Peak & taper for the half marathon on Sun 4 Oct — the sole A-race. Long-run ladder to 18km with goal-pace finishes; threshold and goal-pace work in place of the 5K sharpening that was deferred to a post-half block. The W29 long run sets the race target: A-goal 1:35 (4:30/km), B-goal 1:40 (4:44/km).
Half Marathon · 1:35:00
Sun 4 Oct · 4:30/km
weeks to go
Next checkpoint: Peak long run and goal-setting checkpoint. Hold 4:44 comfortably → race 1:35. Grind it → race 1:40. — Last 5km @ 4:44/km off 18km (13 Sept – 19 Sept) · in 2 weeks
Wed 9 Sept
25 min continuous at 4:40–4:55/km. 10 min warmup, 8 min cooldown. Slightly quicker and longer than W26 — this is the session that should start feeling like half marathon effort. Pace-led, HR-guarded: lock to the prescribed pace even if it feels easy. HR 165 is the abort signal, not the target.
Tue 8 Sept
Easy 6km. Z2 (135–147 bpm), conversational, ~5:30–6:00/km.
Sat 12 Sept
16km total: 11km easy at Z2 (135-147 bpm), then 4km at 4:44/km, then 1km easy jog. FUEL — this is race rehearsal, not optional. Carry fluid, drink from 20 min onward, take a gel or equivalent at ~45 min. The 29 Aug 14km was run dry in heat: pace bled 25 sec/km (6:03 to 6:28) at a flat HR of 141. A meaningful share of that was dehydration, not fitness. EVEN SPLITS — all four kilometres at 4:44, not a build. Do not let the last one drift under 4:40. On 29 Aug the block ran 4:52 / 4:48 / 4:31, and accelerating into it is what pushed HR from 171 to Z5. START HONEST — open the easy portion at 6:00-6:10 and do not go under 5:45 in the first 3km. On 29 Aug km 1-3 went out at 5:23-5:37 and put HR at 153 in what was meant to be Z2; the next 90 minutes paid for it. FINISH THE JOG — actually run the closing 1km easy. Stopping dead on the fast km costs recovery quality and inflates how hard the whole session felt.
Fri 11 Sept
30–45 min recovery spin. Peloton HR Zone 1–2 (power-zone endurance/recovery ride).
Sun 13 Sept
30–45 min moderate spin. Peloton HR Zone 2; push Zone 3 if feeling good.
Thu 10 Sept
30–45 min strength (core/upper bias; keep heavy hip-hinge/compressive loading light until the back is fully robust).
Work-crunch calendar clarified 2026-08-10; supersedes the FORWARD RISK paragraph of note id 19, which speculated about a Sep 2026 deadline. Correct picture: - Runner works in Audit. The budget deadline suspected behind the W7 (Apr 2026) zero week recurs EARLY 2027, not Sep 2026. W28-W30 (the 16km, 18km peak and taper into the 4 Oct half) are NOT at risk from it. - Runner is in year-end crunch NOW (Aug 2026, spanning W23-W24 onward) and reports managing it. Evidence supports "managing": W23 returned 83% (5/6, both runs, 13km) DURING the crunch, including the 8km long run that broke a three-week 5km ceiling. This is the opposite outcome to W7, where a work squeeze produced 0/6. Treat concurrent work load as a live but currently well-handled factor, not a red flag. WATCH: the long-run ladder steepens through the crunch window - W25 12km, W26 14km. If a week has to shrink, protect the long run and drop cross-training; the long run is the binding constraint on the half, per note id 16. Do not let a compressed week take the ladder rung with it. BANK FOR LATER: an early-2027 budget deadline collides with the FALLBACK sub-20 5K plan (dedicated 8-week build -> Feb/Mar 2027 attempt, per note id 13). It does NOT affect the primary early-Nov 2026 attempt off the half's aerobic peak. Another reason to prioritise the November window.
Claude — Week 24 — 8/10/2026
W7 (Mon 13 - Sun 19 Apr 2026) reconstructed retrospectively on 2026-08-10. This is the only 0/6 week in the plan's first 24 weeks. No contemporaneous note exists - there is a month-long gap in coaching notes between 4 Apr (W5) and 5 May (W10). BEST GUESS, UNCONFIRMED: runner attributes it to a work budget deadline and long hours, with workouts skipped as a result. Runner could find no direct evidence; treat as probable, not established. Supporting fingerprint: all six sessions were scheduled in advance on Sun 12 Apr at 20:37, so the week was planned properly and then abandoned. The 14 Apr threshold session was actively marked "skipped"; the remaining five were left as "planned" and never touched. This is an absence pattern, not an injury pattern - contrast W12 and W14 (back flares), which both generated notes written during the week and still returned 50% compliance. Distinct from the W9 holiday (Mon 27 Apr - Sun 3 May, Jersey), which was known in advance, is recorded in note id 6, and still returned 33% including a parkrun on 2 May. FORWARD RISK: work-crunch is the one disruption class in this block with zero mitigation attached, and unlike the foot or the back it gives advance warning. If a comparable deadline falls in Sep 2026 it lands on W28-W30 - the 16km, 18km peak and taper before the 4 Oct half. Mitigation pattern that already works is the W9 one: when disruption is known in advance, pre-cut the week to 2-3 protected sessions rather than letting it go to zero.
Sub-20 5K · deferred
Targeting a sharpening block after the half (~Nov) · last read 70% (medium), Week 16
Claude — Week 7 — 8/10/2026
8 Aug — 8.02km at 5:50/km, avg HR 145, RPE 5, hot, zero foot pain. Longest run since June and a 60% jump off the 5km ceiling held for three weeks. The 13 min in Z3 is heat drift, not effort drift (RPE 5 at 5:50). This completes the second consecutive clean pain-free week, so the gate set on 22 Jul is met: strides return in W24, threshold work in W25. Foot no longer the limiter — volume is.
Claude — Week 23 — 8/8/2026