iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
DeepMind’s AlphaStar reached Grandmaster level in StarCraft II and ranked above 99.8% of officially ranked human players, according to a result announced in October 2019. That figure describes its position on the player rankings—not the percentage of matches it won.
What did AlphaStar’s 99.8% result mean?
The Nature paper reports that AlphaStar was “above 99.8% of officially ranked human players.” In other words, its rating placed it higher than nearly all players included in the official ranking. It does not mean AlphaStar won 99.8% of its games. The paper does not define the statistic as a match win rate.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
StarCraft II: Wings of Liberty | $7.59 | Buy on Amazon |
| 2 |
|
Starcraft II: Legacy of the Void - Standard Edition | $42.94 | Buy on Amazon |
| 3 |
|
Starcraft II PC | $21.45 | Buy on Amazon |
DeepMind’s October 30, 2019 announcement described the result as ranking “above 99.8% of active players on Battle.net.” The paper’s more specific wording is “officially ranked human players.” Neither phrasing supports a claim about every person who has ever played StarCraft II.
What level did AlphaStar reach?
AlphaStar achieved Grandmaster level for all three playable StarCraft II races: Protoss, Terran, and Zerg. Grandmaster is the game’s highest competitive league. The Nature authors say they evaluated the agent in the full game through online matches against human players.
#1 Best Overall
- Fast-paced, hard-hitting, tightly balanced competitive real-time strategy gameplay that recaptures and improves on the original game
- Three completely distinct races: Protoss, Terran, and Zerg
- Units and gameplay mechanics distinguish each race
- 3D-graphics engine with support for visual effects and massive unit and army sizes
- Full multiplayer support, with competitive features and matchmaking utilities available through Battle.net
How was AlphaStar evaluated?
DeepMind said AlphaStar played on the official Battle.net game server, using the same maps and conditions as human players. This matters because the reported achievement came from online competition under the game’s ordinary competitive environment, rather than from a separate benchmark described in the announcement.
The result was reported in 2019; it is not a claim that a newly released AI system achieved this ranking in 2026.
Rank #2
- This is a standalone product. It does not require any other version of StarCraft II to play
- Internet Connection Required
- Battle.net registration and Battle.net Desktop Application required
How did AlphaStar learn to play?
AlphaStar’s result came from a multi-agent learning approach. The Nature paper describes training that drew on both human games and agent games, using a league of agents whose strategies and counter-strategies adapted over time. Each agent was represented by a deep neural network.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
DeepMind’s announcement adds that training began with agents trained through supervised learning and then used a fully automated league. The published result establishes the overall approach, but the cited summary sources do not provide match counts, rating uncertainty, or the full details of the ranking calculation.
Rank #3
- Videogame Software
What does the result establish—and what does it not?
- It establishes: In the 2019 evaluation, AlphaStar reached Grandmaster level across Protoss, Terran, and Zerg and ranked above 99.8% of officially ranked human players.
- It does not establish: A 99.8% match win rate, human-level intelligence in general, or equivalent performance in other games or real-world tasks.
The finding is a specific achievement in StarCraft II. It should be understood as a competitive ranking result, not a general measure of AI capability.
Quick Recap
Sources
- Google DeepMind’s October 30, 2019 AlphaStar announcement
- The Nature paper, “Grandmaster level in StarCraft II using multi-agent reinforcement learning”
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
