Every daily email compared against its own category baseline — apply (coaching), product (Masterclass / Tackling Texting), and content (podcast / video / posts) — plus list growth, churn, and application/intent signals from the weekly BerserkerMail export.
Data cutoff: October 8, 2026
Last two weeks shown (Sep 25 – Oct 8), so recent still-accumulating sends don't shrink the table. Baselines are the trailing 90 days, full-list sends only: apply 32.1% open / 0.12% click · product 33.0% / 0.44% · content 32.1% / 2.55%. Deltas are against the email's own category.
| Date | Subject | Type | Open | Δ Open | Click | Δ Click | Unsubs |
|---|---|---|---|---|---|---|---|
| Sep 25 | Why You Haven't Met Anyone In 2 Years | apply | 32.4% | +0.3 | 0.08% | −0.04 | 6 |
| Sep 26 | I may have done you a disservice | apply | 30.9% | −1.2 | 0.13% | +0.01 | 5 |
| Sep 27 | Why it’s fine to disengage with dating | apply | 30.5% | −1.7 | 0.05% | −0.07 | 7 |
| Sep 28 | Guys Who Get You Laid | apply | 32.0% | −0.1 | 0.08% | −0.04 | 6 |
| Sep 29 | Getting Your Cheerleader | apply | 31.7% | −0.4 | 0.06% | −0.06 | 9 |
| Sep 30 | Get more dates with this riskless approach | content | 32.3% | +0.2 | 2.37% | −0.18 | 9 |
| Oct 1 | God help the morons on Instagram | content | 31.3% | −0.8 | 1.85% | −0.70 | 4 |
| Oct 2 | “Insane” tactic that saved him years | apply | 30.6% | −1.6 | 0.08% | −0.04 | 6 |
| Oct 3 | Getting past being a Flat Earther | apply | 32.0% | −0.1 | 0.05% | −0.07 | 4 |
| Oct 4 | How to tell your wife to shut up (without getting murdered) | apply | 32.3% | +0.2 | 0.06% | −0.06 | 2 |
| Oct 5 | How to Ruin Your Life By Pretending to be a Good Person | content | 32.5% | +0.4 | 2.73% | +0.18 | 3 |
| Oct 6 | Making Western marriage work when the odds are against you | content | 30.1% | −2.0 | 1.39% | −1.16 | 6 |
| Oct 7 | What we can learn from this 29 y.o. hottie * | apply | 29.7% | — | 0.08% | — | 3 |
| Oct 8 | How to beat her list of non-negotiables in a man * | — | — | — | — | — | — |
Highest unsub rates of the last month (Sep 9 – Oct 8, full-list sends). Still a quiet board — nothing over 0.17%. Two matured sends from the Sep 29–30 pair joined mid-board; nothing from October has registered above 0.09%.
| Type | Date | Unsub Rate | Lost | |
|---|---|---|---|---|
| hey. can we talk later? | product | Sep 8 | 0.17% | 11 |
| Get more dates with this riskless approach | content | Sep 30 | 0.14% | 9 |
| Getting Your Cheerleader | apply | Sep 29 | 0.14% | 9 |
| MK Ultra’ed Into Being Undateable | product | Sep 19 | 0.13% | 8 |
| Why Your Woman Stops Wanting Sex | content | Sep 11 | 0.13% | 8 |
| The Secret to Pitching Her Infinite Children | content | Sep 10 | 0.13% | 8 |
Average open rate on full-list sends fell every month from October (39.5%) through June (32.5%) — then July ticked up to 33.1%, the first monthly increase in nine months (helped by the June 30 bot purge cleaning the denominator). August closed at 32.3% with all sends matured — a mild −0.9pt slip from July, far gentler than the ~1pt/month slide of the spring, and no single-week step-down that would signal a filter hit. September closed at 32.0% on 28 sends (−0.3pt vs August — the slide has nearly flattened). October’s bar appears once the month has enough matured data (rule: no current-month bar until month-end; past mid-month it may show excluding the final week’s sends). The page benchmarks against a trailing 90-day baseline, not all-time, for exactly this reason.
The headline metric. For each join-week cohort, we check every subscriber's status at the first snapshot at least 7 days after they joined. Current answer: a new subscriber has an ~80% chance of surviving their first week (323 of 402 measurable). This series gains one bar per week as snapshots accumulate.
Measurable subscribers in parentheses. Jun 29, Jul 27, and Sep 28 cohorts omitted (too few measurable at ~1 week yet; Sep 28 — the biggest intake since June at 23 — gets its real reading next week). Only cohorts that joined within snapshot coverage (Jun 2 onward) can be measured at ~1 week; earlier cohorts appear in the retention-to-date chart below. The Jun 1 paid-spike cohort (75%) and Sep 14 (75%) share the low-water mark; the new Sep 21 bar lands at 82%.
