Standard After the August Bans: Green Was the Target. Green Won the Week.
Wizards banned Badgermole Cub to kill green — then green nearly doubled to 16.5% of post-ban Standard and took three of the week's fourteen Challenge wins, because the real engine was never the Cub: Icetill Explorer plays its lands out of the graveyard, and nobody banned that. Our before/after across 2,310 tournament lists: Izzet Prowess deleted (19.2% → 1.0%), Mardu up elevenfold, Smaug already winning Challenges — and the one untouched engine that should be nervous about October 12.
Ten days ago, Wizards banned Badgermole Cub to kill the best deck in Standard, and the obituaries for green wrote themselves. Here's what our corpus says actually happened: in 520 tournament lists from the first post-ban window (August 11–20), the green Landfall cluster nearly doubled its share of the field, from 8.9% of 1,790 pre-ban lists to 16.5% — the largest archetype in post-ban Standard — and green Landfall decks won three of the fourteen Challenge-level events our corpus has ingested since the ban, a tally matched only by the one engine the ban never touched (more on that below). The deck that was supposed to die is the deck to beat.
Meanwhile the deck everyone had learned to hate quietly stopped existing. Izzet Prowess was the #1 label in our final pre-ban month at 19.2% of all lists. Since August 11 it's at 1.0%. Five lists. That's not a decline; that's a deletion.
So the August 10 bans did exactly what they were supposed to do — just not to the deck everyone was watching. Let's go through the week properly: what left, what filled the space, what The Hobbit actually contributed, and the one engine nobody touched that should be nervous about October 12.
What was banned, and what Wizards said
Standard lost three cards on August 10: Badgermole Cub, Stormchaser's Talent, and Gran-Gran.
The announcement's stated case: Selesnya Offense, the leading Badgermole Cub deck, was winning "over 57% of its matches" on the second-highest play rate — a profile Wizards explicitly compared to Vivi Ornitier and Agatha's Soul Cauldron at the time of the Vivi ban. (That 57% is Wizards' own match-win-rate claim; we track decklists, not match results.) The Cub was called "the clear choice for a card to ban from Selesnya Offense," and the other two came along so that Izzet Prowess and the Lessons decks couldn't simply inherit the throne. All three were format staples by any measure — in our final pre-ban month the Cub sat in 23.3% of all 75s, Stormchaser's Talent in 19.7%, Gran-Gran in 16.9%.
The timing is the sore spot, and Wizards didn't hide it: the announcement admits the data was fuzzier than usual because competitive attention had spent two months on Modern. The Standard RCQ season started August 15 — five days after the ban — and The Hobbit hit paper on August 14 (Arena and MTGO on the 11th). Players who built for the season got four days' notice. The community read, roughly unanimously: right call, wrong clock. CoolStuffInc's Travis Hall called the bans a net positive for a stale format, Wargamer's Matt Bassil called the Stormchaser's Talent hit "a year or more" overdue, and Star City Games filed all three under "no-brainer bans" that would have hurt less earlier. Hard to argue with any of it.
Next window: October 12, 2026. Hold that date; we'll need it below.
The engine the ban missed
Now the part the obituaries got wrong. Badgermole Cub was green's mana: earthbend 1 on entry, and every creature you tapped for mana produced an extra . Stack that on Llanowar Elves and you get the turns Frank Karsten eulogized in his August 13 magic.gg piece — "Standard players won't be adding eight mana on turn three anymore." He predicted "green decks will have to be rebuilt from the ground up," named four post-ban pillars (Four-Color Control, Izzet Spellementals, Dimir Excruciator, Mardu Discard), and pointedly declined to call any green deck one of them.
Here's the thing the week proved: the green deck that mattered was never really a mana deck. It's a lands deck, and its engine didn't lose a single card.
