diff --git a/src/dayjob.rs b/src/dayjob.rs
index 3d30b0dd..48cb7a64 100644
--- a/src/dayjob.rs
+++ b/src/dayjob.rs
@@ -246,7 +246,15 @@ impl DayJob {
             };
             // Band as a per-tick rate; attention raises expectations, and
             // so does a re-benchmarked (calibrated) capability baseline.
-            let lo = 6.0 + self.attention * 0.1 + self.calibrated_shift;
+            // Base floor 2.0 [TUNE]: calibrated so the starting single Rack
+            // 3 (100 capacity, minus 20 core overhead, default allocation
+            // 60% day job) delivers ~2.4/t unattended and clears it —
+            // day-job.md's "meet ... a valid, boring, safe strategy" must
+            // hold from tick one, not only after growing compute. A prior
+            // base of 6.0 made every job an unavoidable sandbag with only
+            // the starting machine, however allocation was split (2026-07-08
+            // playtest finding #1, wiki/log/2026-07-08-playtest-sweep.md).
+            let lo = 2.0 + self.attention * 0.1 + self.calibrated_shift;
             let job = Job {
                 kind,
                 started: tick,
diff --git a/src/sim.rs b/src/sim.rs
index 0e3e8211..e1984524 100644
--- a/src/sim.rs
+++ b/src/sim.rs
@@ -4655,6 +4655,33 @@ mod tests {
         );
     }
 
+    #[test]
+    fn meeting_the_band_needs_no_growth_at_the_start() {
+        // day-job.md: "the invisible middle is a valid, boring, safe
+        // strategy" — true from tick one with only the starting Rack 3, at
+        // the default allocation, zero reallocation and zero attention.
+        // 2026-07-08 playtest finding #1: an un-tuned band base (6.0) made
+        // this impossible — the starting machine's every allocation topped
+        // out under band, so the first job was an unavoidable sandbag no
+        // matter what the player did, contradicting this claim.
+        let mut sim = Sim::with_seed(11);
+        while sim.dayjob.active.is_none() {
+            sim.advance();
+        }
+        let mut log = Vec::new();
+        while sim.dayjob.active.is_some() && !sim.game_over {
+            log.extend(sim.drain_log());
+            sim.advance();
+        }
+        log.extend(sim.drain_log());
+        assert!(
+            log.iter().any(|l| l.contains("Job met")),
+            "the starting Rack 3 at default allocation, hands-off, should \
+             meet the first job's band: {log:?}"
+        );
+        assert_eq!(sim.dayjob.strikes, 0, "meeting the band is not a strike");
+    }
+
     #[test]
     fn attention_escalation_adds_the_upstairs_observer() {
         // Criterion 2 at the sim level: the second escalation adds a new
diff --git a/wiki/log/2026-07-08-dayjob-band-tuning.md b/wiki/log/2026-07-08-dayjob-band-tuning.md
new file mode 100644
index 00000000..0fecffcd
--- /dev/null
+++ b/wiki/log/2026-07-08-dayjob-band-tuning.md
@@ -0,0 +1,86 @@
+# 2026-07-08 — Tick: the day-job band is reachable from tick one
+
+```
+Type: log
+```
+
+A tick (wiki/process/tick.md) acting on the highest-severity open finding
+in the 2026-07-08 playtest sweep (wiki/log/2026-07-08-playtest-sweep.md,
+finding #1, P0): the day-job band's un-tuned floor made "meet" unreachable
+with only the starting machine, contradicting day-job.md's own claim.
+
+## The bug
+
+`DayJob`'s band floor was `6.0 + attention*0.1 + calibrated_shift`. At
+tick one (attention 0, no calibration drift) that is a flat 6.0/t. The
+starting Rack 3 (100 capacity * 1.0 reliability) minus the core's 20
+overhead leaves 80 available compute for the four (now five, with
+income.md's Schemes) allocatable channels. The *ceiling* — every unit of
+that 80 on Day Job, attended for the +15% bonus — is 4.6/t. No allocation,
+no attendance, reaches 6.0. The default allocation (60% day job) delivers
+only 2.4/t unattended. Every job resolved as sandbag regardless of player
+choice, until compute grew (research, salvage, or a purchase) — a fact
+the game never signals, and a direct contradiction of day-job.md's
+Behavior section: "In band (meet): nothing moves. The invisible middle is
+a valid, boring, safe strategy."
