Released in 2017, A Hat in Time is a 3D “collect-a-thon” platformer built around colorful hub worlds and self-contained acts, developed by Gears for Breakfast and published by Humble Bundle. It began on PC, then expanded to consoles over time, and today it is best known across Steam and the major console platforms as a movement-forward platformer where progression is tied to recovering Time Pieces to fuel the protagonist’s journey.
As a speedgame, it shines because the core verbs are fast, expressive, and easy to chain. A typical run is a constant tradeoff between clean movement and smart act selection: runners want the most direct path to the Time Pieces needed for progression while minimizing detours, downtime, and recovery from mistakes. The act structure also makes the game naturally “grindable,” since individual missions can be practiced in isolation, then stitched into an overall route that balances consistency with riskier time saves.
What sets its speedrunning apart from a lot of similar 3D platformers is how often routing decisions are shaped by mission design rather than pure open-ended collection. The game supports multiple competitive flavors that highlight different strengths, from quick-completion approaches to broader collection and challenge-focused formats, which helps the scene stay varied without needing the overview to hinge on any single record or era.
A Hat in Time’s speedrun scene started taking shape before the “final” game was even the version most people would come to know. During the game’s crowdfunding and prerelease period, runners were already organizing around limited-access builds and community events, sharing early routes and comparing times through public video uploads. That early energy is preserved in a separate prerelease leaderboard space on speedrun.com for A Hat in Time (Beta Build), including a dedicated “2017 Speedrunning Event” category with archived run submissions that still link back to original VODs on Twitch.
Once the full release settled in, the public record for the game’s runs consolidated around speedrun.com’s main game page, where the earliest “standard goals” were formalized as full-game categories (with Any% as the baseline completion race, and longer completion styles built out alongside it). Early submissions show a heavy early reliance on PC builds and clearly labeled versions such as the base “1.0,” which matters because route assumptions and available tricks can differ across updates. Timing norms also stabilized through the same public infrastructure: the leaderboard presentation itself tracks both the timer method used and the version context for a run, reflecting how the community tried to keep comparisons fair as the scene expanded.
As the game evolved through updates and expansions, the speedrun scene had to evolve with it. Community discussions and run listings show an ongoing awareness that patches could alter or remove key skips, sometimes forcing runners to change routes or reference older builds when documenting how a trick worked. In practice, that reality helped push version labeling into the culture of the leaderboards, with runs commonly tagged by DLC and build context rather than treated as one timeless, unchanging ruleset.
Most of the scene’s organization lives in two places that serve different needs. speedrun.com functions as the public record, where categories are defined, leaderboards are kept, rules are visible, and runs are submitted and archived alongside forums, guides, and community resources. Alongside that public archive, day to day problem solving and route talk tends to consolidate in real time spaces, especially Discord, including the official server for the game run by Gears for Breakfast, plus runner focused discussion linked through the game’s community hubs.
Moderation and verification follow the same general rhythm as most modern speedrun communities. Runners submit a time through the game page, attach the required details and proof, and then wait while volunteer moderators review whether the run meets the category’s rules. Speedrun.com’s own guidance reflects that this review process is not instant, with expectations around patience and a typical window for moderators to process submissions, which fits the broader culture of trust plus verification that keeps leaderboards credible.
Knowledge in the community is preserved less by scattered tips and more by documentation that can be reused by the next runner. The game’s speedrun hub includes a substantial set of guides and resources that cover practical setup, route learning, and version specific issues, including documentation on downpatching, community allowed mods, and training materials. Tooling also becomes part of the culture, with shared utilities that standardize practice and performance across different PC setups, making it easier for runners to compare runs fairly and to reproduce common strats.
On speedrun.com, the game’s full run structure is built around a small family of headline categories that frame what “completion” means for different runners. The main full game boards center on Any% as the shortest end-to-end goal, then expand outward into broader completion and challenge formats like All Time Pieces, All Rifts, 110%, Death Wish, and Co-op. Those sit alongside a large set of individual level boards, which let runners treat specific acts and time rifts as their own competitive spaces rather than forcing every improvement into a full game attempt.
The scene’s timing identity is tied to the game’s built-in timer logic. Community guidance emphasizes that the in-game timer is expected to be enabled, and that it pauses during loading so that load performance does not decide placements. In practice, runners often mirror that behavior by using a timer setup that compares against “Game Time,” or by using an autosplitter approach that stays in sync with the in-game clock rather than raw real time.
Several “big decisions” are reflected directly in how the boards are organized and labeled. Full game leaderboards are presented with PC as the platform context, while level boards can include additional platform groupings such as PS/Xbox, which matters because hardware and version behavior can differ. The boards also track a run’s version context and timer state in the listings, including distinctions like “Fixed” versus “Default” and version notes such as DLC or build labels, which helps the community keep comparisons meaningful across updates and setups.
