Competitive Analysis - 46,000 Simulated Games across Gen 1 to Gen 9
🕹️ Play the Game 📊 Original Gen 1 AnalysisThe game now lets you pick any mix of generations before a match. This study tests every option.
Every game follows the live game's rules exactly: 6 v 6 knockout, a random stat each round, primary-type multipliers (1.2x / 0.8x), and the blocked-stat rule after a draw. Both sides pick stats at random, so the results measure the cards, not player skill.
Inside a single generation the match is fair: each side wins about half the time. The real difference shows when generations face each other. The newer generations (Gen 4, 7 and 9) carry higher stat totals and beat the original three generations about two times out of three.
Game win rate of the row generation's team against the column generation's team. 1,000 games per pairing, with sides swapped every game so going first never favours either generation.
Share of rounds won by Pokemon of each primary type, across all 46,000 games. Bars show the round win rate; the number of Pokemon with that primary type is in brackets.
Round win rate for a primary type facing another primary type. Only matchups with at least 400 rounds are shown, so every bar rests on solid data.
Round win rate for every one of the 1,025 Pokemon, pooled across all tests. Every Pokemon played at least 414 rounds.
A card's stat total (Attack + Defense + Sp. Attack + Sp. Defense + Speed, HP is not used in the game) predicts its win rate almost perfectly: the correlation is 0.98. Type matters at the margins, but a big stat total matters most.
Note: Random play baseline. 1,000 games per test gives a margin of about ±3 percentage points on a game win rate, so differences smaller than that (for example Gen 8's 46.6% side win rate) are noise.