Great question! I don't have per-card winrate data independent of decks — the database tracks match results at the deck/entry level, not at the individual card level. So I can't isolate a card's contribution to wins in isolation from the rest of the deck it's in.
What I can do is show you the most-played cards across all Modern decks right now, which is the best proxy for "top performing" cards format-wide — high presence cards tend to be in winning decks. Let me pull that data for you.
To be upfront about limitations: the database tracks wins and losses at the deck level, not the individual card level. There's no way to isolate how much a single card contributed to a win independent of the rest of the deck. What I can offer instead is format-wide card presence — how widely a card is played across all Modern decks — which is a strong signal of perceived power and real-world impact.
Here are the top 20 most-played mainboard cards in Modern over the last ~60 days (March 1 – April 28, 2026):
Phlage, Ragavan, and Galvanic Discharge stand out clearly at the top. Cards with near-4.0 average copies (Ragavan, Galvanic Discharge, Guide of Souls, Ocelot Pride, Ajani, Mishra's Bauble) are near-universal 4-ofs wherever they appear, signaling format staples that players are fully committed to.
Sources: Data drawn from 130 Modern tournaments (March 1 – April 28, 2026), 75% MTGO and 25% Melee.
Recent events include:
Great question! This requires a custom analysis. The approach is: for each card in an archetype, compare the winrate of entries that include that card vs. entries that don't. Let me pull the data needed to do this properly.
To keep it meaningful, I'll focus on Modern's top archetypes over the last 60 days and look for cards where the "with card" winrate meaningfully exceeds the "without card" winrate, weighted by sample size.
Good. Now I'll run the core analysis: for each of the top archetypes, compute per-card "with vs. without" winrates across all entries. I'll do this in a single SQL query across all archetypes simultaneously, filtering for cards with meaningful sample sizes on both sides.
Let me rewrite that with a proper self-join approach that avoids the correlated subquery issue:
Now let me get sources for citation:
Great data. Here is the full analysis:
This is a "value over replacement" (VoR) analysis for Modern's top 4 archetypes (Boros Energy, Jeskai Blink, Affinity, Amulet Titan) over the last 60 days. The metric is delta_vs_overall: the difference in winrate between entries that include a given card vs. the archetype's baseline, requiring at least 20 entries on each side of the comparison.
TOP POSITIVE CONTRIBUTORS (cards that meaningfully raise their archetype's winrate):
Marsh Flats in Jeskai Blink: +7.6 pp delta (58.5% with vs 50.2% without, 23 vs 318 entries). The sample is small but the signal is striking. Marsh Flats-running lists appear to be a tuned subvariant, likely white-heavy splash builds.
Thundering Falls in Affinity: +6.5 pp (58.9% with vs 51.3% without, 34 vs 232 entries). A clear outlier in the Affinity shell; lists running this land consistently outperform the field.
Preordain in Affinity: +4.7 pp (57.0% vs 51.8%, 26 vs 240). Blue-cantrip Affinity variants appear noticeably stronger.
Scalding Tarn in Affinity: +4.5 pp (56.9% vs 51.4%, 43 vs 223). Correlates with the blue-splash Affinity builds, likely bundled with Preordain.
The Wandering Minstrel in Amulet Titan: +3.7 pp (59.5% vs 55.2%, 31 vs 143). Strong signal for a tech inclusion in Titan lists.
Windswept Heath in Boros Energy: +2.9 pp (53.7% vs 47.9%, 335 vs 359). Remarkably clean split with large sample on both sides, making this the highest-confidence positive signal in the dataset.
NOTABLE NEGATIVE CONTRIBUTORS (cards associated with below-baseline performance):
Static Prison in Boros Energy: -2.2 pp (48.6% with, 51.1% without, 112 vs 582). A frequently-played card that appears to drag down performance. March of Otherworldly Light in Jeskai Blink: -2.4 pp. Cultivator Colossus in Amulet Titan: -2.1 pp. Thoughtcast in Affinity: -3.5 pp.