Icetill Explorer does three things for four mana: an additional land drop every turn, permission to play lands from your graveyard, and a landfall trigger that mills a card whenever a land you control enters. Read those together and the card feeds itself — every land drop mills you toward more lands, and every milled land is a future land drop. Around it, the shell turns raw land drops into a clock: Sazh's Chocobo grows a +1/+1 counter per landfall, Earthbender Ascension earthbends 2 on entry and tutors a basic straight onto the battlefield, then pays out +1/+1 counters and trample as its quest counters stack, Mightform Harmonizer doubles a creature's power on each landfall, and Sapling Nursery — nominally eight mana, but with affinity for Forests it usually costs two or three — converts every land into a 3/4 Treefolk. Esper Origins // Summon: Esper Maduin is the perfect mill hit: flashback it out of the graveyard Icetill filled and it comes back as a Saga that ramps and pumps the team.
None of that was banned. So the pilots kept the engine, cut the Cub, and moved on inside a weekend. Scorpion100 won the 64-player MTGO Challenge on August 16 with the cleanest version:
The adoption data tells you exactly how the rebuild went. Icetill Explorer and Mightform Harmonizer each jumped from 9.4% of all 75s pre-ban to 18.3% after it. Keen-Eyed Curator — one of the two-drops Karsten suggested as Cub replacements — went 10.7% to 18.5%, and it's a genuinely good fit here: a two-mana 3/3 that exiles graveyard cards for and turns into a 7/7 trample once it's exiled four card types, in a format where (as we'll see) everyone's graveyard is suddenly full of interaction targets. Bristly Bill, Spine Sower went from 7 lists to 30. Llanowar Elves barely moved, 24.0% to 21.2% — Karsten's line that the Elves "remain eminently playable" was the prediction that aged best. His caution about green as a pillar aged worst, and to be fair to him, he wrote it on August 13 with almost no post-ban results to read. The rebuild he said "won't be trivial" turned out to be: cut one card, play the raccoon.
Two footnotes on the green cluster, both honest ones. First, our classifier files this engine under three label variants (Mono Green Landfall, Mono-green Landfall, Mono Green Aggro) — we merged them for the 16.5% figure after verifying the winning lists share the identical sixteen-card core, and there's a Selesnya Landfall variant (7 more lists, including Tree42o's August 17 Challenge win) splashing white for Practiced Offense on top. Second, one week of a format where everyone showed up with the wrong sideboard is not a verdict. It is, however, a much better dataset than a prediction.
The losers: one deletion, one graceful downgrade, one ghost
Izzet Prowess didn't adapt; it evaporated. 19.2% of the field to 1.0% in ten days, and the collateral tells you why. Stormchaser's Talent was the whole curve in one card: a one-mana Class that makes a prowess Otter on entry, rebuys your best instant or sorcery at level 2, and at level 3 turns every instant and sorcery into another Otter. Without it, Slickshot Show-Off fell from 19.9% of all 75s to 1.7%, and Eddymurk Crab — which Karsten floated as a pivot target — fell from 24.8% to 8.1% instead of rising. The Elemental-tribal cousin, Izzet Spellementals, held steady (7.8% → 8.3%) and won a paper RCQ in Austria, but it was already its own deck; nobody meaningfully rebuilt Prowess around the remaining pieces. Sometimes a deck really is one card.
Jeskai Lessons took the downgrade path Karsten mapped. Gran-Gran was the cheapest engine in the deck — she draws-and-discards every time she becomes tapped, and with three or more Lessons in the graveyard she makes every noncreature spell cheaper, which is what turned Accumulate Wisdom into a one-mana draw-three. The Lessons cluster collapsed from about 14.4% of pre-ban lists across its labels to 2.1%. But the shell — Tablet of Discovery plus Jeskai Revelation — is fine, exactly as the post-ban consensus predicted: J_tekt won an August 15 Challenge with Jeskai Tablet, no Gran-Gran, no Lessons package to speak of, just Tablet mana into Thor, God of Thunder and a pile of interaction. Jeskai Tablet more than doubled its share (1.5% → 3.3%) and won twice (Bolas299 on the 13th, J_tekt on the 15th), and its Four-/Five-Color Control sibling held at 4.6% and won the August 18 Challenge in Homaar's hands — three wins for the Tablet shell family, level with green.
