You asked
How does taking a mulligan work?
Answered counted as answeredThe process for taking a mulligan is as follows: 1. The starting player declares whether they will take a mulligan [1][2]. 2. Each other player in turn order declares whether they will take a mulligan [1][2]. (In a multiplayer game using the shared team turns option, each player on the starting team declares first, then players on other teams declare in turn order [3]). 3. All players who decided to take mulligans do so at the same time [1][2]. 4. To take the mulligan, a player shuffles the cards in their hand back into their library [1][4][5]. 5. The player draws a new hand of cards [1]. (The size of this hand is equal to the starting hand size [1], though some sources state you draw one fewer card each time [4][5]). 6. The player puts a number of those cards equal to the number of times that player has taken a mulligan on the bottom of their library in any order [1]. A player may continue this process until their opening hand would be zero cards [1]. Once a player chooses not to take a mulligan, the remaining cards become that player's opening hand, and they may not take further mulligans [1]. The process is repeated until no player takes a mulligan [1].
🔧 debug — where this ruling came from
- outcome
- ANSWERED — Answered
- corpus profile
- live
- answerer pin (deploy-frozen, attribution only)
- gemma4:26b think:false 5571076f3d70
- timings
- search 419 ms · ask 18681 ms · total 19118 ms
- single-flight (PR-5)
- queued 1 ms · generation 19117 ms
- retrieval refs (raw)
- [{"page": 9, "score": 0.016393, "snippet": "(Some effects can modify a player's starting hand size.) A player who is dissatisfied with their initial hand may take a mulligan . First, the starting player declares whether they will take a mulligan. Then each other player in turn order does the same. Once each player has made a declaration, all ", "headingPath": "103. Starting the Game", "sharpsignalDocId": 172}, {"page": 10, "score": 0.016129, "snippet": "103.5d In a multiplayer game using the shared team turns option, first each player on the starting team declares whether that player will take a mulligan, then the players on each other team in turn order do the same. Teammates may consult while making their decisions. Then all mulligans are taken a", "headingPath": "103. Starting the Game", "sharpsignalDocId": 172}, {"page": 28, "score": 0.015873, "snippet": "The place where your mana is stored until you spend it or until the current step or phase ends. A series of games against the same opponent. Most matches are best two out of three, so the first player to win two games wins the match. The loser of the first game decides who goes first in the second g", "headingPath": "M · M · M · M · M · M · …", "sharpsignalDocId": 163}, {"page": 25, "score": 0.015625, "snippet": "A keyword ability seen on creatures. Creatures with double strike deal their combat damage twice. When you reach the combat damage step, check to see if any attacking or blocking creatures have first strike or double strike. If so, an extra combat damage step is created just for them. Only creatures", "headingPath": "D · D · D · D · E · E", "sharpsignalDocId": 163}, {"page": 212, "score": 0.015385, "snippet": "If the back face of the card that represents that spell is an instant or sorcery face, or that spell is a copy of a double-faced card created with an instant or sorcery back face, it doesn't enter the battlefield, and is instead put into its owner's graveyard. Example: A player controls both Mycosyn", "headingPath": "705. Flipping a Coin", "sharpsignalDocId": 172}, {"page": 9, "score": 0.015152, "snippet": "sticker sheets from among those in the sealed product they opened and reveals them. In either case, that player has access to only the stickers on the chosen sheets during the game, and those sticker sheets remain revealed. (See rule 123, 'Stickers.') 103.2e In a Conspiracy Draft game, each player p", "headingPath": "103. Starting the Game", "sharpsignalDocId": 172}, {"page": 198, "score": 0.014925, "snippet": "For example, a d20 is a twenty-sided die with possible outcomes from 1 to 20. 706.1b Players may agree to use an alternate method for rolling a die, including a digital substitute, as long as the method used has the same number of equally likely outcomes as the die specified in the instruction. 706.", "headingPath": "705. Flipping a Coin", "sharpsignalDocId": 172}, {"page": 212, "score": 0.014706, "snippet": "712.10. If a spell or ability instructs a player to transform or convert a permanent, and the face that permanent would transform or convert into is an instant or sorcery card face, or is a double-faced token that was created with an instant or sorcery face, nothing happens. 712.11. A double-faced s", "headingPath": "705. Flipping a Coin", "sharpsignalDocId": 172}]
- citations (raw)
- [{"page": 9, "quote": null, "headingPath": "103. Starting the Game", "sharpsignalDocId": 172}, {"page": 9, "quote": null, "headingPath": "103. Starting the Game", "sharpsignalDocId": 172}, {"page": 10, "quote": null, "headingPath": "103. Starting the Game", "sharpsignalDocId": 172}, {"page": 28, "quote": null, "headingPath": "M · M · M · M · M · M · …", "sharpsignalDocId": 163}, {"page": 16, "quote": null, "headingPath": "Section 3: Playing a Game", "sharpsignalDocId": 163}]
- boost provenance — errata + base rulebook (E2/E3)
-
no registered errata card matched this query; base-rulebook boost: not recorded (this ruling predates it); rulesage.errata.boost-enabled=true, rulesage.errata.supersedence-enabled=true
base-rulebook boost: not recorded (this ruling predates it)
- answered-before recall
-
this ruling was answered, so past rulings were not offered
similarity threshold in force: 0.55 (rulesage.answered-before.similarity-threshold; the exact-question tier is not gated by it)
- debug row retained until
- Oct 31, 2026 18:37
retrieval — what the search arm returned
score = the fused retrieval score (reciprocal-rank fusion over the search arms) — NOT a cosine similarity. How many arms fused this ruling was NOT RECORDED (it predates the boost trail, ran on the path that stored none, or stored one that could not be read), and the ceiling depends on that count (ceiling = arms/(k+1) at k=60): 2 arms top out at 0.03279; 3 arms top out at 0.04918. So a score above 0.03279 here is not an anomaly: it is arithmetic proof a boost arm voted, and the measured band 0.016–0.033 (taken on two-arm asks) is what does not apply. A chunk only one arm ranked highly lands near 0.016 (that one arm's #1 vote), so 0.03 here is an excellent hit, not a 3% match.
