Coding-decoding questions give you a rule applied to one word and ask you to apply the same rule to another. The "code" is just a systematic transformation — shift every letter by a fixed amount, replace every letter with its alphabetical position, reverse the word, or some combination. Your job is not to memorize codes; it is to reverse-engineer the transformation rule from the given example, then apply it cleanly.
Think of it like a lock-and-key analogy. The examiner shows you one locked-and-opened pair (e.g., PLANE → QMBOF). The lock is the rule. Your task is to identify what kind of lock it is, then open the second word with the same key.
In AFCAT's reasoning section, coding-decoding questions are almost always from one of four families:
The vast majority of AFCAT questions fall into families 1 and 2. If you can decode the rule within 20 seconds of reading the given pair, you will answer the question in under 60 seconds total. That is the target.
One more thing to internalize before going deeper: the alphabet has 26 letters. A=1, Z=26. After Z comes A again (wrap-around). This circular nature is what makes +1 shifts on Z give A, not some fictional letter. Keep that loop in your head at all times.
This is the most common AFCAT type. Every letter in the original word is shifted by a constant value k to produce the coded letter.
How to find k: Take the first letter of the original word and the first letter of the coded word. Count the alphabetical distance between them. Verify with the second letter. If the same k holds, you have your rule.
Example: TIGER → UJHFS
k = +1Now apply to any word — take each letter's position, add 1, convert back to letter.
Wrap-around rule: If Z (26) shifts by +1, you get 27, which wraps to A (1). So Z+1 = A, Y+2 = A, etc. For backward shifts: A−1 = Z.
Spotting the shift quickly: Don't count letter by letter through the alphabet. Use the table of common positions you should already know:
A=1, E=5, I=9, J=10, M=13, N=14, P=16, R=18, S=19, T=20, Z=26
With these anchors, you can calculate any shift in 2–3 seconds without counting on fingers.
Here, letters are replaced by their position number in the alphabet. A=1, B=2, C=3, ... Z=26. No shift involved — it is a direct lookup.
Example: BOOK → 2-15-15-11
To go in reverse: given a number, find the letter at that position.
Variants to watch for:
The word itself is reversed (PLANE → ENALP) or letters are swapped in pairs (1st↔2nd, 3rd↔4th, etc.). These are less common in AFCAT but appear occasionally.
Detection: If the coded word contains all the same letters as the original, just in a different order — suspect reversal or rearrangement, not a shift.
Two rules applied together. The most common combo is: reverse the word, then shift each letter. Or: shift letters, then reverse.
Detection strategy: If a single shift or reversal doesn't explain the pattern fully, try combining two operations. Apply one, check if the result is a recognizable transformation of the coded word.
When you see a new coding pair, run through this sequence:
k.These cut your computation time in half:
| Letter | Position | Letter | Position | |--------|----------|--------|----------| | A | 1 | N | 14 | | E | 5 | P | 16 | | I | 9 | R | 18 | | J | 10 | S | 19 | | M | 13 | T | 20 | | K | 11 | Z | 26 |
The pairs that add up to 27 (A+Z, B+Y, C+X...) are also useful — they tell you the "opposite" letter, which matters in reverse-alphabet coding.
When you see a coded pair, immediately check if the first coded letter is one ahead of the first original letter. If yes, test the second pair. If two consecutive pairs confirm +1, stop — apply +1 to every letter of the new word without checking further.
Worked example: HORSE → IPSTF. H(8)→I(9): +1. O(15)→P(16): +1. Confirmed in 4 seconds. Now encode the target word by adding 1 to each letter position. Standard approach (checking all 5 letters, then encoding): ~40s. This shortcut: ~15s.
Memorize only the vowel positions: A=1, E=5, I=9, O=15, U=21. Every consonant's position is then calculable by counting from the nearest vowel anchor.
Example: R is 3 after O (15+3=18). S is 4 after O (19). T is 5 after O (20). G is 2 after E (7). This beats counting from A every time.
For number-substitution questions: standard method of counting A-B-C-D... from scratch takes 8–12 seconds per letter. Anchor method: 2–3 seconds per letter. Over a 6-letter word, that saves ~40 seconds.
When options differ in only one or two letters, the examiner is testing whether you make an arithmetic slip on one specific letter. After encoding your answer, compare it to each wrong option — note exactly which letter they changed. Then recheck your computation for that specific letter only.
Example: If your answer is GMJHIU and the options include GMJIHU (letters 4 and 5 swapped), the trap is a transposition error. You already got all letters right — don't second-guess the others. This eliminates re-encoding the full word: saves 20–25 seconds.
In reverse-alphabet codes (A=26, B=25...), the position of any letter in the reverse system = 27 − (standard position).
So R (standard=18) in reverse = 27−18 = 9. T (standard=20) in reverse = 7.
You never need to count backward from Z. Standard method: count Z=1, Y=2, X=3... for every letter (up to 13 steps). This formula: one subtraction, 2 seconds. Saves 5–10 seconds per letter on reverse-coding questions.
When applying a shift near Z or A, do a quick boundary check before writing your answer: if any letter in the word is in the last k positions of the alphabet (for a +k shift) or the first k positions (for a −k shift), it will wrap around.
For +1: only Z wraps (→A). For +2: Y and Z wrap. For +3: X, Y, Z wrap.
Scan the original word first. If none of its letters are in the danger zone, skip the wrap-around mental check entirely and encode straight. This saves the "did I go past Z?" hesitation, which commonly costs 10–15 seconds of doubt.