Cohort size in parentheses. Jun 29 and Jul 27 cohorts omitted (under 10 subscribers). Each cohort's bar updates with every new snapshot, turning this into a survival curve. Mixed week: the September cohorts kept settling (Sep 14: 75 → 70%, Sep 21: 100 → 82%) while Aug 24 recovered 65 → 71% (re-subscribes are real) and the veteran flush hit pre-coverage cohorts instead. The fresh Sep 28 cohort — 23 joined, the biggest intake since June — opens at 96%.
% of each join-week cohort carrying the "Downloaded Self-Audit" tag as of Oct 6. The tag went live ~Jul 19, and the step-change is visible: cohorts joining before it ran 0–13%, while the Jul 27 and Aug 3 cohorts download at 50–67% — the self-audit is now in the new-subscriber path, and when it's actually in front of new opt-ins, roughly half take it. (Jul 27 is a 6-person cohort; the Aug 3 read — 10 of 20 — is the solid one.)
| Window | Real New / wk | Unsubs / wk | Net / wk |
|---|---|---|---|
| Jun 9–23 (14d, averaged) | +34.5 | −20 | +14.5 |
| Jun 23–30 | +26 | −19 | +7 |
| Jun 30–Jul 7 | +3 | −23 | −20 |
| Jul 7–14 | +23 | −21 | +2 |
| Jul 14–28 (14d, averaged) | +16.5 | −18.5 | −2 |
| Jul 28–Aug 4 | +4 | −31 | −27 |
| Aug 4–11 | +20 | −21 | −1 |
| Aug 11–18 | +11 | −24 | −13 |
| Aug 18–25 | +26 | −18 | +8 |
| Aug 25–Sep 1 | +16 | −29 | −13 |
| Sep 1–8 | +11 | −23 | −12 |
| Sep 8–15 | +14 | −25 | −11 |
| Sep 15–22 | +19 | −15 | +4 |
| Sep 22–29 | +16 | −17 | −1 |
| Sep 29–Oct 6 | +22 | −22 | 0 |
Every full-list send is classified on two axes. Tone: light (warm, funny, celebratory), medium (analytical or story-driven with stakes), heavy (dread — divorce, betrayal, loneliness). Selling strategy: proof (testimonials, track record, community, credentials), insight (selling by demonstrating the thinking), tension (analyze the problem deeply, build tension, pitch into it). Splits below: last week vs. the trailing 4 weeks.
| Tone | Last week (Oct 2 – Oct 8) | Last 4 weeks (Sep 11 – Oct 8) |
|---|---|---|
| Light | 1 of 7 (14%) | 6 of 26 (23%) |
| Medium | 6 of 7 (86%) | 19 of 26 (73%) |
| Heavy | 0 of 7 (0%) | 1 of 26 (4%) |
| Selling strategy | Last week (Oct 2 – Oct 8) | Last 4 weeks (Sep 11 – Oct 8) |
|---|---|---|
| Proof | 1 of 7 (14%) | 3 of 26 (12%) |
| Insight | 6 of 7 (86%) | 20 of 26 (77%) |
| Tension | 0 of 7 (0%) | 3 of 26 (12%) |
_intake/ finals drop is the fix. Segmented sends (under 1,000 recipients) are excluded from splits. Classifications live in the Tone/Strategy columns of sent-emails.csv.
Categories. Every email is classified by its CTA: apply → patstedman.com/apply (1-on-1 coaching), product → the Masterclass or Tackling Texting sales pages, content → a podcast, video, or post. Comparisons are always within category — a content email's 2% click rate and an apply email's 0.15% are both normal.
Baselines. Trailing 90 days, full-list sends only. All-time averages are inflated by the higher-open era of late 2025.
Excluded outliers. Sends under 1,000 recipients are excluded from all averages — these are segmented sends to buyers, event registrants, or sales-page visitors (e.g., "The Dynamic Man Masterclass is HERE" at 83% open to 249 buyers, the Forbidden Intimacy Deep Dive emails at ~380 registrants, "Your Free Call" to 24 people). They're a different audience, not a signal about copy. The most recent 2 days of sends are shown but not scored, since opens are still accumulating.
Conversion proxy. BerserkerMail doesn't report per-email revenue, and buyer counts aren't reported here (retired 2026-10-09 — too small-n to mean anything weekly). The coaching-side conversion metric is "Application Submitted" tags; product-side signal is intent clicks on the sales pages.