Quordle Rescue Hints: How to Save a Lost Board

What Quordle Rescue Actually Is

Rescue is the mode that exists because the daily does not forgive. In a normal Quordle round you get one shared budget of guesses for four hidden words, and when that budget runs out the round is over — the two boards you never finished simply stay unfinished. Rescue takes those same boards and gives you another attempt at them. Nothing is reset. The greens you earned are still green. The yellows still point at the right letters in the wrong places. The greys are still eliminated. You are not starting a puzzle; you are inheriting a wreck and asked to finish the job.

That single sentence — inheriting a wreck — is the whole strategy. Most players approach rescue as if it were a fresh round with a head start, and then play it exactly like the daily: two wide openers, then branch the boards, then close. That is the wrong shape for the problem. A rescue round is not a search problem, because most of the searching has already been done for you, badly and under time pressure. It is an execution problem. The letters are largely known; the difficulty is that they are known unevenly, across four boards in four different states of disrepair, with a budget too small to fix all of them.

So before you type anything, the question is not "what is a good word?" It is "which of these four boards can I actually finish, and in what order?" Everything below is an answer to that question.

Why the Mode Exists at All

A daily word game has a structural problem: failure is terminal. Once you lose, there is nothing to do until tomorrow, and the thing that made the game worth playing — the tension of a live position — evaporates at exactly the moment the player is most engaged. Most games solve this with a restart button. A restart button is a bad solution for a daily puzzle, because it converts a test into a grind: you can simply replay until the board is easy, and the shared result stops meaning anything.

Rescue is the more interesting fix. Instead of erasing the failure, it keeps it and changes the task. You do not get a clean board; you get your own broken one, with all of its awkward consequences — the duplicate letter you guessed too early, the two boards that both narrowed to the same ending, the vowel you never tested. The mode turns a lost round into a different puzzle with the same vocabulary: a repair puzzle. It rewards a skill the daily barely tests, which is reading a half-finished position accurately, and it punishes a habit the daily actively encourages, which is spending guesses to generate information you should already have.

There is a design elegance to it that is worth noticing, because it explains why rescue feels different rather than just easier. In a normal round, every guess is an investment in future information. In rescue, the information is mostly sunk cost already paid. What is scarce is not knowledge but closure: the number of moves it takes to convert a nearly-known word into a green row. Players who understand that shift do well. Players who keep playing for coverage tend to run out of budget with four boards at eighty per cent.

How a Rescue Round Differs From a Normal Quordle

Five differences matter, and each one changes a decision you would otherwise make on autopilot.

1. You begin mid-game. Your first rescue guess is the most informed guess you will ever make in the Quordle family, because you choose it while looking at three or four boards' worth of constraints. In the daily, guess one is chosen in total darkness. Here, guess one is a decision with evidence behind it — and most players waste it by replaying their standard opener out of habit.
2. The budget is separate and tight. Rescue gives you a fresh allowance rather than continuing the daily's count, but it is an allowance sized for repair, not for a full search. Count it before you move. The number on the screen should determine your triage, and triage done after the first guess is triage done too late.
3. The information is asymmetric. One board may be four greens deep while another is almost untouched. In a normal round the boards are roughly equally informed at every stage, because every guess hit all four. In rescue they are not, and that asymmetry is the resource you are managing.
4. Absences are as valuable as presences. A grey letter on a damaged board eliminates a huge slice of the remaining candidate pool, because the pool was already small. In the daily, a grey is one data point among many. In rescue, it is often the deciding one.
5. You are allowed to lose a board. A rescue round can be a partial save. That changes the objective from "solve everything" to "maximise boards saved within budget", which is a fundamentally different optimisation — and the single biggest source of strategy most players never use.

If you take one thing from this section: do not play rescue like the daily with fewer guesses. Play it like triage.

Reading a Damaged Board Properly

Before you can rank the four boards you have to describe each one accurately. Vague impressions ("this one looks close") are how players talk themselves into spending three guesses on a board that was never going to close. Sort every board into one of four states.

• Closed. Every letter is known and placed. This board costs one guess, and that guess should be spent last, not first, because it teaches you nothing.
• Anchored. At least one green, and the remaining unknown positions are few. The board has a shape. You can list its candidates on one hand. This is the board you should normally save first.
• Constrained but shapeless. Several yellows, no greens, or greens that do not pin the word down. You know which letters are in it and roughly where they are not, but you cannot yet see the word. These boards are expensive: they need one good discriminating guess before they become anchored.
• Cold. Mostly greys, maybe one yellow. You do not have a board here, you have a letter-exclusion list. Cold boards in rescue are usually the ones to sacrifice, unless the budget is unusually generous.

There is a fifth state that trips people up, and it deserves its own name: the poisoned board. A poisoned board has plenty of colour but no usable anchor — for example, three yellows that all but rule out every position the letters could occupy, leaving a handful of candidates that share a shape you have not tested. It looks solved-adjacent. It is not. The tell is that when you try to write the candidate list, you can only produce words that share one unresolved letter, and you have no guess available that tests that letter without being a coin flip on the answer itself.