per-arm ranks (the rank each search arm gave a chunk): NOT DISCLOSED by the gateway. Fusion happens gateway-side and only the FUSED score crosses the wire, so rulesage cannot show them without a gateway change. The rank column below is the FUSED rank (position in the returned list), which is real.
token count per chunk: NOT DISCLOSED — chunk token counts are not reported on the ask path. The character counts below are of the bounded snippet/quote rulesage actually received — a real measurement of what we hold, never an estimate of the chunk.
| fused rank | document | heading | page | fused RRF score | snippet chars |
|---|---|---|---|---|---|
| #1 | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 103. Starting the Game | p.9 | 0.016393 (inside the measured healthy band 0.016–0.033 (measured on two-arm asks; this ruling's arm count was never recorded)) | 300 |
| #2 | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 103. Starting the Game | p.10 | 0.016129 (inside the measured healthy band 0.016–0.033 (measured on two-arm asks; this ruling's arm count was never recorded)) | 300 |
| #3 | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 163) | M · M · M · M · M · M · … | p.28 | 0.015873 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded) | 300 |
| #4 | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 163) | D · D · D · D · E · E | p.25 | 0.015625 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded) | 300 |
| #5 | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 705. Flipping a Coin | p.212 | 0.015385 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded) | 300 |
| #6 | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 103. Starting the Game | p.9 | 0.015152 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded) | 300 |
| #7 | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 705. Flipping a Coin | p.198 | 0.014925 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded) | 300 |
| #8 | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 705. Flipping a Coin | p.212 | 0.014706 (below the measured healthy band (< 0.016) — a weak fuse on a two-arm ask, and this ruling's arm count was never recorded) | 300 |
chosen chunks — what the answer actually cited
token count per chunk: NOT DISCLOSED — chunk token counts are not reported on the ask path. The character counts below are of the bounded snippet/quote rulesage actually received — a real measurement of what we hold, never an estimate of the chunk.
| # | document | heading | page | quote chars |
|---|---|---|---|---|
| [1] | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 103. Starting the Game | p.9 | no quote on the wire |
| [2] | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 103. Starting the Game | p.9 | no quote on the wire |
| [3] | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 172) | 103. Starting the Game | p.10 | no quote on the wire |
| [4] | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 163) | M · M · M · M · M · M · … | p.28 | no quote on the wire |
| [5] | RULEBOOK Magic: The Gathering — Rulebook (corpus doc 163) | Section 3: Playing a Game | p.16 | no quote on the wire |
generation — timing and the path taken
- timing split
- retrieval 419 ms · first delta 16941 ms · total 19118 ms (retrieval and first-delta both measured from the START of gateway work, so first-delta INCLUDES retrieval — do not subtract)
- full breakdown
- queued 1 ms · retrieval 419 ms · ask 18681 ms · gateway work 19117 ms · total 19118 ms
- tokens in (the prompt the answerer read)
- 4037
- tokens out (what the answerer generated)
- 290
- tok/sec
- 136.3 tok/sec — 290 output tokens in 2.128 s of generation
- live seat (read from the running child)
- answering — gemma4:26b think:false · num_ctx 32768 (live-verified 6s ago)
- stream path
- streamed — the gateway pushed at least one accumulated draft
- watchdog
- first-token watchdog: 30s per arming window (rulesage.table.first-token-watchdog-seconds), re-armed once by the passages frame — so the worst case before a reader falls back to polling is 60s. Whether it actually fired is a BROWSER-side decision that never reaches the server, so it is not recorded here. The stream path above is the server's half of that story: whether a delta was emitted at all.
- num_ctx · think posture
- both are the CHILD's settings, not rulesage's, and no ask reply carries them — an answer never states which context window or think posture produced it. The live-seat line above reads them from the running child instead, where the gateway can answer for itself; it describes the child NOW, not this ask.
disposition
- outcome
- ANSWERED — Answered
- abstain rationale (n/a — this ruling answered)
- answered — no abstain rationale applies
setup-shape — why the checklist band did or didn't render
- decision
- NOT_SETUP_SHAPED — no setup-shape pattern matched this question
errata registry — what the gate above had to match against
this game registers NO errata cards (edition 85), so "no registered errata card matched" above could never have matched one
Errata scope: the gate above matched this ruling's own edition only. When an ask also reads an expansion, that expansion's own registered errata cards stay outside the match — a stated phase-1 limitation, not a failed lookup.