Selesnya Offense — the deck the ban was actually aimed at — is the ghost. Our classifier's label for it (Selesnya Nature's Rhythm) was 9.2% of the pre-ban field, dead even with Jeskai Lessons. Post-ban it doesn't clear our two-deck reporting threshold. Zero confirmed lists. The turn-three kills built on Cub mana into Leatherhead, Swamp Stalker wearing Practiced Offense are gone. But note what didn't die: Practiced Offense itself still sits in 12.9% of post-ban 75s, because putting a +1/+1 counter on your whole board and handing something double strike is a fine Magic card at fair speeds too — the white-splash Landfall lists play it, and so does the week's biggest overperformer, which brings us to the winners.
The winners: Mardu called its shot
Karsten declared Mardu Discard the bans' "clear winner," and for once a prediction cashed immediately. The Mardu cluster went from 0.6% of the pre-ban field — ten lists in a month — to 7.1%, and its pilots stacked trophies: Zonda won the August 14 Challenge, then won another one on August 15 with a six-slot retune of the same 75, while the paper RCQs in Ohio (Austin Daily) and Cardiff (Stanley Franks) went to the same deck under its Mardu Discard label.
The deck is a discard-synergy machine wearing an aggro suit. Cool but Rude draws you a card for every attack-step discard and, at level 2, burns each opponent for 2 per discard; Marauding Mako grows a counter per discarded card; Hardened Academic puts a +1/+1 counter on a creature whenever cards leave your graveyard; and the discards themselves are free real estate — Bloodghast returns on any land drop, and Moonshadow is a one-mana 7/7 with menace that enters under six -1/-1 counters and sheds them as permanent cards hit your graveyard. Discarding your hand is the plan, the payoff, and the mana curve all at once. It's fast, it's redundant, and the deck that used to punish it — Cub decks going over the top on mana — no longer exists.
The rest of Karsten's pillar list mostly showed up too. The Doomsday Excruciator deck (our corpus labels it Superior Doomsday, after the Superior Spider-Man reanimation package around the six-black-pips Demon) lost nothing in the bans and held ~6.2% across its labels, winning a Last Sun qualifier in Fukuoka. The Dimir midrange family — our classifier splits it across Dimir Aggro (5.2%), Dimir Mid (3.8%) and smaller labels, about 11% combined — took the August 15 Challenge with unagieel's list. And the Tablet control decks, as covered above, won two of the fourteen ingested events. The format Wizards said it wanted — slower, wider — is visible in the numbers: fourteen Challenge-level events since the ban produced eight different winning deck families once you merge label variants, and no family took more than three. The interaction suite shifted with it, from red/green hate toward black answers: Strategic Betrayal — it forces an opponent to exile a creature they control and their entire graveyard — nearly doubled from 18.8% to 35.8% of all 75s, right behind Flashfreeze and Disdainful Stroke.
The Hobbit: one dragon is doing all the work
The Hobbit went legal mid-window (paper August 14), and MTGGoldfish's podcast summarized its competitive debut in a title: "The Hobbit Sees No Play\* Post Bans." Our corpus agrees with the asterisk more than the headline — because one card very much plays.
Smaug the Magnificent — for a 4/3 with flying and haste that makes a Treasure on each of your upkeeps and, whenever he attacks, deals damage equal to your Treasure count to any target — went from unreleased to 4.8% of all post-ban 75s in six days. That's a faster first week than most headline mythics get, and the results are real: SpockVidaLoka won the August 14 Challenge with Mono Red Dragons running the full four, plus a second top-8 the same day, while the Boros version put up an August 15 top-4 (fazparte) and more. Two shells, same dragon. Mono Red pairs him with Magda, the Hoardmaster, whose crime-Treasures feed Smaug's attack trigger until three of them cash out for a 4/4 Scorpion Dragon. Boros goes wider and cheats on time instead: Momo, Friendly Flier knocks off the first flying creature spell you cast on each of your turns, which puts Smaug onto the battlefield — hasty, attacking — on turn three, with Sarkhan, Dragon Ascendant beholding him for a Treasure on arrival and growing off every Dragon that enters, Nova Hellkite next in the chain.
