After thousands of AMC and AIME attempts logged in Amc2Aime, a small set of mistakes accounts for most of the avoidable point losses. None of them are about "not knowing the math." They're about execution — and execution is improvable in days, not months. Here are the seven worth memorizing.
1. Solving for the wrong quantity
The single most common mistake among trained competitors. The problem says "find the sum of x and y" and you compute the product. Or it asks for the area of the smaller region and you report the larger. Or it gives a constraint "n is a positive integer less than 100" and your answer is 102.
Countermeasure: after every answer, before bubbling, re-read the last sentence of the problem. Out loud if possible. The 8 seconds this costs catches 80% of these.
2. Sign and parity slips on the AIME
AIME answers are 0–999 integers. A wrong sign produces a different valid integer. A factor-of-2 slip produces a different valid integer. Both feel correct.
Countermeasure: mod-9 or mod-11 sanity check before bubbling. Compute your answer's digit sum (mod 9) two ways — once from the final number, once from a clean intermediate step. If they disagree, you have an error. 30 seconds, catches a non-trivial fraction of arithmetic slips.
3. Spending 8+ minutes on a problem you can't solve
The trap: you can feel you're close. You've already invested 5 minutes; the sunk-cost instinct says to push through. Meanwhile, problem 14 — which you could have nailed in 4 minutes — gets a hurried 90-second attempt and goes wrong.
Countermeasure: set a hard internal time budget per problem (2 minutes on AMC, 12 on AIME) and enforce it. When the budget runs out, move on. If you have time at the end, return. Banking minutes for double-checking earlier problems beats grinding on a stuck one.
4. Random guessing on AMC 10/12
The blank-credit rule (+1.5 per blank) is mathematically a safety net only if you don't blind-guess. Expected value of a 1-of-5 guess is 6 × (1/5) = 1.2 points, strictly worse than 1.5.
Countermeasure: commit to the rule before the test starts. Blank if no eliminations. Guess only with 3 or fewer candidates remaining. Disciplined competitors gain 4–6 points over those without this rule.
5. Reading the problem too fast on the easy ones
Problems 1–10 on AMC 10/12 (and 1–6 on AIME) are designed to be confidence-builders, but they're also where careless errors lurk. Students see "easy" and skim. The problem mentions one constraint that flips the answer; the skim missed it.
Countermeasure: read every problem twice on the first pass through the easy ones, even when you already see the approach. The 30 seconds spent re-reading is cheaper than the 6 points lost to a misread.
6. Not committing to one method
On a hard problem, you start by trying coordinates. Two minutes in, you abandon and try synthetic geometry. Two minutes later, back to coordinates with a different setup. By minute 8, you have three partial attempts and no answer.
Countermeasure: spend the firstminute deciding which approach to take. Write down the chosen approach as a single noun phrase ("coordinates with origin at A", "Vieta's on the polynomial"). Commit to it for at least 5 minutes before switching. Most rabbit holes happen because the switch-cost burned the time, not because the original approach was wrong.
7. Skipping the review pass
You finish problem 25 with 3 minutes left. The instinct is to spend those 3 minutes on the problems you skipped. The better instinct is to spend them re-checking arithmetic on problems 1–15.
Why: a problem you skipped will most likely stay skipped — it was skipped for a reason. A problem you got "right" in 90 seconds may have a sign error or a misread. Recovering one of those gains 6 points; attempting a hard skipped problem in 3 minutes gains 1.2 expected points.
Countermeasure: when you finish, don't return to skipped problems first. Return to problems 1–15 and verify each answer. Only after the easy zone is checked should you spend any leftover seconds attacking skipped ones.
The bigger picture
Notice what these seven aren't. None of them are "learn this new technique." All of them are about being a more disciplined executor of techniques you already have. That's why execution wins prep races: it scales fast (you can fix this in a week) while learning new content scales slowly (months).
For AMC competitors who've already invested in topical knowledge, the next 5–10 points typically come not from another month of Number Theory but from internalizing these seven. Try them on the next mock. Track which ones you violate. Most students discover one or two are responsible for half their losses.
Inside Amc2Aime, the L4 Socratic follow-ups specifically attack execution failures rather than teaching new content — they ask you questions like "you wrote x = 3, but the problem asked for 2x — what should the answer have been?" That's exactly the category of feedback that fixes mistakes 1, 5, and 7 above. Run a full mock, click the coach, ask for the L4 review, and pay attention to which of the seven you keep committing.