climbing // training log
← active plan

Mesocycle 5 — Convert: board 7A depth + rebuild the 6C+ rung

Span: 31-08-2026 → 26-09-2026 (4 weeks: 3 load + 1 deload) Sessions: 16 (Mon/Wed/Fri/Sat) Theme: cycle 4 fixed the thing it set out to fix and still produced no 7A send. Depth worked — 6 named problems, every one repeated, against cycle 3's 12 problems appearing exactly once each, and the 3-problems × 3-goes cap executed perfectly on both full load Mondays. What failed was everything around it. Success = at least one 7A redpoint, a rising high-point trend on three named problems that are still on the wall at the end, and a deload week that actually lands under 6.5.

What changed from cycle 4 (and why):

  • 7A depth moved to the Kilter. Two things killed cycle 4's gym-set projects and neither was effort. Yellow crimpy — the one genuine 7A with a real series, high-water 8 of 9 — was stripped off the wall before this cycle could start. And three of three named "7A/6C+" projects turned out to be 6C: Technical crimps (cycle 3), Un cast shadow and Power long reach (cycle 4). Board problems cannot be stripped and their grades are standardised, so a name written in W2 is still there in W3 and still means what it says. You also flashed a 7A there on 31-08.
  • The 6C+ rung is back, and it is the quiet headline. Cycle 3 built 6C+ to 20 logged rows; cycle 4's narrowing starved it to 9. In the same window you redpointed 14 × 6C and went 0-for-8 at 7A. The rung that converts one into the other is exactly the one that got cut. It now has a standing Saturday target: three distinct 6C+, attempted and logged, every week.
  • One gym project stays. The board is steep, crimpy and narrow. CLAUDE.md names movement vocabulary as the bottleneck, and slab, sloper and cross-through do not live on a 40° board.
  • The ramp is written down. Cycle 4 ran 7.80 → 8.00 → 7.50, so W3 was easier than W1 and the load weeks never ramped at all. Every week now carries a stated RPE target and every session is rated against it.
  • The deload carries numbers instead of an adjective. W3 → W4 fell exactly 0.50 in cycle 3 and exactly 0.50 in cycle 4, and both times the deload week's hardest session was the weekend one. "Take it easy" has failed twice, so deload Saturday now has an RPE target, a grade ceiling and a problem cap.
  • The finger edge steps BACK to 25mm. Cycle 4 ran every repeater at 20mm bodyweight and completed 15s, 7s, 5s, 6s against a 6 × clean-7s standard. Under CLAUDE.md that means the edge is too small, not that load is missing.
  • The pinch block is pinned at 20mm. Cycle 4 logged 30mm once and 20mm twice, so the three rows never compared and pinch is still unjudgeable after two dedicated cycles.
  • Weighted pull-ups progress. The 31-08 retest broke a two-test plateau: 36.0 → 38.5 → 40.0 → 40.0 → 42.0 kg. Working sets move +25.5 → +27.5 kg.

Glossary: open /protocols → glossary for any term (half-crimp, open-hand, repeater, PAP, contact strength, RFD, completion-anchored, etc). Keep on a phone tab during sessions.

Baseline reference (pull-up retested 31-08-2026; no finger retest this cycle)
Test Standing anchor Retest Notes
Pull-up 1RM (max_single) +40.0 kg (03-08) +42.0 kg · 31-08 [id 6] total 104.0 kg · 167.7 % BW. Working sets ≈ 86 % of total = +27.5 kg × 3; ramp +13.5 kg
Half-crimp 20mm 7s max +15 kg · 124 % (06-07) [id 6] not retested 4 flat reads; no max hangs while the finger is unresolved
Open-hand 20mm 7s max +15 kg · 124 % (06-07) [id 7] not retested level with HC since 06-07
Repeaters 25mm · bodyweight completion-anchored NOT derived from any max — see the Wednesday notes
Pinch block 20mm · bodyweight completion-anchored pinned for all three weeks
Bodyweight 62.0 kg confirmed 01-09 unchanged since 08-05; every pull load above is computed against it and is correct as written

Do not paste recompute_loads.mjs finger output into this table. It currently reports half-crimp and open-hand maxes as +0.0 kg · 100 % BW, because hangboard_tests has no test_type column and the script returns the latest row per grip — and every cycle-4 row is a bodyweight repeater, not a max. The real anchor is the 06-07 pair above. The pull-up line is computed correctly and is safe to trust.

