Commander Goldfish Simulator
See how reliably your deck makes land drops, casts ramp, and reaches the mana it needs.
This measures early-game consistency, not deck strength or multiplayer win rate.
Paste a Moxfield or Archidekt URL, upload a file, or load the sample deck to get started.
Decklist
Paste plain text. Put commanders in the commander section.
Edit decklist
Results
Run a simulation to see your deck's consistency
Paste a decklist or URL above, choose a test, and run the simulation to see your early-game consistency.
- Import a deck
- Select a test
- Run the simulation
The simulation could not finish
Primary metrics
More at a glance
Consistency by turn
See how land drops and ramp affect your deck across the opening turns.
Why games failed
View detailed breakdown Confidence, mana distribution, sample turns
Confidence interval
Keep-rate breakdown
Mana distribution at the target turn
Detected ramp cards
Cards not included in ramp calculations
0 excludedThese were not counted as ramp, so the result may slightly underestimate the deck.
Cards that could not be matched
0 unresolvedSimulation assumptions
Sample turn sequences
Up to 5 sample games, described in plain language.
Run the simulator to see sample turns.
What a casual Commander goldfish simulator measures
This tool runs solitaire test games for an ordinary Commander deck and reports how reliably it develops its mana in the first several turns. It does not model opponents, combat, or table decisions, and it is not a win-rate predictor. Instead it answers practical questions: how often do you make your first four land drops, how often can you produce four mana by turn four, and how often does an unusually swingy hand show up as an early screw or flood.
Every run is a batch of browser-side simulations, from 250 to 5,000 games. The default health check calculates every major metric in one pass, so you never need to run five separate checks to get a full picture.
How land-drop consistency is calculated
Land-drop consistency is the percentage of simulated games that successfully play a land on every turn through the selected turn, plus a per-turn breakdown so a slipping curve is visible at a glance. A deck that plays its land every turn through turn four, but not turn five, will show a high early number that starts to fall later, which is a useful signal on its own.
How ramp is detected
A short list of extremely common staples is recognized directly. Everything else is classified conservatively using structured Scryfall data and oracle text: a mana rock or dork must have a real, repeatable tap ability, and a land-ramp spell must put a land onto the battlefield, not just into your hand. A card only counts as ramp once it has actually been cast under the simplified sequencing policy, not merely drawn.
Not every card containing the word Treasure or mana is treated as ramp. Combat-gated or conditional effects this simulator cannot verify are excluded outright and listed under Cards not included in ramp calculations, along with a plain-language reason, so the result never silently invents a card's behavior.
What flood and mana screw mean in this tool
Mana screw, by default, means a simulation misses a land drop at or before turn three. Flood uses a few fixed checkpoints: five or more lands in the opening hand, five or more lands seen by turn four, and six or more lands seen by turn five. Both are adjustable in Advanced settings, and a deck that makes all of its land drops is not automatically labeled flooded just for doing so.
Why simulation results differ from real games
This is a goldfish: one deck, no opponents, and a simplified sequencing policy that plays a land every turn, then casts the best available ramp with the mana on hand. Real games add interaction, mulligans made with table information, and choices this simulator cannot see. Treat the result as a consistency signal for the deck's own resources, useful for comparing two versions of a list rather than predicting any single game.
How to compare two deck versions
Run the simulation once. Edit the decklist, import a different version, or adjust the mana base, then run it again. The result panel shows the difference from your previous run for the main metrics, labeled Compared with your previous run, stored locally in your browser with no account needed. Use Reset comparison to clear it.
Related Commander tools
If a metric here points at a specific fix, one of these tools goes deeper.
Commander Goldfish Simulator FAQ
What does a casual Commander goldfish simulator measure?
It measures the health and consistency of a deck's early mana development: how often it makes its land drops, casts ramp on curve, reaches a target amount of mana by a chosen turn, and how often it gets mana screwed or flooded. It is not a win-rate simulator.
How is land-drop consistency calculated?
Across every simulated game, it is the percentage that successfully play a land on every turn through the selected turn, plus a per-turn breakdown so you can see exactly where drops start slipping.
How does the simulator detect ramp?
Common staples are recognized directly. Everything else is classified conservatively from Scryfall card data and oracle text: only real, repeatable tap abilities and land-search-to-battlefield effects count. A ramp card must be legally cast to count, not merely drawn. Anything that cannot be classified with confidence is excluded and listed under Cards not included in ramp calculations, so the result may slightly underestimate the deck rather than invent behavior.
What do flood and mana screw mean in this tool?
Mana screw, by default, means a simulation misses a land drop at or before turn three. Flood is tracked at a few fixed checkpoints, such as five or more lands in the opening hand, or five or more lands seen by turn four. Both thresholds can be adjusted in Advanced settings.
Why does simulation results differ from real games?
This tool goldfishes a single deck with a simplified sequencing policy and no opponents, interaction, or table decisions. Treat the result as a consistency signal for the deck's own resources, not a prediction of any specific game.
How do I compare two deck versions?
Run the simulation once, edit or re-import the deck, then run it again. The result panel shows the difference from your previous run for the main metrics, stored locally in your browser with no account needed.