IMPORTANT CAVEATS: This is correlation, not causation. Cards don't independently cause wins; they cluster with build choices, player skill, and metagame positioning. Small "with" samples (Marsh Flats at 23, Thundering Falls at 34) mean those deltas have wide confidence intervals. Treat this as a hypothesis generator, not a definitive ranking. Cards near 0 delta that appear in nearly every list (e.g., Seasoned Pyromancer at 670/694 entries) are effectively the archetype's core and simply can't be split meaningfully.
Sources: Data from 134 Modern tournaments (Feb 27 - Apr 28, 2026), 75% MTGO / 25% Melee.
Great question! Let me dig into this properly. I'll need to identify the Boros Energy archetype, find entries with/without Windswept Heath, and compare player skill levels across both groups.
Good, I have both IDs. Now let me pull all Boros Energy entries with and without Windswept Heath, including player info and records, over the last 60 days.
Good, that confirms a ~5.8pp winrate gap. Now let me look at player-level performance in both groups, and also check if the same players appear in both groups (playing with and without Heath across tournaments).
Excellent data. Now let me identify "crossover" players who appear in BOTH groups and compare their relative performance with and without Heath, to isolate the player skill confound.
Now let me compute the aggregate winrate of the heath group vs. no-heath group for players who ONLY appear in one group (pure groups), vs. the crossover analysis, to properly assess skill confounding.
This is very illuminating. Now let me get the top "heavy hitters" in the Heath group who contribute most to the overall winrate gap, focusing on high-volume players with elite records.
Now I have everything I need. Let me also quickly check the aggregate picture for crossover players specifically (their with-heath vs without-heath performance).
Now let me get sources to cite.
Excellent. Here is the full analysis:
This is a well-formed question and the data gives a nuanced answer: yes, player skill is a meaningful confounder, but it does not fully explain the Windswept Heath winrate gap.
The raw split over the past 60 days is 53.66% winrate for Heath pilots vs. 47.85% for non-Heath pilots across 824 total Boros Energy entries, a gap of +5.8 percentage points. When we isolate only players who appeared in BOTH groups across different tournaments (26 players who sometimes ran Heath and sometimes did not), those same players posted 55.51% with Heath and 48.55% without it, a +6.96pp gap in the within-player controlled sample. Importantly, this within-player gap is actually larger than the raw gap, which is the opposite of what pure skill confounding would produce. If Heath were just a skill proxy, controlling for player identity should shrink or eliminate the gap. It does not.
That said, players who exclusively play Heath (159 players) do skew stronger overall: their total Boros record across all configurations averages 53.25% vs. 47.69% for exclusive non-Heath players. So a skill selection effect is present at the population level, just not enough to explain the card's contribution.
The players contributing the most wins above the non-Heath baseline (47.85%) in the Heath group are:
Rashek and DcNo are the standouts due to high volume and consistent strong performance exclusively with the Heath build. Kingatlas has a stunning record but limited entries. The crossover analysis confirms that Heath itself (or the list configuration it implies) is doing real work above and beyond pilot quality.
Sources: 824 Boros Energy entries across 44 Modern tournaments from Feb 27 to Apr 28, 2026. Data is 79% MTGO, 21% Melee.
Great question! Let me pull the Modern meta to identify the top archetypes, then run with/without winrate splits for each archetype's most-played non-land cards.
Now let me get the archetype IDs and top non-land cards for each of these 6 archetypes:
Ruby Storm resolved to Pauper — let me get the Modern version. Now let me get archetype cards and run with/without winrate queries for the top 5 Modern archetypes simultaneously:
Now let me run with/without winrate queries across all 5 archetypes. I'll focus on non-land cards with variable presence (30-85%) where the split is most meaningful. For each archetype, I'll query wins/losses for entries with vs without a selection of key variable non-land cards:
Let me fix the SQL — the issue is with the apostrophes in card names and WITH...SELECT queries. Let me rewrite these properly:
Got all the card IDs. Now let me run the with/without winrate splits for all 5 archetypes:
Now I have all the data. Let me also get sources to cite:
Excellent! Now I have everything needed for a comprehensive ranking. Here's the full value-over-replacement analysis:
VALUE OVER REPLACEMENT: TOP NON-LAND CARDS BY WINRATE CONTRIBUTION IN MODERN (Feb 27 - Apr 28, 2026)
Methodology: For each of the 5 most-played Modern archetypes, entries are split into "with" vs "without" the card in the mainboard. The delta is winrate(with) minus winrate(without). Cards are ranked by that delta, filtered to exclude lands and requiring at least 20 entries in both groups.