The tracked projects — picked Mon 07-09, then fixed for the cycle
Slot Where Grade Picked Why it cannot rotate
A Kilter 7A Mon 07-09 Board problems are permanent and named in the app
B Kilter 7A Mon 07-09 Same, and a different crux family from A
C Gym set 6C+ scouted Sat 05-09 Gym sets live ~8 weeks; one scouted 05-09 outlasts the cycle
  • Nothing is named in this file yet, and that is deliberate. The gym rotated its sets before the cycle started, so there was no honest inventory to name at authoring time. Cycle 3 proved that unnamed rotating projects produce zero depth, so the naming discipline is preserved by a different route: you pick on 07-09, write the three names down once, and they get written into the W3 steps as chips afterwards.
  • Read the grade off the board, not off the feel. Three for three named projects across cycles 3 and 4 were a grade below what the plan claimed. That is the single most likely reason the 7A ladder has produced no sends.
  • If one sends: log it as redpoint, not project, then carry on with the other two. A send is not permission to rotate.
  • If one stalls three sessions with a flat high-point: it is the wrong problem for now. Say so in the note; it gets replaced in cycle 6, not mid-cycle.
Weekly RPE targets — the ramp, written down
Week Target Why
W1 (31-08) 7.5 re-anchor week; Monday already ran at 8
W2 (07-09) 8.0 first full grind
W3 (14-09) 8.5 peak — send week
W4 (21-09) 6.0 deload, and it has to land under 6.5

Cycle 4 ran 7.80 / 8.00 / 7.50 / 7.00. W3 came in below W1, so there was no progressive overload, and the deload sat 0.50 below the load week for the second cycle running. Rate every session; a blank RPE disarms the load-watch guards and this cycle is judged on these four numbers.

Movement-type tags — one tag per go, filled in at the log step

Hold crimp · pinch · sloper · pocket · undercling · gaston Move dyno · bump · compression · static reach · cross-through · lock-off · mantle · high step · drop knee · sit-start Foot heel hook · toe hook · smear Style/wall coordination · balance · slab · tension

Crimp has been the #1 crux tag in cycles 2, 3 and 4 (9, 9, 8 tags) and has not moved. Tag coverage thinned in cycle 4 as manual volume fell, so the re-rank got weaker just as it mattered. Every hard go gets a tag, but you do NOT stop to log between goes: climb the goes, then the log step opens the rows at once and you tag them there.

Priority hierarchy
  1. Depth on three problems — Mon, 3 goes each, high-point logged every go
  2. The 6C+ rung — Sat, three distinct 6C+ every week, logged whatever the outcome
  3. Conversion at 7A — two board 7As, ground across W2 and W3
  4. Finger re-anchor — Wed repeaters at 25mm, completion-anchored, NO max hangs
  5. Pinch benchmark — Wed, 20mm pinned, three comparable rows
  6. Maintain — weighted pull-ups (now progressing) + front-lever + antagonists
  7. Movement skill — Fri Perfect Repeats + foot precision
  8. Light PE — Fri only, easy dose
Hard rules (override everything below)
  • No full crimp — half-crimp + open-hand only, through 09-11-2026 (6-month rule from 09-05-2026)
  • No max hangs anywhere in this cycle — repeaters only, completion-anchored, and the finger is still unresolved
  • Open-hand goes first every Wednesday, half-crimp second and droppable (finger gate, 01-09)
  • Hangboard + loaded pinch Wed gym only — never on consecutive or rest days
  • Monday is capped at 3 problems. Not "about three"
  • Limit-volume cap: ≤ ~9–10 hard attempts on the Monday project block
  • Rate every session. This cycle is judged on four weekly RPE averages; a blank is not a neutral outcome
  • Front-lever = first cut if Monday limit quality or Wednesday first-hang quality dips
  • Stop on sharp finger / elbow / shoulder pain (not soreness, not pump)
  • Missed sends 2× in a row OR sleep <6 h × 2 nights → next session −1 level
  • Sun = full rest, no exceptions
Cut the front-lever / legs block if any of these flip
  • First hang of the day feels heavy → cut front-lever, keep fingers
  • Monday first project go is worse than last week's last go → cut front-lever
  • Elbow or shoulder niggle → cut both, keep climbing only
  • Session running past its stated total → cut the accessory block, never the project block

