Chaos/Order, the identity with one answer: is it actually viable?

Issue #200, 2026-09-12. Constructed, Duel (485). Rules version 2026-07-16. Card text verbatim from data/corpus_flat.txt; rules pasted from data/Riftbound-Core-Rules-2026-07-16.txt.

Run plays exist to check whether a line is viable in a real game. This is the first one written for a specific line rather than for the baseline, and the line is the one written for the only identity that had no legal finisher surviving a stalled board — ivern-ride-the-wind-double-conquer. The catalogue says it reaches eight points. A run play asks a different question: on which turn, against an opponent who is doing something, and does that beat what they are doing?

Needs

The scenario is a stalled board, because that is what the line is for: the opponent controls both battlefields and garrisons them, so 315.2.b.2 Holds nothing for you and your Hold curve pays zero.

The mana, so every line below is checkable

From the unopposed clock: 315.3.b channels 2 runes a turn, 161.2.a caps the deck at 12, 164.2.a taps a rune for 1 Energy and 164.2.b recycles it for 1 Power of its domain — and 164.2.b's cost is the RECYCLE, not an exhaust, so one rune pays Energy and Power in the same turn. 161.2.b sends a recycled rune back to the Rune Deck, so every Power you spend costs you a rune on the board. Going first, which 485.7 makes the slower seat.

Runes available: T1 = 2, T2 = 4, T3 = 6, T4 = 8, T5 = 10, T6 = 12.

The turns

Correction caught while writing this table, and it is the sort a run play exists to catch. OGN-210 Daring Poro carries the tag Poro, not Bird. So the three Iverns must choose Bird, Cat and Dog, and the Poro supplies the fourth. Reading a card's tag off its name is exactly the kind of slip that survives a rules check, because every paragraph cited is still correct.

1. Standard-Move all three Iverns to the first battlefield — one game action (144.3, with 144.3.a requiring the shared destination), each paying its own exhaust (144.2). 450 applies Contested.
2. Cleanup: 323.9 stages a Combat (opposing units are present), it opens, and your summed Might is 18 against their garrison (465.2.c). You win under 466.3.a, 466.5 establishes Control, 466.5.d makes it a Conquer.
3. 1 Score + 3 Ivern Gains = four points. 383.4.c.2.a pays every Ivern present, not the one that walked in.
4. Still your Main Phase: cast three Ride the Wind — E6 and 3 Chaos Power, which 12 runes cover with room (tap 6 for Energy, recycle 3 Chaos for Power, leaving 9 on the board). Each moves one Ivern to the second battlefield and readies it. The move is by effect (449, 420.2.b), so 144.4's base↔battlefield restriction does not bind it and 420.3.a's exhaust is not charged — which is the only reason three already-exhausted Iverns can make the trip.
5. Cleanup: the second battlefield Conquers the same way. 470 permits it because it is a different battlefield. 1 Score + 3 Gains = four more.

Eight points on turn 6.

Breaks to

• One removal spell aimed at the Daring Poro, cast on their own turn. Ivern's four-tag clause follows the effect verb, so 383.2.a.1 makes it part of the Effect, checked on resolution — kill the Poro and all six Ivern Gains blank, leaving the two bare Conquer Scores. Two points instead of eight, for one card. OGN-229 Vengeance (Order, E4 + 2 Power, "Kill a unit."), VEN-154 Public Execution (Body/Order, E2 + 1 Power) and SFD-158 Sandshifter (Order, E5 + 2 Power) all reach a base.
• It cannot be done at instant speed, which is the only reason the line survives: swept over every non-banned card carrying [Reaction] in its own text, the only one reaching a base is OGN-033 Shakedown, and its own text lets the defender decline by having the caster draw 2. So the answer is cast a full turn early and is fully telegraphed.
• The answer to the answer costs two Energy: a second Daring Poro (103.2.b allows three).
• A garrison of summed Might 18 or more at either battlefield stops a leg outright.

The verdict: it is a race, and going first it is roughly a tie

This is the part the catalogue cannot say. The opponent on this board is not idle — they are holding two battlefields, which is the do-nothing Hold curve, and that curve pays 2 points a turn from their own Beginning Phase (315.2.b.2 with 470). If they took both battlefields on their turn 2 or 3, they are at eight on turn 5 or 6 — for free, having spent nothing but the bodies.

So the honest reading of "is it viable":

• It is not fast. Turn 6 is the floor and it cannot be improved: three E6 bodies cannot land before T5 on 2-runes-a-turn, and 143.4 costs a turn after that. T2 is dead.
• It does not interact. Nothing in the seven cards slows the opponent down. It races a curve that is already ahead and wins only if their start was slow or their garrison thin.
• It is still the identity's best answer, which is the actual finding: Chaos/Order owns no board-independent point source (they are all Mind), and its only other finishers are Hold-gated and therefore dead on exactly this board. The line is correct and the identity is the problem.

The constructive version, and it is a design note rather than a line: at E26 the deck has four to six unspent Energy on turns 4 and 5 and every card in it is a payoff. Chaos is the domain of movement and denial; spending T2's dead four Energy and T4-T5's spare on anything that costs the opponent a Hold would move this from a tie to a win, and none of that is in the entry because the entry prices a combo rather than a game.

What writing this play found in the catalogue's own instrument

npm run adversarial -- --turns reports this line at T8. Played by hand it is T6, and the difference is a defect in the clock rather than in the line.

deployTurn() in scripts/adversarial-check.mjs is a greedy allocator: each turn it buys the first affordable card from a list sorted by total cost. Traced on this line it buys a Ride the Wind on turn 1 — and since the model does runes -= c.p when a Power cost is paid (correctly, per 161.2.b), spending Power on turn 1 drops the board to one rune and stunts rune growth for the rest of the game. It then cannot afford an Ivern until T5, and finishes deployment on T7:

T1: runes 1, bought [RideTheWind]      <- a human never does this
T2: runes 2, bought [RideTheWind]
T3: runes 3, bought [RideTheWind, DaringPoro]
T4: runes 5, bought [-]
T5: runes 7, bought [Ivern]
T6: runes 9, bought [Ivern]
T7: runes 11, bought [Ivern]           -> reported as deploy+1 = T8

A spell is not deployed; it is cast on the turn it is used. Treating spells as finishing-turn costs and permanents as deployment gives T6 here, and across the catalogue it moves 25 of the 80 finishers and flips 8 out of the "slower than doing nothing" bucket — which is §7's headline, so the 33-of-45 figure is an over-count by something like that margin.

That correction is NOT being shipped, because it is not right yet: the same change sends gutter-palace-keeper-time-warp and time-warp-hold-burst to Infinity (it demands every spell be payable in a single turn, which a Time-Warp CHAIN spreads across several) and moves grand-plaza-recruit-vanguard the wrong way. The defect is real and demonstrated; the replacement model needs a turn-by-turn allocator rather than a sorted greedy pass, and that is a bigger change than this play should make.