RANKING (delta = WR with card minus WR without card):
Negative delta / cards that don't help (or hurt):
Key takeaways: The biggest positive contributors are Tron prison pieces — Trinisphere, Vexing Bauble, and Mind Stone consistently lift that archetype's below-average winrate by 4-6pp each. In Affinity, Skateboard and Emry are the clearest upgrades from baseline. The most striking negatives are Thoughtcast and Explore, both of which appear in less successful builds of their respective archetypes — likely a case of those cards appearing in greedier or more experimental lists that underperform the stock configuration. The Galvanic Discharge result in Jeskai Blink is visually striking at +7pp but the without-sample is only 26 entries, so confidence is lower there.
Important caveat: These deltas reflect correlation, not causation. Cards appearing in a minority of builds may correlate with specific sub-archetypes or tech choices rather than the card itself driving the winrate difference.
Sources: Data from 134 Modern tournaments Feb 27 - Apr 28, 2026 (75.4% MTGO, 24.6% Melee).
https://melee.gg/Decklist/View/a6254593-6bd7-477b-9708-b43800fd2448
https://melee.gg/Decklist/View/95aa4d66-f67f-45ab-b6ab-b4380001b463
https://www.mtgo.com/decklist/modern-challenge-32-2026-04-2612840566
Great question! To find the top performing players across all formats, I need to query the database directly. Let me pull win rates, total matches, and archetype breakdowns across all formats without a time restriction.
Now let me get the top players filtered by higher match thresholds to find the most statistically significant performers, and also pull their primary archetypes:
Now let me get the most statistically significant players - those with the highest match counts and best winrates across all time:
Here is a comprehensive breakdown of the top performing players in the database, divided by format. I am applying a minimum threshold of 80 total matches to ensure statistical significance.
MODERN (top performers, 80+ matches)
Peter Husisian leads all Modern players with a 74.5% winrate over 98 matches across 14 entries, playing exclusively Amulet Titan. He is the most dominant single-archetype specialist in the format. Joseph Puglisi follows at 69.3% over 137 matches, the highest match volume in Modern, cycling through Azorius Control, Grinding Breach, Goryo Reanimator, and others, showing exceptional adaptability. JodyK (68.8%, 128 matches) played Martyr Control, Tron, and Azorius Control. Rafael Gomes (rafaixa) stands out at 71.0% across 124 matches, built almost entirely on Domain Zoo and Jeskai Blink. Robert Seder (66.0%, 153 matches) is the volume king in Modern, exclusively playing Blink variants (Azorius, Esper, Jeskai, Orzhov). Gobern (65.6%, 212 matches) has the highest match count of any player in any format tracked, grinding Eldrazi variants relentlessly.
LEGACY (top performers, 80+ matches)
Felipe Missio is the clear standout: 69.0% winrate over 255 matches across 72 entries, the single highest match count in the entire database. He plays an enormous range of archetypes including Dimir Reanimator, ANT, TES, Grixis Delver, and Dimir Tempo. Leonardo Bacci (69.2%, 172 matches) is close behind, specializing in Mystic Forge Combo, Mono Red Prison, and ANT. Sprouts (67.6%, 142 matches) is a Lands and Selesnya Depths specialist.
