PokeBonk Guide is an independent reference site focused on starter choices, scoped recommendations and a searchable Pokédex. We show the conditions behind our recommendations so readers can decide whether the advice fits their own run.
PokeBonk Guide is maintained by the pokebonkguide team. Contact details are published on the Contact page. We are independent from the PokeBonk creator and the owners of the Pokémon franchise; we do not operate the game or handle game-account requests.
We separate menu checks, recorded runs and editorial recommendations. Before a comparison, we state the goal, mode, starting progress and decision policy. We rotate starter order in small blocks and record interruptions or changes to the protocol. Planned sessions are never counted as completed playtests.
The frozen protocol PB-BASE: goal first clear (defeat the final-zone legendary guardian, take the clear portal) · mode Classic · difficulty Normal · progress base Lab · tester one operator supervising an automated input agent (agent v2/v2.1) on keyboard and mouse · game version v0.8.13.
Description checked — a detail was read in the game interface or client strings at a stated version. Recorded run — the claim links to an observation under stated conditions. Seen in records — a species name appeared in the text of our HUD records; it is not a playability or acquisition check. Not checked — no reviewed fact is recorded for that field. Editorial recommendation — our interpretation of those observations. Needs recheck flags a detail that may no longer apply. Source categories and judgment categories are kept separate.
Tiers answer a specific choice within a stated setup. Each assessment includes a reason and a trade-off. Entries may share a tier, and Unranked is not a negative rating. Build routes show priorities and alternatives, not a promise that random choices will arrive in a fixed order. Tiers are editorial judgments — no formula assigns them from win rates.
A small set of runs by one operator and an automated input agent is not a population-wide win rate. Learning, progression, random choices and version changes can affect the results. Our records contain zero clears in 48 comparable runs — a count of 0 of n describes the checked records in that group; it does not promise the same outcome for any other player. Medians use only included runs with a recorded level; an unknown level is not treated as zero.
Run totals are generated from the reviewed ledger: 53 attempts, 48 comparable, 0 targets reached, 2 operator-retired, 2 technical, 1 protocol deviation.
| Phase | Runs | Disposition summary |
|---|---|---|
| Pilot (v0.8.5) | D01–D03 | D01–D02 excluded technical (key-code bug), D03 excluded protocol |
| Baseline (v0.8.13) | B01–B10, C01–C10, S01–S10 | 28 included failures; B02 and B08 excluded (retired) |
| Build exploration (v0.8.13) | R01–R15 | 15 included failures (Squirtle only) |
| Targeted check (v0.8.13) | X01–X05 | 5 included failures |
Every run has a stable id and a public summary anchor (#run-<id>), rendered from the same release data as the rest of the site. Raw recordings, original notes and screenshot filenames stay private; only the reviewed public projection is published.
Familiarisation on v0.8.5 (D01-D03); never merged into baseline groups.
Pilot run on v0.8.5. A key-code mapping bug prevented movement (technical in nature); excluded from comparison groups.
Pilot run on v0.8.5. A key-code mapping bug prevented movement (technical in nature); excluded from comparison groups.
Pilot/familiarisation run on v0.8.5 with kite-circle autopilot assist; pilot phase is never merged into baseline groups.
10 rotating blocks with one run per starter (B01-B10, C01-C10, S01-S10).
Reviewed loss under the frozen protocol; no deviation recorded.
Session retired by the operator (time limit or interruption); not counted as a completed attempt and excluded from comparison groups.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Session retired by the operator (time limit or interruption); not counted as a completed attempt and excluded from comparison groups.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Squirtle only: move-priority (R01-R05), planned fusion / actual survival (R06-R10), vitamin (R11-R15). Squirtle was selected for build exploration after showing the deepest baseline survival; this is an exploratory follow-up, not proof of overall best.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
X01-X05 follow-up checks after questions recorded during the earlier phases.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
Reviewed loss under the frozen protocol; no deviation recorded.
This log records material changes to the guide: what was corrected, which page was affected and why. Game release dates, our observation dates and article update dates are kept separate. An older observation is not silently relabelled as a test of the latest version.
The launch version is free to read and does not sell access to the game. It has no on-site advertising, paid memberships or affiliate links, and it measures site traffic with Google Analytics 4 only — no ad networks or data brokers. Game names and third-party assets remain with their respective rights holders. Report a factual correction, attribution issue or website problem through the Contact page.