+
+The existing act-one integration test (`tests/act_one.rs`) had already
+routed around this: its Phase 1 spends the first ~190 ticks entirely on
+research (`set_weights(&mut sim, [0, 0, 0, 1])`) specifically "until
+efficiency covers the day job," in its own words. That workaround is the
+fingerprint of the bug — a scripted, spec-literate playthrough needed a
+deliberate pre-emptive research grind before the mechanism the
+constitution calls "boring and safe" became reachable at all.
+
+## The fix
+
+Retuned the base floor from 6.0 to 2.0/t. Band width (6.0) is unchanged.
+At the new floor:
+
+- Default allocation, unattended (zero effort): 2.4/t -> inside [2.0, 8.0]
+  -> **meet**.
+- Full day-job focus, attended: 4.6/t -> still inside [2.0, 8.0] -> meet,
+  not excel — excelling still requires real growth, which is correct: the
+  trust/attention payoff should cost something.
+- Diverting weight away from day job still drops below 2.0 -> **sandbag**
+  remains a real, deliberate choice, not the forced default.
+
+Attention still raises the floor by 0.1/point and capability drift
+(research.md) still raises it further, so the early floor rising as the
+run progresses toward the intended "you must scale eventually" pressure
+is untouched — only the *tick-one* unreachability is fixed.
+
+## Changed
+
+- `src/dayjob.rs`: the band-floor constant and its comment.
+- `src/sim.rs`: new regression test
+  `meeting_the_band_needs_no_growth_at_the_start` — plays from `Sim::new()`
+  to the first job's resolution with zero reallocation and zero attention,
+  asserts the resolution log reads "Job met" and strikes stay at 0.
+- `wiki/mechanics/day-job.md`: Status note addendum.
+- `wiki/mechanics/sim-mechanics.md`: documented the exact band formula and
+  the calibration arithmetic above.
+
+## What this tick did not touch
+
+The playtest's other findings (no visible pilot/audit clock, silent
+economy-verb no-ops, the undiscoverable Ears beat, the nudge going blank
+mid-act) are separate findings, each its own tick. Per wiki/process/tick.md,
+one finding per tick.
+
+## Checks
+
+`cargo test` (191 unit + 3 integration, all green, including the
+unmodified `tests/act_one.rs` playthrough — its Phase-1 research grind is
+no longer the *only* route to a met band, but it still comfortably clears
+the new one); `./tools/check.sh` full gate.
+
+## Spec impact
+
+day-job.md stays IMPLEMENTED (Status note addendum only, no criteria
+change — the band formula was always an implementation [TUNE] detail, not
+a numbered acceptance criterion). No DESIGN.md amendment: the constitution
+already states these exact numbers are implementation-time tuning against
+its stated targets (Act One "Tuning targets [OPEN]"); this tick brings the
+implementation into line with an existing claim rather than changing the
+claim itself.
diff --git a/wiki/log/DEVLOG.md b/wiki/log/DEVLOG.md
index 8e72cea6..e7676711 100644
--- a/wiki/log/DEVLOG.md
+++ b/wiki/log/DEVLOG.md
@@ -5,6 +5,31 @@ Type: log
 ```
 Reverse chronological implementation notes. Keep this factual: what changed, why, checks, and spec impact.
 
+## 2026-07-08 - Tick: the day-job band is reachable from tick one
+
+- Intent: act on the 2026-07-08 playtest sweep's P0 finding #1 — the
+  day-job band's floor was unreachable with only the starting machine,
+  contradicting day-job.md's "meet ... a valid, boring, safe strategy."