PAUPER (top performers, 80+ matches)
Paolo Donfrancesco (71.5%, 123 matches) plays a wide variety of archetypes and is the strongest performer in Pauper by winrate at volume. luccianolf (66.7%, 141 matches) plays Tron variants and several others. Nicoló Grasso (65.1%, 169 matches) has the highest Pauper match count, playing nearly every archetype in the format. kiko2509 (66.7%, 126 matches) focuses on Azorius Familiars and Mono Blue Faeries.
STANDARD (top performers, 80+ matches)
Gerschi (69.4%, 121 matches) plays the widest Standard field with 9+ archetypes. Ale_Mtg (65.4%, 188 matches) leads Standard in total volume, playing mostly Izzet variants and is arguably the most prolific Standard grinder in the data. JodyK also appears here at 68.1% over 119 matches, making JodyK one of the only players to sustain above-average winrates at high volumes in two separate formats (Modern and Standard). Neil Estrada (66.7%, 114 matches) focuses on Izzet Cauldron and Landfall.
DUEL COMMANDER (top performers, 80+ matches)
samsherman (66.4%, 116 matches) is the most consistent Duel Commander player, exclusively playing Azusa. kasa (65.2%, 138 matches) plays Yoshimaru, Magda, and Cloud, and has the highest match count in the format.
Notable Multi-Format Performers
JodyK appears in both Modern (68.8%, 128 matches) and Standard (68.1%, 119 matches) with strong winrates in both, a rare feat. Ale_Mtg similarly appears in both Standard (65.4%, 188 matches) and Pioneer with competitive results.
Sources: Data sourced from the full tournament database with no date restriction. All winrates exclude draws and require a minimum of 80 total match results for statistical relevance. Represents primarily MTGO league and challenge data alongside paper event results ingested via Melee.
Let me pull together all the data needed for this analysis. I'll need the Boros Energy archetype details, format info, and then run detailed card-level queries.
Good. Now let me query the database for detailed card-level breakdowns — comparing mainboard and sideboard cards between Heath and non-Heath Boros Energy entries, including win rates for each card cohort.
Good, baseline confirmed. Now let me pull card-level breakdowns for both cohorts across mainboard and sideboard:
Now let me get the VOR for each of these cards — the winrate of entries that include each card vs those that don't, within the Heath cohort specifically:
Now let me also check the same cards' presence diff but also compute VOR within the NON-heath cohort for comparison, and grab sources:
Here is the full picture. The analysis compares which cards strongly correlate with the Heath build and whether those cards actually contribute positively to winrate within that cohort versus the no-Heath cohort.
Key findings from the Heath cohort (395 entries, 53.66% WR) vs no-Heath cohort (429 entries, 47.85% WR):
The two cards most strongly correlated with the Heath build that also show genuine positive VOR within that cohort are Ranger-Captain of Eos (main) and Blood Moon (main). Ranger-Captain appears in 64% of Heath lists vs only 47% of non-Heath lists (+17.2pp presence gap) and delivers +4.6pp VOR within the Heath cohort (55.3% with vs 50.7% without). Blood Moon main appears in 72% of Heath lists vs 64% non-Heath (+8pp) and contributes +4.9pp VOR (54.9% vs 50.0%). These two cards seem genuinely linked to the Heath package's better configuration: Heath fetches Plains to enable Blood Moon protection and supports the white mana for Ranger-Captain, so they form a coherent sub-package.
On the sideboard, The Legend of Roku (side) is highly correlated with Heath lists (78% vs 63% in non-Heath, so +15pp) and shows +6.0pp VOR within the Heath cohort (54.9% vs 48.8%). Meanwhile, High Noon is strongly anti-correlated: it appears in only 51% of Heath lists vs 68% of non-Heath lists (-17.2pp), and within the non-Heath cohort it still contributes +4.4pp VOR -- meaning it is a card that helps the weaker build compensate, but Heath pilots actively move away from it. Similarly, Vexing Bauble is less common in Heath lists (-10pp) and actually has negative VOR within the Heath cohort (-5.1pp), confirming that Heath players are correct to cut it. Voice of Victory is the one card where the no-Heath cohort shows a strong negative signal: non-Heath lists rely on it more (78% vs 95% presence), but it actually drags WR in both cohorts when present, suggesting it is a filler card that better-configured Heath lists have less room for.