Finally, when the community wants a stricter interpretation of what is allowed, it often lives as a parallel rule set rather than a vague expectation. A clear example is the separate “Category Extensions” space that includes variants like Any% NMG (a “No Major Glitches” ruleset), alongside core boards that also use subcategory filters such as “Lagless” and “Lag Abuse.” Even without turning the rules section into a legal document, those labels signal the same underlying principle: the leaderboard’s identity is defined by what techniques are permitted, and the community encodes that identity into categories and subcategories rather than leaving it to guesswork.
As the scene matured, A Hat in Time routing became less about simply “doing the acts fast” and more about building a repeatable chain of decisions that keeps momentum high. Runners learned which mission objectives can be cleared with minimal setup, where a quick reset is faster than salvaging a mistake, and how to keep movement flowing through hub transitions and act openings. That kind of optimization is reflected in how the community preserves knowledge, with consolidated Any% and All Time Pieces learning material living alongside practical setup advice rather than being treated as scattered personal notes.
The biggest strategy shifts have often come from the intersection of tricks and version behavior. Certain skips and movement glitches became route-defining when they were reliable, then stopped being central when later builds changed what was possible. Community discussion about versions highlights this clearly, including examples like “finale skip” and “bag skip” no longer being possible on one major version line, while other movement oddities such as “bucket flying” remained part of the toolbox. Because of changes like that, downpatching and version awareness became a practical skill, not a niche hobby, and it is documented as part of the runner workflow rather than treated as secret knowledge.
Technique discovery also moved from one-off clips to a more formal research culture. The forums function as a long-running lab notebook where players describe how a trick works, what conditions trigger it, and what it is actually good for in real runs. A good example is the community explanation of “Pattycake Hover,” which ties the trick to specific in-game states and then immediately frames its value in terms of category utility, like enabling an early pickup that matters for formats such as Bingo.
Over time, tooling helped push the game toward consistency and comparable timing. The community treats the in-game timer as the baseline reference, since it pauses during loading and keeps hardware differences from deciding placements, and runners align external timing by comparing against game time or using an autosplitter. Discussion around LiveSplit setup, plus a dedicated autosplitting component hosted on GitHub, shows how the scene built infrastructure that reduces human error and standardizes splits. Alongside that, the speedrun.com resources list includes dedicated tools for things like in-game timer fixes and other quality-of-life utilities, which is a hallmark of a community that expects runners to practice, test, and submit under shared technical standards.
One of the most important milestones for A Hat in Time was the point where “what counts as a run” stopped being informal and became encoded into public structure. On speedrun.com, the game’s identity as a speedgame is expressed through a stable set of full-game categories and a deep bench of individual-level boards, with built-in subcategory filters that communicate expectations at a glance, like “Lagless” versus “Lag Abuse.” Even at the level-board level, the leaderboard itself tells a story about rule boundaries by tracking route-defining allowances in a formal way, such as whether a run uses “Finale Skip” or not. That kind of visible taxonomy is a milestone because it raises verification standards and gives runners a shared language for comparing like with like.
Another defining milestone type has been the scene’s response to version change. When major tricks become impossible or substantially different across builds, it forces the community to decide whether to adapt the main route, recommend a “best beginner” version, or preserve older techniques as historical knowledge rather than active strategy. In A Hat in Time, community discussion around patched-out skips and version differences became part of the scene’s collective memory, and it helped normalize version labeling as a practical part of understanding runs rather than a footnote.
A parallel milestone has been the creation of “alternate identities” for the game through rule variants rather than one monolithic leaderboard. The existence of a dedicated Category Extensions space, including formats like Any% NMG, represents the moment where the community formalized different philosophies of play as legitimate, trackable competition instead of leaving them as unofficial challenges. That matters historically because it broadens participation without diluting standards: runners can choose a ruleset that matches their preferred style, and the community can enforce those choices cleanly.
Finally, the game’s marathon showcases function as watershed moments for visibility and culture. A Hat in Time appearing on Games Done Quick stages, including runs at Summer Games Done Quick and Awesome Games Done Quick, signaled that the game’s routes were mature enough for a broad audience and rigorous enough for a high-scrutiny setting. Showcases in adjacent GDQ events like Flame Fatales also highlighted how the game supports different “showcase flavors,” including cooperative formats, which reinforces the idea that the scene is bigger than a single category or one kind of run.
elisamiau – Any% (Lagless) – 33m 45s 720ms – Date achieved: “1 month ago” – PC, DLC 2 w/ mods, Fixed timer – A modern Any% benchmark that reflects the current DLC-era routing and execution expectations.
Fuff – Any% (Lagless) – 44m 36s 300ms – Date achieved: “7 years ago” – PC, Version 1.0, Default timer – A clear snapshot of the early, launch-version era before later DLC and modded-version norms became dominant.
Zoro64 – All Time Pieces (Base Levels, Lagless) – 1h 12m 47s 160ms – Date achieved: “3 months ago” – PC, DLC 2 w/ mods, Fixed timer – A defining All Time Pieces run in the DLC-era ecosystem, showing how full-collection routing is optimized at the top end.
elisamiau – All Time Pieces (Base Levels, Lagless) – 1h 13m 22s 880ms – Date achieved: “3 months ago” – PC, DLC 2 w/ mods, Fixed timer – A near-top ATP performance that helps mark the shape of the current record-era competition.