Write the four states down. Literally, on paper or in a note. The act of labelling forces you to stop confusing "colourful" with "close", which is the most common misread in the mode.

Triage: Which Board Do You Save First?

The naive rule is "save the closest board first". It is not wrong, but it is incomplete, and the missing half is what separates a good rescue from a lucky one. The right unit is cost to close: how many guesses will this board consume before it turns fully green, including any discriminating guess it needs first?

Estimate that number for each board, then compare it to the budget. Three worked shapes:

• Four greens plus one unknown consonant, with two candidates. Cost to close: one guess, maybe two. This is the cheapest board on the table. Close it when you have a spare guess, not now — but never sacrifice it.
• Two greens and a yellow, with a candidate list you can count on one hand. Cost to close: one discriminating guess, then one answer. Two guesses for a guaranteed save is usually the best rate available in a rescue round, and it is where your first move belongs.
• One yellow, no greens, and no plausible shape. Cost to close: unknown, and plausibly more than half your remaining budget. This is the sacrifice candidate.

Now the part most players skip: the order is not just cheapest-first, because closing a board has a second effect. Every guess you spend on a board that is not yet closed is a guess that could have been spent on a board that is. That means the value of closing a board early is not the board itself — it is the guess it frees. A board closed on move two donates every subsequent guess to the remaining boards. A board closed on the final move donates nothing.

So the ranking rule is: close the board that returns the most future guesses, subject to not losing a board you could have saved. In practice, that means anchored boards with two-guess cost go first, closed boards go last, and shapeless boards get exactly one probe each — no more — until something turns them into anchored boards.

There is one exception worth knowing. If two boards share a pattern — both end in the same three letters, or both are missing the same consonant — then neither should be closed first. One burner word can split both, and that burner is worth more than either close. Look for the shared pattern before you commit to an order; it is the highest-leverage read in the mode and it is invisible if you triage boards one at a time.

Your Opener When the Board Is Already Dirty

The standard advice for Quordle — two zero-overlap openers, ten letters covered, then branch — assumes a clean board. In rescue, replaying that pair is usually a mistake, and it is worth understanding exactly why.

A normal opener is chosen to maximise expected information over the space of all possible hidden words. That is the right objective when you know nothing. In rescue you know a great deal, and the space of possible words has already collapsed on each board. Maximising information over the original space is now wrong, because most of that space is already excluded. What you want is the guess that maximises information over the remaining space.

That difference produces a concrete procedure. Build a letter census:

1. Dead letters. Which letters are grey on every board? They are worth nothing. A guess containing them is wasting slots, no matter how good the word looks.
2. Untested letters. Which letters have never appeared in any guess? These are your candidates for the probe. In a typical damaged board there are fewer than you think — often eight to twelve.
3. Live letters. Which letters are confirmed present on at least one board? These do not need testing; they need placing.

Your rescue opener should be a word that covers as many untested letters as possible, ideally four or five, while containing no dead letters and — this is the subtle part — while not being a plausible answer on any board that is one guess from closing. If your probe is itself a candidate on the anchored board, you have turned a diagnostic move into a gamble, and you lose the information you were buying.

The practical version: pick a probe whose letters are common enough to be plausible across several boards but whose shape does not match any near-solved board. A word with an unusual ending is often ideal, because it tests four or five live letters while being obviously wrong for every board you are close on. That is the opposite of the daily's advice, where you want the most plausible word. Here you want the most informative implausible one.

One more refinement. If your budget is very short — three or four guesses — do not probe at all. A probe is an investment, and investments need time to pay off. With three guesses left and two boards to save, every guess must be aimed at a close, and the correct opener is whichever answer you believe most. Rescue rewards recognising when the information phase is over.

The Known-Letters Advantage and How It Changes Guess Economy

Here is the mental model that makes rescue click. In a normal round, a guess buys two things: letters and placement. In rescue, placement is often already bought, so the same guess buys mostly letters, and letters are cheaper than they used to be because the excluded set is large. Information per guess is therefore higher in rescue than in the daily — which sounds like good news, and is, but it creates a specific trap.

The trap is over-investing in information on a board that only needs closure. Suppose a board has four greens and one unknown consonant, and you have a guess that tests three consonants at once. It looks efficient. But the anchored board only needs one guess to close, and your multi-consonant probe closes nothing — it tells you which of three letters it is, and then you still need a second guess to enter the word. You have spent two guesses where one would do, on the board that was supposed to be donating guesses to the others.

So the economy flips. In the daily, the currency is coverage: how many letters have I tested? In rescue, the currency is solves per guess, and coverage is only valuable insofar as it converts into a solve. A guess that resolves three letters on a shapeless board and closes nothing is worth less than a guess that closes an anchored board outright, even though it "learned more".

Three rules follow:

• Never buy information you cannot spend. If you have two guesses left and two boards, information is worthless. Close.
• Buy information only where it unlocks a close. A discriminating guess is justified when it reduces a board's candidates from many to one and you have a guess left to enter the answer.
• Prefer the guess that splits two boards to the guess that splits one. Because you are choosing a word rather than a letter, and a word can carry five probes, a single well-chosen burner can convert two shapeless boards into anchored ones. That is the best rate in the game.