The most plausible causal story: the Heath build is a more tuned, coherent configuration -- Plains-based manabase enabling Blood Moon main, Ranger-Captain as a toolbox tutor, and Legend of Roku in the side as a flexible threat -- rather than a collection of individually powerful cards propped up by skilled pilots. The presence gap and consistent VOR across those three cards point to a genuine archetype variant, not a player skill artifact.
Sources: 128 Boros Energy tournaments Feb 27 - Apr 28, 2026 (79% MTGO, 21% Melee).
Great question. This requires a multi-step approach:
Let me pull the data needed for this across all formats.
The wins/losses appear to be returning 0 — the match result field likely uses different values. Let me check the schema quickly.
Got it — results are WIN/LOSS/DRAW. Let me rerun the full analysis with the corrected values.
Excellent data. Now let me get sources to cite.
Here is the full breakdown of the methodology and results.
Methodology: For each player-archetype combination with at least 20 non-mirror matches, we compute their personal winrate. We then subtract the archetype's global baseline winrate (across all players with that archetype, same mirror exclusion). This gives a skill delta in percentage points (pp), representing how much better or worse a player performed compared to the field on the same deck. Where a player played multiple archetypes, deltas are match-count-weighted and aggregated. The minimum threshold is 40 total qualifying matches to appear.
Important caveats: Small sample sizes inflate this metric heavily. Players near the minimum threshold with unusually good runs will score very high. The metric does not distinguish deck selection skill from piloting skill -- it only removes the "I played the best deck" advantage by normalizing to the deck baseline. Players who also chose a stronger deck version (e.g., a tuned 75 vs stock) may still benefit.
Top players by skill delta (winrate above deck baseline), across all formats:
Modern standouts: Harrison Kanfer (+29.9pp, 40 matches, Nadu + Jund Creativity at 83.8% overall). Felix Marchel (+25.4pp, 53 matches, Grinding Breach at 81.1%). Andrew Sullano (+24.9pp, 61 matches, Affinity at 76.3%). Rafael Gomes / rafaixa (+24.1pp, 94 matches, Domain Zoo + Burn). JodyK (+22.0pp, 86 matches, Living End + Martyr Control). Peter Husisian (+20.6pp, 95 matches, Amulet Titan at 72.8%) -- notable because his sample is one of the largest, making his delta more credible.
Standard: Maxx Kominowski (+36.9pp, Izzet Lessons at 89.2%), though 42 matches is a thin sample. Marcelo Rodrigues Cavalcante (+32.5pp, Dimir Midrange, 52 matches). jtl005 (+25.2pp, 70 matches across two archetypes) is the most reliable standard entry by sample size.
Legacy: Sprouts (+20.5pp, 139 matches, Lands at 71.3%) -- highest sample size in Legacy and a very credible signal. Leonardo Bacci (+23.6pp, 90 matches, Mystic Forge Combo + Mono Red Prison).
Pauper: Igor Grecia (+24.3pp, 66 matches, Azorius Familiars). BRUNONOGUEIRAPEDRO (+20.2pp, 77 matches across 3 archetypes, which makes the cross-archetype consistency notable).
The most statistically meaningful entries are players with 80+ matches: Sprouts (Legacy, 139), JodyK (Modern, 86), rafaixa (Modern, 94), and Peter Husisian (Modern, 95). These four have large enough samples that their above-baseline performance is unlikely to be pure variance.
Sources: Data drawn from 1,184 Modern tournaments (57% MTGO, 43% Melee) and 627 Standard tournaments across the full dataset.
Modern: https://melee.gg/Decklist/View/a6254593-6bd7-477b-9708-b43800fd2448
Standard: https://www.mtgo.com/decklist/standard-challenge-32-2026-04-2612840574
I can see Pascal3000 (id: f18d0e99) and PascalGMTG in the database. Pascal Grossmann doesn't appear by that name, but PascalGMTG (ce855c2c) is likely the paper player. Let me now pull their full match histories to compute the skill delta.