Step 1 — Classify (10 seconds): Is the code letters-to-letters, or letters-to-numbers? If numbers, go to number-substitution pathway. If letters, proceed.
Step 2 — Find the rule (15 seconds): Check the first two letter-pairs of the given example. Compute the shift for each. If both shifts are equal → Caesar cipher with that shift value. If unequal → check for reversal, alternating shifts, or mixed code.
Step 3 — Verify (5 seconds): Apply your identified rule to one more letter-pair in the given example to confirm. Do not skip this — one wrong rule identification costs you the entire question.
Step 4 — Encode (20 seconds): Apply the rule to each letter of the target word. Write the coded form.
Step 5 — Match and eliminate (5 seconds): Compare your result to options. If exact match found, mark it. If not, check for a computation slip on the letter where options differ.
Total target time: under 60 seconds per question.
Why this question: Tests the most fundamental AFCAT pattern — +1 letter shift. If you can't identify this in 10 seconds, all other coding types will also slow you down.
Solving path: Check P→Q: that's +1. Verify L→M: +1. Rule confirmed after 2 checks. Now apply to FLIGHT: F(6)→G(7), L(12)→M(13), I(9)→J(10), G(7)→H(8), H(8)→I(9), T(20)→U(21). Result: GMJHIU. Match with option A.
Why this question: Standard number-substitution. AFCAT uses this type to test whether you've memorized the A=1 through Z=26 table. The question rewards preparation over in-the-moment reasoning.
Solving path: Confirm the rule from RADAR: R=18, A=1, D=4 — matches alphabetical positions exactly. For SONAR: S=19, O=15, N=14, A=1, R=18. Answer: 19-15-14-1-18. Option A.
Why this question: Tests +2 shift, which is the second most common shift value in AFCAT coding questions. The trap here is misidentifying the shift as +1 by looking only at M→O (which could be +2 or a vowel-skip pattern — verify with the second pair to be certain).
Solving path: M(13)→O(15): +2. O(15)→Q(17): +2. Confirmed as +2 shift. Apply to FATHER: F(6)→H(8), A(1)→C(3), T(20)→V(22), H(8)→J(10), E(5)→G(7), R(18)→T(20). Result: HCVJGT. Option A.
Why this question: A letter-to-number question that tests both encoding and retrieval speed. The options are designed to catch off-by-one errors on specific letters (R, G, S).
Solving path: PENCIL: P=16, E=5, N=14, C=3, I=9, L=12 — all match option values. So rule = standard alphabetical position. For ERASER: E=5, R=18, A=1, S=19, E=5, R=18. Answer: 5-18-1-19-5-18. Option A. Watch: option B has R=17 (wrong), option C has E=4 (wrong), option D has S=18 (wrong — that would be R).
Why this question: Confirms you can apply Caesar cipher with +3 shift. The PEACE→SHDFC explanation in the spec resolves to a +3 Caesar pattern for WORLD, making it a direct application question.
Solving path: The +3 Caesar cipher: W(23)→Z(26), O(15)→R(18), R(18)→U(21), L(12)→O(15), D(4)→G(7). Result: ZRUOG. Option A. Note: no wrap-around needed here since no letter in WORLD falls in the last 3 positions of the alphabet.
Why this question: Another +1 shift question — but AFCAT repeats this type because it's a baseline filter. Do not spend more than 30 seconds here. It should be automatic.
Solving path: T→U: +1. Verified. HORSE: H(8)→I(9), O(15)→P(16), R(18)→S(19), S(19)→T(20), E(5)→F(6). Result: IPSTF. Option A. Options B, C, D each alter one letter — classic trap design.
Why this question: Tests number-substitution and whether you can distinguish P=16 from P=15, and G=7 from G=6. Wrong options are built around off-by-one errors on exactly these letters.
Solving path: BOOK: B=2, O=15, O=15, K=11 — confirmed as standard positions. PAGE: P=16, A=1, G=7, E=5. Answer: 16-1-7-5. Option A. Option D has G=6 (wrong). Option B has O=15 shown as first digit, treating P=15 — wrong. Option C has A=2 — wrong.
Checking only one letter-pair to identify the shift. One pair is not enough — M→O could be +2, or it could be a vowel-skip pattern, or a reverse-code coincidence. Always verify with at least two pairs before committing to a rule.
Forgetting wrap-around at Z. Applying +3 to X gives 24+3=27, which must wrap to A (1), not some non-existent 27th letter. Students who skip the boundary check write incorrect encoded letters for words containing X, Y, or Z.
Confusing standard and reverse numbering. A=1 (standard) vs A=26 (reverse) look superficially similar unless you check carefully. If the given pair shows A coded as 26, it is reverse numbering — do not auto-apply A=1.
Transposing adjacent letters in the answer. When encoding quickly, it is common to swap two adjacent letters (writing the 4th letter's code in position 5 and vice versa). Always write the encoded letters in strict left-to-right order, pausing after each one rather than encoding all letters in your head and writing at the end.
Spending time decoding options instead of encoding the target. Some students decode all four options to check which one, when decoded, gives back the original word. This is 3–4 times slower than encoding the target word directly and matching. Encode forward, never decode the options.
Assuming every coding question uses the same rule as the previous one. Each question establishes its own rule in the given pair. Do not carry over the +1 shift from question 3 to question 4 without verifying the new pair first.