Every color is a constraint
Green fixes a position, yellow confirms the letter but rejects a position, and gray removes the letter unless duplicate logic says otherwise.
Guide Strategy Dashboard
CornerstoneIntroduction
The concept in practical Wordle terms.
Candidate reduction is the process of removing impossible Wordle answers after each guess. Every green, yellow, and gray tile should shrink the answer pool. The better your guess, the more useful that shrinkage becomes.
The concept is simple, but strong play requires nuance. Removing a lot of words is good; removing the right words is better. If five candidates remain and they differ only by the first letter, the best move is the one that separates those five, not the one that scores best against the full dictionary.
Why It Matters
How this idea changes real solving decisions.
Candidate reduction matters because Wordle is a constrained search problem. You are not trying to think of every English word; you are trying to keep only words that still match the evidence. Players who ignore candidate reduction often guess words that have already been ruled out indirectly.
It also helps explain why good second guesses can feel unintuitive. A guess may not look like a likely answer, but if it removes multiple answer families, it can be the strongest move available.
Core Principles
Use these rules before choosing the next guess.
Reduce families, not isolated words
A single guess can remove a whole family such as _ATCH, _OUND, or S_A_E when it tests the right separator.
Use negative information
Gray letters are powerful. An all-gray SLATE can be a strong result because it removes several high-frequency letters at once.
Recalculate after every turn
The best next guess changes when the candidate pool changes. Do not stay attached to a plan that the board has disproved.
Real Examples
Board situations that show the strategy in action.
| Scenario | Board | Lesson | Move |
|---|---|---|---|
| All-gray opener | SLATE -> ----- | S, L, A, T, and E are removed, which reshapes the entire search. | Use C, R, O, N, I, Y, P, or U coverage instead of retesting removed letters. |
| Ending trap | __ING with B, C, F, W possible | The pattern is narrow but not solved. | A move that tests multiple first letters reduces more candidates than one direct guess. |
| Duplicate possibility | E is green, no unique candidates fit | Candidate reduction points toward a repeated E rather than a new consonant. | Test an EE word that still respects known positions. |
Common Mistakes
The habits that make this concept harder to use.
Ignoring yellow positions
A yellow letter is not just present; it is absent from one specific slot. Forgetting that creates fake candidates.
Keeping words with gray letters
Unless duplicate feedback applies, gray letters should remove candidates completely.
Not separating close families
When many candidates share four letters, ordinary frequency advice becomes less important than the one changing slot.
Expert Tips
Advanced habits that improve repeated play.
Write the remaining pattern mentally
After each guess, translate the board into fixed positions, required letters, forbidden positions, and eliminated letters.
Look for family names
If you can name the trap family, such as _ATCH or OUND, you can choose a better separator.
Treat duplicate letters carefully
A gray duplicate copy does not always remove the letter if another copy scored yellow or green.
Use tools after the game
Reviewing the actual candidate count teaches you which guesses reduced the pool and which only looked plausible.
Comparison Section
Related concepts that players often mix together.
| Comparison | First idea | Second idea | Takeaway |
|---|---|---|---|
| Candidate reduction vs entropy | Candidate reduction focuses on the remaining answer count. | Entropy estimates expected information across feedback patterns. | Entropy predicts useful reduction; candidate reduction measures what happened on this board. |
| Letter frequency vs candidates | Frequency values common letters in general. | Candidate reduction values letters that split the current pool. | The best letters can change after each clue. |
| Solving vs filtering | Solving guesses try to end the game. | Filtering guesses intentionally narrow the pool. | Both are valid when used at the right time. |
Practical Applications
How to apply the concept in real games.
Turn two decisions
After the opener, use candidate reduction to decide whether you need a broad repair word or a direct pattern guess.
Pattern pages
When a pattern like CK, SH, or EA appears, reduce by family rather than treating the page as a simple word list.
Труден режим
Hard mode candidate reduction is narrower because legal guesses must include known clues, so each legal word should split as much as possible.
Related Tools
Use these tools to turn the strategy into repeatable decisions.
Wordle Candidate Reduction Guide FAQs
Short answers for common questions about this topic.
What is candidate reduction in Wordle?
Candidate reduction is the process of removing impossible answers after each clue.
Why is candidate reduction important?
It keeps your guesses focused on words that still fit the board and helps avoid wasted turns.
Is candidate reduction the same as entropy?
No. Entropy estimates expected information; candidate reduction tracks how much the actual pool shrinks.
How do gray letters reduce candidates?
Gray letters usually eliminate every candidate containing that letter, unless duplicate-letter feedback says one copy may still exist.
What is a trap family?
A trap family is a group of answers that share most letters, such as CATCH, WATCH, PATCH, and MATCH.
Should I always reduce candidates instead of guessing?
No. If only one or two answers remain, a direct solve can be best.
How does hard mode affect candidate reduction?
Hard mode restricts legal guesses, so you must reduce candidates while preserving all known clues.
Which tool shows candidate reduction?
Wordle Analyzer is designed for reviewing candidate reduction after a finished game.