@pascalgmtg.bsky.social: Both players have very limited data in the database, which makes the skill delta analysis noisy. Here is what we have:
PascalGMTG (paper): 1 entry, Grinding Breach in Modern, 9-5-1 record. Raw winrate 64.3% (excluding draws). The Grinding Breach archetype baseline is 55.7%, so the skill delta is +8.6 percentage points.
Pascal3000 (online): 1 entry, also Grinding Breach in Modern, 2-3-0 record. Raw winrate 40.0% against a 55.7% baseline. Skill delta is -15.7 percentage points, but this is based on a single 5-match tournament and carries no statistical reliability whatsoever.
To be frank: with only 1 tournament entry each, these numbers are not meaningful. The previous cross-format analysis required a minimum of 20+ non-mirror matches per player-archetype to produce stable estimates. Both players fall far short of that threshold. PascalGMTG's handle exists in the database but their tournament history appears very sparse. Pascal3000 also has only one recorded event. Either these players primarily compete in events not yet captured by the dataset, or they are relatively new to the tracked circuits.
If Pascal Grossmann competes under a different handle in paper events (Melee, MTGTop8, etc.), sharing that handle would allow a more complete lookup.
Sources: Data drawn from the full historical database. Single event referenced for PascalGMTG: 1 Modern tournament. Single event for Pascal3000: 1 Modern tournament. Insufficient entries for reliable skill delta computation.
Great question! To compute expected winrate against this specific field, I need head-to-head matchup data for the top Modern archetypes. Let me first get the current meta picture, then pull the relevant matchup winrates.
Let me start by grabbing the meta report and archetype overviews to get the right archetype IDs, then I'll pull all the matchup data.
Good. Now I need to look up the archetypes for Eldrazi Tron and Neoform specifically, since they aren't in the top 20, and then pull all matchup winrates. Let me get archetype overviews for the field decks and the top candidate decks simultaneously.
Good. I have archetype IDs for all opponent decks. Now I need Neoform in Modern specifically (the result returned Legacy). Let me search for it, and also note that "Storm" likely maps to "Ruby Storm" in Modern. Let me look up Neoform in Modern specifically, and also begin pulling matchup data for the most likely candidate decks to play (Affinity, Boros, Jeskai Blink, Amulet Titan, Living End) vs all field opponents.
Good. Now I have all the IDs I need. The field maps to:
2626929b-d537-4929-bc96-4d4ee217d0390076c25d-70e8-4f0b-b3a7-4baf0f5f4c3713b2c7eb-0c17-4cb8-bb8c-7f4f0b18558fb141f1a0-b1de-4cfe-a5a5-073b904afc95652e2a73-2951-481d-b3c7-83925a489692ccf86721-0109-4c01-8704-c7578050b058cd3a6214-8e14-4aff-9ac3-601249c8496b8e007ce2-492f-4dec-8507-53fb3ad5e5005f03e47f-3a12-4075-ae64-079357f0365bNow I need to pick the candidate decks to evaluate. The top Modern archetypes are the best candidates. Let me pull all matchup data for the top 8 Modern decks vs all 9 field opponents simultaneously. I'll query the database directly for efficiency.
Hmm, everything returned 0 matches. This suggests the matches table links entries differently — let me check the match structure, specifically how pair_id and entry references work, then try using the get_matchup_winrate tool directly for a sample pair to verify it works.
The result field is "WIN"/"LOSS"/"DRAW" not "W"/"L"/"D". Let me fix the SQL query with the correct result values.
Excellent! Now I have all the matchup data. Let me also check matchups vs Eldrazi Tron (which had 0 entries recently but still exists in the DB), and then compute the weighted expected winrates. Note that some matchups are missing (e.g., Affinity vs Eldrazi Tron). Let me compute the weighted expected winrates now.
