Frequently Asked Questions
Everything you need to know about MODOMeta's tournament data coverage, metagame metrics, deck classification engine, and third-party credits.
1 Tournament Data & Coverage #
Is the Magic Online Challenge data complete? #
Daybreak Games (the studio operating MTGO) typically publishes only the top 32 decks from Magic Online Challenges, regardless of total tournament players. Whether a Challenge draws 64, 128, or 250+ players, only the top 32 decklists and standings are published in the official MTGO tournament results dumps.
Because finishes below 32nd place are omitted from official results dumps, our Challenge Share and metagame metrics reflect only these top 32 lists. Decks that finish outside the top 32 are effectively invisible to MODOMeta. For head-to-head match data, official dumps only include the 7 playoff bracket matches (Quarterfinals, Semifinals, Finals).
How does MODOMeta have so much data on Vintage and Legacy? #
For Legacy, we augment the official matchup data with Swiss results captured by the Legacy Data Collection Project for players with published decklists (typically top 32). Similarly, for Vintage, we use matchup data compiled by the Vintage MTGO Community. Across other formats, MODOMeta uses only the official published playoff results. This means we have about 10x the data on archetype and player performance in Vintage and Legacy compared with other formats.
Our figures, in the archetype matrix for example, may not precisely agree with data from the Legacy Data Collection Project or the Vintage MTGO Community, but they should be close. We do our own archetype classification based on the published decklist and as a result only count matchups between players finishing in the top 32, while those projects collect data on all matchups in the events. Most challenges in those formats have fewer than 64 players although showcase challenges usually have much more.
What about Preliminaries and Last Chance Qualifiers (LCQs)? #
Preliminaries (usually 4-round events) and Last Chance Qualifiers (LCQs) are largely excluded from MODOMeta's metagame statistics because there are relatively few of them in the dataset. While we ingest these events for tournament and player records, they are excluded from Challenge and League share metrics.
How do League 5-0 decks work? #
Daybreak publishes decks that achieved a 5-0 match record in constructed Leagues. Our League Share metric counts the number of 5-0 decks published for each archetype over the total 5-0 league decks. However, decks that don't achieve a 5-0 are not published, so we don't have visibility into the total number of league decks played.
Will you expand to cover paper events? #
Never say never, but probably not. The MTGO dataset is already cleaned (virtually no invalid cards, consistent player names) and pre-processed. There's also a clear hierarchy of the competitiveness of the events (Challenges > Leagues). Even though some paper event data are available (from Melee.gg for example), it would be a lot of work to cover them and MTGO by itself gives a pretty clear idea of the current metagame.
Can you add <Feature>? #
Create an issue to discuss how the feature would fit into the site. If it fits with the data MODOMeta already has and it's useful across formats, I'll probably add it.
Can I explore the data behind MODOMeta myself? #
The data for MODOMeta is built entirely from public sources: Scryfall card data a cache of MTGO tournament results, and player-driven data collected by the Legacy Data Collection Project and Vintage MTGO Community (see this repository for more details).
You can also download the exact database dump (~800 MB) used to serve MODOMeta. It's an SQLite database that can be opened with standard tooling.
2 Metagame Metrics Explained #
Conversion Rate #
Calculated as Top 8 Finishes / Challenge Appearances.
Measures an archetype's or player's win efficiency: how likely is a player or archetype to convert a run into a Top 8 appearance?
Since there are usually only 32 decklists posted per challenge, a conversion rate significantly above 25% (8 / 32) is very good.
Match Win % #
This stat gives a good idea of a player's strength in challenges although it will overstate a player's true match win percentage. Daybreak / WotC post full win-loss standings for the top 32 players in a challenge. If a player drops from an event after losing in the early rounds and places outside the top 32, those losses will not count against their match win percentage because their standings were never posted.
3 Classification & Color Inferences #
How does MODOMeta identify archetypes and colors? #
Decklists are processed through a multi-stage classifier:
- Explicit Archetype Rules: Formats define archetype signatures with mandatory cards, minimum signature card thresholds, and anti-signatures (cards that exclude a deck from an archetype, like preventing Delver lists with Bowmasters from matching Izzet Delver). The archetype signatures are public and powered by the open-source mtg-archetypes library.
- TF-IDF & k-Nearest Neighbors (kNN) Closeness: If a deck does not match an explicit rule, it is compared against verified classified decks in the same format using TF-IDF cosine similarity. Decks with ≥ 85% card similarity to an established archetype list inherit that classification.
- Tactical Posture & True Color Fallback: Decks not matching explicit rules or the 85% kNN closeness threshold fall back to tactical posture analysis (Combo, Tempo, Control, Aggro, Midrange) combined with true color deduction based on castable spells and mana-producing lands (e.g. “Abzan Midrange”, “Grixis Control”).
If you think a certain classification could be improved (without becoming too specific or brittle), feel free to create an issue or pull request on the mtg-archetypes repository.
4 About #
What does MODOMeta do that other aggregators don't? #
MODOMeta was built because the existing deck aggregators like MTGGoldfish and MTGTop8, while powerful and with features MODOMeta won't ever have, don't do the kind of deeper analysis on archetype strength I wanted to see. Specifically, I wanted Challenge conversion rates and separated League and Challenge play rates.
5 Credits & Acknowledgments #
MODOMeta would not be possible if Daybreak Games didn't publish as much data openly as they do. In addition, I'd like to thank the following projects for making it possible to build something like MODOMeta without having to build everything else as well:
Scryfall
Card dataCard metadata, imagery, legality rules, and mana costs are powered by Scryfall bulk data.
Mana Font
SymbolsAll MTG mana symbols, color icons, and card costs are rendered using the Mana font created and maintained by Andrew Gioia. These symbols are copyright Wizards of the Coast.
MTG Decklist Cache
ArchiveTournament decklists and tournament standings are sourced from the modometa-mtgo-data archive, building upon the format established by Francisco Bettega's MTG_decklistcache.
Legacy Data Collection Project
Detailed player matchup results and archetype head-to-head data would be far more limited without the LDCP archive of match results.
Vintage MTGO Community
Similar to the LDCP for Legacy, detailed matchup results are sourced from the Vintage MTGO Community player-driven spreadsheets.
mtg-archetypes
ClassifierDeclarative archetype classification rules, card requirements, and priority matching are powered by the mtg-archetypes library.