Your MTG Deck Doesn’t Have a Power Level. It Has Enemies.

Published 2026-07-06

Most Magic players talk about decks like they have one true power level.

“This is a seven.”

“This is casual.”

“This one is tuned.”

“This one is disgusting and should be buried in a shallow grave behind the LGS.”

But after looking at 4,914 simulated games from 229 player-submitted decks, the Grim.Cards data points to a more uncomfortable truth: your deck may not be “good” or “bad” in the abstract. It may just be great into one table, dead into another, and pretending otherwise is how cardboard gets expensive. The dataset covered simulations from May 14 through July 2, 2026, including 201 Commander decks and 28 Standard decks submitted by 109 users. Commander decks posted a 45.7% win rate across 4,415 games, while Standard decks posted 37.3% across 499 games.

That overall win rate is not the interesting part. In a fixed gauntlet where one challenger deck runs into multiple meta opponents, a sub-50% average is not automatically a sign that the submitted decks are trash. The gauntlet is designed to kill things. That is its whole miserable job. The interesting part is where the bodies piled up.

The Same Commander Deck Pool Beat Breya — and Got Mauled by Atraxa

The Commander gauntlet tested the same 201 submitted Commander decks against five opponents: Breya Artifact Combo, Derevi Bant Control, Aesi Landfall, Edgar Markov Vampires, and Atraxa Superfriends. Because every deck faced every opponent, the matchup table gives us one of the cleanest signals in the whole report: matchup identity mattered a lot.

Against Breya Artifact Combo, submitted Commander decks won 54.7% of games. Against Derevi Bant Control, they won 53.2%. Those are winning records. The same pool of decks then dropped to 47.3% against Aesi Landfall, 36.1% against Edgar Markov Vampires, and 36.0% against Atraxa Superfriends. That is an 18.7-point spread between the easiest and hardest Commander matchups, using the same 201-deck challenger pool.

That is not a “red decks good” or “blue decks bad” kind of finding. In fact, color identity had a much narrower spread than matchup identity. Red-containing Commander decks led at 48.6%, while black- and white-containing decks tied at 43.9%, a spread of only 4.7 points. Meanwhile, the difference between Breya and Atraxa was almost 19 points. In this dataset, what your deck faced mattered more than what colors your deck contained.

That should change how players think about testing. A deck that looks impressive against one archetype may be quietly doomed against another. A deck that beats value engines may fold to planeswalkers. A deck that survives control may not survive vampire pressure. “My deck is strong” is not specific enough. Strong into what?