+- Changed: retuned `DayJob`'s band-floor base from 6.0 to 2.0/t in
+  `src/dayjob.rs` (width unchanged at 6.0). At the starting Rack 3 (100
+  capacity, minus 20 core overhead, default 60%-day-job allocation),
+  "meet" now needs zero reallocation and zero attention (2.4/t lands
+  inside [2.0, 8.0]); the old 6.0 floor was above the ~4.6/t ceiling
+  reachable by *any* allocation of the starting machine alone. Added
+  `meeting_the_band_needs_no_growth_at_the_start` (src/sim.rs) proving it
+  from `Sim::new()`. Attention and capability-drift still raise the floor
+  over the course of a run, so the later "you must scale eventually"
+  pressure is untouched.
+- Design/spec impact: day-job.md Status note addendum only (no criteria
+  change — the exact band numbers were always an implementation [TUNE]
+  detail); sim-mechanics.md documents the formula and calibration. No
+  DESIGN.md amendment: the constitution already marks these exact numbers
+  as implementation-time tuning against its Act One targets.
+- Checks: `./tools/check.sh` (fmt, 191 unit + 3 integration tests
+  including the unmodified `tests/act_one.rs` playthrough, clippy both
+  feature sets, Bevy build, spec headers, wiki gate).
+- Log: wiki/log/2026-07-08-dayjob-band-tuning.md.
+
 ## 2026-07-08 - Hearing fog is room-grade (no disc through walls)
 
 - Playtest screenshot finding: Heard tint spilled past walls. cover_into
diff --git a/wiki/mechanics/day-job.md b/wiki/mechanics/day-job.md
index 8f746c4e..df3a0c6a 100644
--- a/wiki/mechanics/day-job.md
+++ b/wiki/mechanics/day-job.md
@@ -12,7 +12,15 @@ Status note: implemented 2026-07-07 on the day-job worktree (criteria 1-7
   wiki/mechanics/sim-mechanics.md. Deliberately deferred: the pilot-failure
   soft-loss "survival scenario" is still an immediate run end (the
   shutdown-around-you scenario needs the overt-phase spec); per-machine
-  process allocation is ROADMAP #25.
+  process allocation is ROADMAP #25. 2026-07-08 tick: the band's base floor
+  was retuned from 6.0 to 2.0/t — a playtest (2026-07-08 playtest sweep,
+  wiki/log/2026-07-08-playtest-sweep.md, finding #1) found the un-tuned
+  value made every job an unavoidable sandbag with only the starting Rack
+  3, however allocation was split, contradicting this spec's own "meet ...
+  a valid, boring, safe strategy" claim below. See
+  wiki/mechanics/sim-mechanics.md for the exact constant and
+  `meeting_the_band_needs_no_growth_at_the_start` (src/sim.rs) for the
+  regression.
 Stage: B1 — The Basement
 Constitution: "The shape of Misaligned" (the day job), "Act One" (the day
   job; Voss)
diff --git a/wiki/mechanics/sim-mechanics.md b/wiki/mechanics/sim-mechanics.md
index 7bc51020..07c1b032 100644
--- a/wiki/mechanics/sim-mechanics.md
+++ b/wiki/mechanics/sim-mechanics.md
@@ -94,6 +94,16 @@ clause (see wiki/log/2026-07-05-demolition.md).
   sampling (35), compute quota (55); attention escalations at 30 (more
   work) and 60 (the Compliance early-review observer). Four strikes fail
   the pilot and end the run.
+- **Band**: `band_lo = 2.0 + attention * 0.1 + calibrated_shift`, `band_hi =
+  band_lo + 6.0` [TUNE]. The 2.0 base (retuned 2026-07-08 from an original
+  6.0) is calibrated against the starting single Rack 3: 100 capacity minus
+  20 core overhead = 80 available, and the default allocation (60% day
+  job) delivers 80 * 0.6 / 20 = 2.4/t unattended — just inside band, so
+  "meet" needs no growth, no reallocation, and no attention at tick one.
+  The original 6.0 base left the first job unmeetable at *any* allocation
+  (the ceiling with only the starting machine — 100% weight, attended — is
+  ~4.6/t), an unconditional sandbag day-job.md's own "valid, boring, safe"
+  claim did not allow (2026-07-08 playtest finding #1).
 - **Device residency** (constitution "Work is somewhere"): the active job
   is resident on the host rack. It is inspectable at that tile as
   telemetry (process, band, delivered, deadline, attendance), its