Eight weeks out, the preparation question changes. You’re no longer working through the syllabus-you’re managing a finite pool of post-2021 authentic IB Math AA HL sessions, all aligned to the current specification. Every paper you sit under timed conditions spends one of those sessions permanently, which means what matters isn’t how many scripts you touch, but how deliberately you deploy each one.
Prediction papers can expose unfamiliar framings and Paper 3-style investigations, but they don’t reliably match examiner intent or difficulty calibration. The allocation logic runs by fidelity and phase: use older authentic sessions for weeks 1-2 diagnostics-you’re measuring gaps, not simulating exam conditions-and protect the most recent authentic sessions for weeks 6-8 full simulations, when you’re actually stress-testing readiness. Use prediction papers mainly for Paper 3 practice in weeks 3-5 and occasional exposure to unexpected framing, not as a substitute for authentic Paper 1 and 2 simulations. If your authentic archive is smaller than you’d like, run fewer full simulations but keep conditions strict and complete the full review loop every time; spending scarce authentic papers on untimed practice is functionally the same as not using them at all. Treat full, timed IB Math AA HL practice exams as a stress-test of closed gaps-sequencing, not volume, is the lever.
Weeks 1-2 – Directed Diagnostic Phase
The score from your first timed paper is almost useless by itself. What matters is the taxonomy underneath it: which losses are procedural slips, which are concept gaps, which are method-selection failures, and which are communication breakdowns. Take one older authentic Paper 1 and one Paper 2 in timed chunks by section-not as full papers-and link every lost mark to the question families that carry most weight in AA HL: multi-step calculus, complex numbers and differential equations, vector geometry, probability distributions. The output is a ranked list of three or four priority gaps, not a total score-and that structure aligns with how many practitioner revision plans handle the diagnostic phase before moving into systematic corrections and topic work.
- Triage (5 minutes): Circle only the top two mark-loss causes (most marks lost or most repeated), ignore the rest for now.
- Fix (20-40 minutes): Reattempt those questions closed-book; if stuck, consult notes or solutions only after writing what you tried.
- Schedule (2 minutes): Plan for your next session to start with 10 minutes on those same causes.
- Cadence rule: Do not take a second full simulation or timed set until you have completed this triage → fix → schedule loop for the previous one.
- Boundary note: If you don’t have markschemes, use teacher or peer checking instead so you can still run the same loop.
Weeks 3-5 – Integration, Gap-Closure, and Paper 3 Preparation
Gap-drilling alone has a ceiling. Knowing a topic works fine until you’re sitting a paper where nothing tells you which topic is relevant-and that’s precisely where AA HL marks are won and lost. In weeks 3-5, use your priority gaps to drive short focused sets, but shift quickly into mixed-topic integration so your default becomes unlabeled sets where questions jump between algebra, calculus, functions, vectors, and probability. In a randomized classroom trial, students using interleaved, mixed-topic practice scored 61% on a later unannounced test, compared with 38% for classmates who used blocked, single-topic sets-indicating that mixed practice can substantially improve long-term retention and the ability to identify what a problem is actually asking before choosing a method.
Alternate the days deliberately: targeted drills on a single gap one day, then a timed mixed set the next that weaves that gap with two or three other areas. Keep sets short enough to mark and review immediately, then translate each error into a micro-skill and feed it back into your gap list-so the same weakness can’t quietly resurface under a different topic label.
Run a separate Paper 3 strand in parallel. Paper 3 is a 60-minute component with two extended investigation questions worth 20% of the final grade, and fluency on Papers 1 and 2 doesn’t automatically transfer to this format. Practitioner guidance recommends reading the whole investigation first, using earlier parts as stepping stones for later ones, and making reasoning and communication explicit throughout. From week three, include at least one timed Paper 3 investigation per week-often drawn from prediction papers or other investigation resources-so this protocol is second nature before the simulation phase, not a scramble during it.
Weeks 6-8 – Full Simulation and Score Translation
The full simulation phase is not where you get better at AA HL-it’s where you confirm whether the work in weeks 1-5 has closed the right gaps. Reserve authentic post-2023 AA HL papers for weeks 6-8: separate sittings for Papers 1, 2, and 3, correct calculator rules, no pauses mid-paper. Scores from these runs are meaningful only when conditions are this close to the live exam; otherwise they mostly measure environment effects. Practitioner 30-day revision plans typically expect around five to seven full timed papers in total, each followed by systematic corrections, so treat this phase as your opportunity to reach the lower end of that range with high-quality, authentic runs rather than churning through every paper available.
