and more. Each game documented with its known solved status, branching factor, optimal strategy summaries, and competitive records where they exist.
What We Track
Six statistical lenses on every game
DataPlane organizes its data into six core categories — each answering a different question serious players actually ask.
-
Opening Statistics
Which openings win most often at which level? How does win rate shift between amateurs and grandmasters? How has opening popularity evolved decade by decade? We track frequency, performance, and trend data for every major opening system.
-
Engine Evaluations
What does the machine think? Engine scores at multiple search depths, comparison across engines (Stockfish, Leela Chess Zero, KataGo), and evaluation breakdowns by game phase. Understand not just the verdict but the reasoning behind it.
-
Win Rates by Color / Side
First-move advantage is real — but how real, and does it change by time control, skill level, or opening choice? DataPlane quantifies the color advantage across all skill brackets and game formats with statistical confidence intervals.
-
Historical Match Records
Archival data from competitive play stretching back over a century. Filter by player, year, event, opening, or result. Explore how the game has changed — and which ideas have stood the test of time.
-
Endgame & Tablebase Data
Precise outcomes from known endgame positions. Forced wins, draws under optimal play, distance-to-mate calculations. The part of the game where opinion ends and mathematics takes over completely.
-
Comparative Game Theory
Cross-game analysis — how branching factors compare, what solved games reveal about unsolved ones, how AI performance scaled differently across Chess vs. Go vs. Checkers. For the analytically curious.
Understanding the Numbers
What engine scores actually mean
Engine evaluations are the most powerful tool in modern game analysis — and the most frequently misread. Here's how DataPlane presents them and what to look for.
-
Centipawn loss (CPL)
The standard measure of move quality in Chess. One pawn equals 100 centipawns. Average centipawn loss per move tells you how closely a game was played to engine-perfect. DataPlane shows CPL by phase — opening, middlegame, and endgame — so you can see exactly where decisions diverged from optimal play.
-
Win probability curves
Rather than a single evaluation number, DataPlane shows win probability as a curve across the entire game. Watch the moment a position tipped — not after the fact, but move by move. Drawn from neural network evaluations that mirror how modern engines actually assess positions.
-
Depth matters
An evaluation at depth 20 and depth 40 can disagree significantly in complex positions. DataPlane always shows the search depth alongside any evaluation, so you know how much to trust the number you're looking at.
-
Engine disagreement
When Stockfish and Leela Chess Zero assess the same position differently, that's often where the most interesting chess lives. DataPlane flags positions where top engines diverge — these are the positions worth studying most carefully.
-
Go-specific: policy and value networks
For Go, DataPlane presents both the policy network output (what move the AI prefers) and the value network output (who is winning and by how much). Understanding the difference between the two is essential for reading AI analysis in Go correctly.
The Archive
A century of decisive games
DataPlane maintains a structured archive of competitive games across all covered titles. Not a raw PGN dump — a searchable, filterable, annotated record with statistical context attached to every game.
-
Go — the AI turning point
The 2016 AlphaGo matches didn't just produce results — they revealed that human Go theory had systematic blind spots at the highest level. DataPlane charts the win-rate implications of AI-introduced moves that professional players had considered suboptimal for decades.
-
Chess — 1850 to present
From the Romantic era of gambits and sacrifices to the engine-assisted precision of modern super-tournaments. Search by player, event, year, ECO code, or result. Every game includes engine evaluation overlay and opening classification.
-
Milestone games documented
Deep Blue vs. Kasparov 1997. AlphaGo vs. Lee Sedol 2016. Marion Tinsley's undefeated Checkers record. The moments where human and machine understanding met — annotated with the data that made them decisive.
-
Opening evolution over time
How has the Sicilian Defense's win rate changed over 60 years? When did the King's Indian fall out of top-level favor and why? DataPlane's historical opening charts answer these questions with actual data, not retrospective opinion.
-
The Checkers proof
In 2007, Jonathan Schaeffer's team at the University of Alberta proved that perfect Checkers play always results in a draw. DataPlane documents the game tree statistics, the computational scale of the proof, and what it means for how we understand the game today.
Our Readers
Serious players. Curious minds.
DataPlane is built for people who find the question “what does the data say?” more interesting than “what does the expert think?”
-
Club & tournament players
Use opening statistics and win-rate data to build a repertoire grounded in performance numbers, not just style preferences. Know which lines give you the best practical results at your specific rating range.
-
Coaches & instructors
Build lesson plans around engine evaluation data. Show students exactly where their guesses diverge from optimal play — and quantify improvement over time with CPL tracking.
-
Game researchers & academics
DataPlane's structured datasets and historical archives support research in game theory, AI development, cognitive science, and the mathematics of abstract strategy. Cite real numbers, not anecdotes.
-
Developers & designers
Building a strategy game? DataPlane's cross-game branching factor analysis and first-move advantage data give you a rigorous framework for understanding balance and depth in your own designs.
-
Enthusiasts & lifelong learners
You don't need a rating to find this stuff fascinating. If you've ever wondered why computers play h4 in positions that look completely quiet, DataPlane has the answer — and the data behind it.