Sessions

End-of-cycle checkpoint (Sat 26-09-2026 evening)

Run /checkpoint with the cycle selected. Four questions, four places to look:

  1. Did a 7A go down? query.mjs sends --since 2026-08-31 --until 2026-09-26 --grade 7A. Cycles 3 and 4 produced zero. One redpoint is the bar. Count board and gym separately — they are different scales.
  2. Did the high-point trend rise on the three named problems? Needs ≥2 sessions per problem and a non-zero Δ. Check the denominators first: cycle 4's high_point_total moved on the same problem (9 → 8, and 14/12/12 inside one session), which makes reached/total unreadable. If a total moved again, the trend is not the finding — the input is.
  3. Did the 6C+ rung rebuild? Target is 9 distinct 6C+ engagement rows across the cycle against 9 in all of cycle 4 and 20 in cycle 3.
  4. Did the deload land? W4 average must come in under 6.5, and Saturday 26-09 must not be the hardest session of its week. It was in cycle 3 and again in cycle 4. If it is three for three, the deload Saturday cannot be social and cycle 6 drops it.

Also read the pinch rows: three Wednesdays at 20mm, one comparable series at last. Flat across three weeks means the block does not earn a slot in cycle 6.

Compare volume on MANUAL rows only. TopLogger backfills closed history — the cycle-3 window read 45 TL rows on 03-08 and 56 on 01-09 — so any raw total taken at a checkpoint decays. Manual was 50 (cy3) → 39 (cy4).

Logging discipline

  • Tap the project chip, never retype the name. From W3 the three names arrive as chips; retyping is what flatlined cycle 3.
  • High-point on every project go, not just the good ones. A go that ends at hold 3 is data.
  • Keep the high-point denominator fixed per problem. If you counted 9 holds in W2, it is 9 in W3. A moving total silently destroys the trend.
  • A send is redpoint, not project. The style field is what the pyramid counts.
  • Tag the crux move-type on every hard go (target >75 %).
  • Rate every session. Four weekly averages are the spine of this cycle.
  • Board goes log as board, gym sets as bouldering. Cycle 5 runs projects on both, so this matters more than in any previous cycle: getting it wrong corrupts every discipline-partitioned chart and makes the board-vs-gym 7A comparison meaningless.

Red flags — pause the cycle, message Claude

  • The middle-finger ache still present at the end of W1 — it will have outlasted a full cycle plus a deload
  • Sharp finger, pulley, elbow or shoulder pain that persists past a session
  • Two consecutive Mondays at RPE 9 with no high-point progress
  • Sleep <6 h for two nights running
  • Any joint that is still complaining on the next rest day

Rest days

Tue, Thu, Sun. Sunday is a full rest day before Monday — no exceptions. Bands and mobility only on rest days; loaded hangs stay Wednesday-gym-only.

Top rope

Not scheduled at authoring, deliberately — but Wed 02-09 was called as a top-rope day by the athlete on 01-09 and is written up as one (fingers first, routes as the session tail). The standing position for the rest of the cycle is unchanged: keep TR opportunistic. There are 3 Routes sessions on record (29-07, 08-08, 02-09) and, before tonight, zero sends carrying discipline:top_rope, so there is still no route signal to plan against. 02-09 is the session that fixes that: log every route with discipline:top_rope and a French grade, and put the flash / redpoint route grades on /baseline. That is what makes a scheduled TR slot arguable at cycle-6 authoring. If you take another TR day, say so in the note and do not silently displace a bouldering day.