Use the same post-paper loop you established in diagnostics: triage the script to identify the top one or two reasons you lost marks, reattempt those questions closed-book before checking solutions, and only schedule another full simulation once that cycle is complete. Students who reach weeks 6-8 with no prior full papers should expect three or four authentic simulations here, topping up with one or two prediction-paper runs if needed to approach the five-to-seven benchmark. After each simulation, translate raw scores into an error budget by comparing them with current grade-boundary ranges-how many marks per paper you can still afford to lose while reaching your target grade-and let that budget dictate which remaining gaps to address. That kind of discipline only holds when it’s embedded in a repeatable daily cadence rather than a weekly intention.
The Eight-Week Schedule Template
Treat ‘Day 1-Day 7’ as a sequence, not literal weekdays-map them onto your actual timetable while preserving the core rhythm of timed work, review, and fixes. The point is a repeatable cadence per phase, with Paper 3 built in from week three and full simulations concentrated where they generate the clearest feedback: the end.
- Weeks 1-2 (Directed diagnostic cadence)
- Day 1: Paper 1 timed by section (single section only) → mark immediately → log errors by type and question family.
- Day 3: Paper 2 timed by section (single section only) → mark → update the same error log.
- Day 5: Mixed mini-check (30-45 minutes): 6-10 questions spanning at least three topics → purpose is method-selection stress-test, not score.
- Day 7: Off or light work (formula and technique refresh, no new timed paper).
- Weeks 3-5 (Integration + Paper 3 strand cadence)
- Day 1: Targeted drilling on Gap #1 (short sets) → finish with two mixed questions that include Gap #1.
- Day 2: Mixed-topic unlabeled set (timed) → mark → convert mistakes into one or two concrete micro-skills for the log.
- Day 3: Paper 3 investigation session (timed 60 minutes) → annotate where results are carried forward and where communication earns marks.
- Day 4: Targeted drilling on Gap #2 or #3 → finish with two mixed questions.
- Day 5: Mixed-topic unlabeled set (timed) → mark → update the error log and adjust gap rankings.
- Day 6: Paper 3 investigation session (timed 60 minutes) or, if Paper 3 is already stable, a third mixed set.
- Day 7: Off or light correction-only catch-up on earlier errors.
- Weeks 6-8 (Full simulation + score translation cadence)
- Day 1: Full Paper 1 simulation (timed, authentic conditions) → same-day marking and error log.
- Day 2: Full Paper 2 simulation → same-day marking and error log.
- Day 3: Review day: calculate your error budget from those scores and spend 60-90 minutes fixing the top two mark-loss causes.
- Day 4: Full Paper 3 simulation (timed 60 minutes) → annotate carry-forward structure and time allocation.
- Day 6: Optional third simulation (authentic or prediction) only if review quality has stayed high; otherwise use this slot for targeted fixes.
- Day 7: Off or light work.
- Switching criteria
- Move from Weeks 1-2 to Weeks 3-5 when your Top 3-4 gaps stay the same for two consecutive review sessions.
- Move from Weeks 3-5 to Weeks 6-8 when your mixed sets show fewer ‘I didn’t know what topic this was’ moments and you have completed at least four Paper 3 investigations.
The structure’s real work is cumulative. By the time you reach week seven, your error log has been running for six weeks-which means a full simulation is stress-testing a prioritized, narrowed gap list, not an open-ended question about everything you might not know. That’s the difference between a mock that gives you something specific to do next and one that just confirms you have a lot left to learn.
Making Every Remaining Session Count
Whether you used your papers to find and close the right gaps, or to fill revision time with the feeling of effort-that distinction is harder to see in the moment than it sounds. The review loop, the ranked gap list, the Paper 3 strand running from week three: each is designed to make a fixed archive of authentic material do more work per session. The syllabus has been stable since 2021; the pool of authentic post-2021 sessions available to any student sitting the May 2026 exams isn’t going to grow. What’s in your hands now is how much signal you extract from each one before exam day-and whether that signal shapes your next session or just gets logged and forgotten.