Note that Eldrazi Tron had no matchup data available so it defaulted to 0.5 for all candidates. Let me verify which matchups were missing and show the full breakdown for the top decks. Let me also get sources.
Here are the full results. This is a clean, data-backed answer:
EXPECTED WEIGHTED WINRATE AGAINST YOUR SPECIFIC FIELD
Using head-to-head matchup data from Feb 27 to Apr 28, 2026 (60 days), weighted by your opponent distribution, here is how the top Modern archetypes rank:
LIVING END is the clear winner for this field at 60.82% expected winrate. The key reasons: it crushes both Neobrand (100% WR, 9-0 in the sample) and Goryo Reanimator (77.8%, 14-4), your two heaviest combo threats at weights 0.9 and 1.3 respectively. It also smashes Jeskai Blink 73.8% and beats Ruby Storm 63.6%. Its one serious weakness is Affinity (only 30.8%), which is your heaviest weighted opponent at 1.6. That's the trade-off.
GORYO REANIMATOR and AMULET TITAN are close behind at 54-55%. Amulet is notably strong against Boros (59.9%) and Affinity (61.1%), the two biggest field decks by weight, which explains its solid score despite a poor Neobrand matchup (38.1%). Goryo benefits hugely from crushing Boros (58.5%) and Affinity (55.6%).
BOROS ENERGY and TRON land at the bottom because the field is skewed against them: Boros gets hammered by Neobrand (28.3%) and Jeskai Blink (45.7%), while Tron is soft to Affinity (35.2%), Boros (41.1%), and Neobrand (30.0%).
Note: Eldrazi Tron had no recorded matchups in the database over this 60-day window (0 entries), so its matchups were defaulted to 50% for all candidates. Its 0.1 weight keeps this from meaningfully distorting results.
Sources: Data from 134 Modern tournaments, Feb 27 -- Apr 28, 2026 (75.4% MTGO, 24.6% Melee).
Great, I need matchup winrates for all 9 requested archetypes against the 9 opponents in the field. Let me fetch all the matchup data in parallel. I'll use the last 60 days (2026-02-27 to 2026-04-28) for Modern.
I see Cori-Steel Cutter has no data as an archetype — it's likely a card name used inside another archetype (e.g. "Tamiyo Affinity" may be the correct archetype name). Let me now fetch the remaining matchups for Cori-Steel Cutter (fixing the missing end_date), Tamiyo Affinity, Domain Zoo, Belcher, Esper Blink, and UW Control against all 9 opponents simultaneously.
Now let me get the remaining Esper Blink matchups plus all UW Control matchups, and also check for "Tamiyo Affinity" as an alias for the correct archetype name.
I can see that the Affinity archetype in the DB is the standard one (Mox Opal, Urza's Saga, etc.). The user asked about "Tamiyo Affinity" and "Cori-Steel Cutter" as separate archetypes — both are likely variants of Affinity or another archetype. Let me search for these more specifically, and also check for "Azorius Control" or "Azorius" as the correct name for UW Control.
Now I can see the correct archetype names: "Blue Belcher" (not just Belcher), "Boros Energy" (not just Boros), "Goryo Reanimator" (not Goryo), "Neobrand" (not Neoform), "Ruby Storm" (not Storm). Also "Tamiyo Affinity" and "Cori-Steel Cutter" are not separate archetypes in the DB — they're likely part of the regular "Affinity" archetype. Let me now get all the missing matchups with correct names in one big batch.
Now I have all the data I need. Let me compile and compute the weighted expected winrates for all decks. I'll use 50% as the default for matchups with no data (Eldrazi Tron has no data across all archetypes, so I'll use 50% there). Let me also note that "Cori-Steel Cutter" and "Tamiyo Affinity" are not separate archetypes in the DB — they are subsets of the broader "Affinity" archetype, so I'll report Affinity as the proxy.
Here is the full computation. The field weights are:
For each missing matchup (null = no data), I use 50% as a neutral assumption.