One consequence surprises people: duplicate letters matter more in rescue than in the daily. If a damaged board already contains a yellow that implies one copy of a letter, and the plausible candidates all contain two, you have a contradiction — and contradictions are gold, because they collapse the candidate list to a handful. Check duplicates before you check anything else. A miscounted double is the most common reason a rescue board stops making sense.

When to Abandon a Board

Abandoning feels like giving up, which is why players do it too late and too rarely. But a rescue round is scored on boards saved, and a board you cannot save is a resource sink that costs you the boards you could. The skill is knowing when the maths has turned.

A board is a candidate for abandonment when all four of these are true:

1. Its cost to close exceeds its share of the budget. With three guesses left and one board needing three, there is nothing left for the others.
2. It has at least two independent unknowns. One unknown can be guessed. Two independent unknowns mean the candidate list is multiplicative, and a single guess cannot resolve both.
3. It has no anchor to build on. No green, or a green in a position that does not constrain the rest of the word.
4. Saving it would force a coin flip elsewhere. If the guesses required here leave another board at one-guess-from-done with two candidates, you are trading a probable save for an improbable one.

When you abandon, abandon cleanly. Do not half-feed the board with a guess that might help "just in case" — that is the worst of both worlds, because it neither saves the board nor advances the ones you are actually saving. Write it off, and spend everything on the boards with a path.

There is a mirror-image error: abandoning a board that is one discriminating guess from closing because it looks shapeless. Before you write anything off, do the candidate count. If you can list five candidates and one guess splits them, the board is not shapeless — it is cheap, and you were about to throw it away.

Five Mistakes That Kill Rescue Rounds

1. Replaying the daily opener pair. It burns two of a short budget re-testing letters you already know, and it delays triage until the round is half gone.
2. Ignoring greys. On a damaged board the greys are the strongest information you own. Players skip past them because they are not colourful, then guess a word containing two dead letters.
3. Closing the closest board first. It feels productive and donates nothing. Close it when it is free, not when it is tempting.
4. Guessing an answer when you need a letter. If a board is blocked by an untested vowel, no amount of plausible consonants will open it.
5. Never abandoning. Four boards at eighty per cent is a zero. Two boards saved is a two. The mode rewards the player willing to make that trade.

The Rescue Hint Ladder

No answers on this page, by design. What follows is a ladder you run yourself, one rung at a time, stopping as soon as the next move is obvious.

1. Rung 1 — the budget. How many guesses do you actually have? Write the number down. Everything below is arithmetic against it.
2. Rung 2 — the greens count. For each board, how many letters are confirmed and placed? Four boards, four numbers. This is your first sort.
3. Rung 3 — the dead-letter list. Which letters are grey on every board? Cross them out of your mental alphabet. You will be surprised how small the live alphabet has become.
4. Rung 4 — the untested letters. Of the letters still live, which have never been played? One untested vowel can be holding up two boards at once.
5. Rung 5 — candidate counts. For each unsolved board, how many possible words remain? A number, not a feeling. Boards you cannot count are boards you cannot yet help.
6. Rung 6 — cost to close. Turn each candidate count into a guess cost. Which board is cheapest per save? That is your next move, unless a burner can split two boards at once.
7. Rung 7 — the duplicate check. Does any board have a yellow that contradicts a double letter in its likely candidates? Resolve the contradiction before anything else; it usually collapses the list.
8. Rung 8 — the sacrifice test. Is any board's cost greater than its share of the remaining budget? If yes, write it off deliberately rather than by drift.

If all eight rungs are clear and you still cannot move, the problem is almost always a misread rather than a missing word. Go back to rung 3 and re-check which letters are actually dead — a single letter you believe is eliminated but is not will make every candidate list wrong at once.

Using These Hints Without Spoiling the Round

Rescue is the one mode where a spoiler is especially costly, because the whole pleasure is the repair. If someone hands you the four words, you have not been rescued; you have been skipped past the only interesting part. So this page deliberately contains no answer, no letter-by-letter reveal, and no daily solution of any kind.

Use it as a procedure instead. Read the triage section before you open the game, then run the ladder live. When you stall, climb one rung rather than reaching for a solver. When a round ends, review it against the ladder: which rung did you skip, and what did skipping it cost? That review is worth more than ten more rounds, because rescue failures are almost always structural — a wrong order, a late sacrifice, an information guess with no close behind it — rather than lexical.

If you want to practise the reading skill specifically, play a normal Quordle round and deliberately stop at six guesses, then look at the position and triage it as though it were a rescue. You will get the same four board states, the same asymmetric information, and none of the pressure. Do that ten times and rescue rounds stop feeling like a different game, because you will have already solved the only genuinely new problem the mode presents.

For the clean-round version of this method, see our Quordle hints and strategy guide. If you want to compare how the same triage logic scales to eight boards, the Octordle hints page is the natural next step, and you can build the reading habit under no pressure in unlimited Quordle practice.