And the rest of the set? The Queen of Dale is the only other Hobbit card with a competitive pulse — 11 post-ban lists (2.1%), and her timing is on theme: every turn an opponent casts their first noncreature spell, she recruits (draw a card, then discard one; a 1/1 Soldier if you ditched a nonland), which is a fine place to stand in a format drifting back toward instants and sorceries. Gandalf, Goblins' Bane // Flameshape and Kíli the Resourceful, the spec darlings of preview season: zero lists each. Bilbo, Thief in the Night: two. The market hasn't caught up to that split — Bilbo still asks $18.95 and The Queen of Dale $20.20 on Scryfall's paper aggregate, while Smaug, the one actually cashing Challenges, sits at $28.30 for the regular printing. One of those prices is earned.
Speaking of prices: Badgermole Cub, the fifty-dollar mythic this announcement executed, has gone $53.23 → $44.94 in our Scryfall paper aggregate across the ten days since the ban (TCGplayer trackers report it lower still). Slow, grinding, no floor in sight — the standard trajectory for a Standard-only ban casualty.
What to watch
Three things, in order of how fast they'll answer themselves.
Does green survive contact with a prepared field? Week one was played by a format that sideboarded for Izzet Prowess and Cub mirrors that no longer exist. 16.5% across 520 lists is a real number, but the field will re-aim — Flashfreeze didn't fall from 47.4% to 36.3% of 75s because people stopped playing counterspells; it fell because half its targets left. The first big paper read is Magic Spotlight: The Hobbit in Brisbane, August 28–30. If the Landfall cluster still clears double digits there against decks built to beat it, it's a pillar, whatever the predictions said.
Does the Excruciator/Dimir block cap the aggro decks? Mardu's 0.6% → 7.1% rocket happened in a week when nobody respected it. Strategic Betrayal at 35.8% of 75s says the format is already reaching for the answer — a card that exiles the creature and the graveyard is aimed squarely at Bloodghast and Moonshadow mathematics.
Tablet of Discovery, October 12. The Cub decks are dead, the Talent decks are dead, the Gran-Gran engine is dead — and the Tablet of Discovery/Jeskai Revelation shell that many players expected to see on the August list not only survived untouched but won three of the week's fourteen events across its variants, as many as green. It's the last pre-ban engine still standing at full power. If control's share keeps climbing through the RCQ season, that's the card the October conversation will be about, and this time nobody will get to say the data was fuzzy.
We'll run this exact before/after again when the paper RCQ results are in. Same windows, same corpus, same rule as always: when the week contradicts the predictions — ours included — the week wins.
Data: DeckCortex scraped corpus, canonical tournament lists only. Post-ban window 2026-08-11 → 2026-08-20, n=520; pre-ban control window 2026-07-11 → 2026-08-10, n=1,790; both pulled 2026-08-20. All presence figures are decklist-appearance shares of full 75s (main + sideboard); archetype shares are shares of lists carrying the scraper's archetype label, with the green Landfall and Mardu label merges verified list-by-list as described inline; neither is field share, and we track no match win rates — Selesnya Offense's "over 57%" is Wizards' claim from the August 10 announcement. Fourteen Challenge-level events had ingested for the window at pull time; second daily MTGO Challenges arrive via a top-cut-only feed and the August 17–18 second challenges were still incomplete, so their winners are not counted. One stray post-ban list still shows Badgermole Cub (0.2%) — an artifact of the effective date, not a legal deck. Karsten quotes are from "Metagame Mentor: Standard After the Bans and Four Pillars for the RCQs" (magic.gg, 2026-08-13); podcast title from MTGGoldfish Podcast 601 (2026-08-17); community reads from CoolStuffInc (2026-08-14), Wargamer, and Star City Games (2026-08-10/11). Prices are Scryfall paper aggregates via our proxy, pulled 2026-08-10 and 2026-08-20. Card wording validated against live Oracle text on 2026-08-20. Next B&R: October 12, 2026.