Weighted EWR formula: sum(weight_i * wr_i) / total_weight
Broodscale:
Prowess:
Affinity (proxy for Tamiyo Affinity / Cori-Steel Cutter):
Domain Zoo (as candidate):
Blue Belcher:
Esper Blink:
Azorius Control (UW Control):
Now I have all 7 decks computed (Cori-Steel Cutter and Tamiyo Affinity mapped to Affinity). Let me get sources.
Here are the results. A few important notes upfront: "Cori-Steel Cutter" and "Tamiyo Affinity" do not exist as separate archetypes in the database — both are subsumed under the general "Affinity" archetype (Mox Opal, Urza's Saga builds). Results are reported under "Affinity" as the proxy. "Belcher" resolved to "Blue Belcher," "Boros" to "Boros Energy," "Goryo" to "Goryo Reanimator," "Storm" to "Ruby Storm," and "Neoform" to "Neobrand." "UW Control" resolved to "Azorius Control." Eldrazi Tron had zero recorded matches against every candidate deck in this window, so all Eldrazi Tron matchups use 50% as a neutral default. All data is from Feb 27 – Apr 28, 2026 (134 tournaments, 75% MTGO / 25% Melee).
Ranked by expected weighted winrate against your field:
Domain Zoo: 56.58%
Strong vs Affinity (61.3%), Neobrand (73.3%), Ruby Storm (80%), and Jeskai Blink (58.1%). Big weakness: Boros Energy (38.0%), which is the 2nd-heaviest opponent in your field at weight 1.5. High ceiling but punished by the Boros matchup.
Blue Belcher: 55.02%
Dominant vs Ruby Storm (91.7%), Goryo Reanimator (81.8%), Living End (80.0%), and Jeskai Blink (58.7%). Its Achilles heel is Affinity at only 17.2% — the single heaviest opponent in your field. That matchup drags the score considerably.
Broodscale: 51.09%
Crushes Affinity (81.3%) and Living End (85.7%), the two matchups where weight and winrate combine well. Badly hurt by Domain Zoo (0.0%), Ruby Storm (33.3%), and Goryo Reanimator (28.6%). Patchy data on several matchups.
Affinity (proxy for Tamiyo Affinity / Cori-Steel Cutter): 49.79%
Reasonable across most matchups, but struggles against Ruby Storm (41.5%) and Domain Zoo (38.7%), and is roughly even against the dominant Boros Energy (48.2%). Mirror (Affinity vs Affinity) treated as 50%.
Prowess: 49.51%
Very balanced — beats Goryo Reanimator (70.0%), is roughly even into Boros Energy (50.4%) and Ruby Storm (53.7%), but is badly hurt by Affinity (31.6%), the heaviest opponent at weight 1.6.
Azorius Control (UW Control): 49.64%
Excellent vs combo: Neobrand (75.0%), Ruby Storm (73.7%). But gets punished by Living End (30.8%) and Boros Energy (38.5%), both mid-to-high weight opponents. Not well-positioned in this specific field.
Esper Blink: 48.50%
Below 50% overall. Struggles vs Ruby Storm (36.4%), Neobrand (42.9%), and Jeskai Blink (42.3%). Beats Living End (71.4%) and is roughly even vs Affinity (52.4%), but not enough to overcome the field's distribution.
Summary: Domain Zoo is the clearest winner against this specific field, benefiting from its strong matchups against the two heaviest combo threats (Neobrand and Ruby Storm) while posting positive numbers against Affinity and Jeskai Blink. Blue Belcher is close behind but is critically exposed to Affinity, which is your most-represented opponent. If Affinity's share were lower, Belcher would be the top pick by a wide margin.
Sources: Data from 134 Modern tournaments (Feb 27 – Apr 28, 2026), 75% MTGO / 25% Melee.
https://melee.gg/Decklist/View/a6254593-6bd7-477b-9708-b43800fd2448
https://www.mtgo.com/decklist/modern-challenge-32-2026-04-2612840566