Doka – Death Wish Any% (Lagless) – 55m 55s 370ms – Date achieved: “1 year ago” – PC, CDLC 1, Fixed timer – A standout Death Wish Any% that represents the category’s highest consistency-and-risk balancing at the front of the board.
elisamiau – Death Wish Any% (Lagless) – 1h 00m 37s 020ms – Date achieved: “6 years ago” – PC, DLC 1.5, Default timer – A historically useful marker from an earlier Death Wish meta, showing what “fast” looked like before later refinements and leadership changes.
elisamiau – 110% (Lagless) – 3h 32m 18s 810ms – Date achieved: “3 months ago” – PC, CDLC 1, Fixed timer – The headline completion-style category run, useful as a reference point for how the full-checklist route is currently standardized.
Doka – All Rifts (Base Levels, Lagless) – 15m 24s 620ms – Date achieved: “3 months ago” – PC, CDLC 1, Fixed timer – A top-end “micro-category” performance that highlights how refined rift movement and transitions have become.
Fuff – All Rifts (Base Levels, Lagless) – 17m 54s 940ms – Date achieved: “7 years ago” – PC, Version 1.0, Default timer – An early-era counterpart that helps illustrate how much rift routing and movement tech improved over the game’s lifespan.
Doka + HardcoreCheese5 – Co-op Any% (Local, Lagless) – 30m 58s 800ms – Date achieved: “6 months ago” – PC, CDLC 1, Fixed timer – A marquee co-op run that reflects how coordination and co-op-specific movement choices shape the fastest completions.
Doka – Hatless Badgeless – 48m 08s 070ms – Date achieved: “1 month ago” – PC, DLC 2 w/ mods, Fixed timer – A strong example of a challenge-style extension category where routing has to work around major capability constraints.
elisamiau – Any% NMG – 37m 48s 830ms – Date achieved: “5 years ago” – PC, DLC 2.1, Fixed timer – A key “rule-set identity” run that’s useful when explaining how the community separates core Any% from a more restricted, glitch-policy-driven variant.
Gears for Breakfast. “A Hat in Time.” April 8, 2022. Accessed February 1, 2026. https://gearsforbreakfast.com/games/a-hat-in-time/
Gears for Breakfast. “A Hat in Time – Seal the Deal OUT NOW!” September 13, 2018. Accessed February 1, 2026. https://gearsforbreakfast.com/2018/09/13/a-hat-in-time-seal-the-deal-out-now/
Gears for Breakfast. “A Hat in Time – Nyakuza Metro + Online Party Announcement!” April 25, 2019. Accessed February 1, 2026. https://gearsforbreakfast.com/2019/04/25/a-hat-in-time-nyakuza-metro-online-party-announcement/
Gears for Breakfast. “An Update for A Hat in Time Is Now Available!” October 14, 2021. Accessed February 1, 2026. https://gearsforbreakfast.com/2021/10/14/an-update-for-a-hat-in-time-is-now-available/
Gears for Breakfast. “A New Update Now Available!” January 11, 2022. Accessed February 1, 2026. https://gearsforbreakfast.com/2022/01/11/a-new-update-now-available/
Gears for Breakfast. “Online Party Outage RESOLVED.” April 8, 2022. Accessed February 1, 2026. https://gearsforbreakfast.com/2022/04/08/online-party-outage-resolved/
Steam. “A Hat in Time.” Steam Store. Accessed February 1, 2026. https://store.steampowered.com/app/253230/A_Hat_in_Time/
Steam. “A Hat in Time – Nyakuza Metro.” Steam Store. Accessed February 1, 2026. https://store.steampowered.com/app/922710/A_Hat_in_Time__Nyakuza_Metro/
Steam. “A Hat in Time – Seal the Deal.” Steam Store. Accessed February 1, 2026. https://store.steampowered.com/app/816720/A_Hat_in_Time__Seal_the_Deal/
SteamDB. “A Hat in Time (App 253230) Patch Notes.” Accessed February 1, 2026. https://steamdb.info/app/253230/patchnotes/
speedrun.com. “A Hat in Time.” Accessed February 1, 2026. https://www.speedrun.com/ahit
speedrun.com. “A Hat in Time — Guides.” Accessed February 1, 2026. https://www.speedrun.com/ahit/guides
speedrun.com. “A Hat in Time — Levels.” Accessed February 1, 2026. https://www.speedrun.com/ahit/levels
speedrun.com. “A Hat in Time — Category Extensions.” Accessed February 1, 2026. https://www.speedrun.com/ahitce
speedrun.com. “A Hat in Time (Beta Build).” Accessed February 1, 2026. https://www.speedrun.com/ahitbb
speedrun.com. “Time’s End — The Finale (Level Leaderboard).” Accessed February 1, 2026. https://www.speedrun.com/ahit/Times_End_-_The_Finale
Doka. “List of Community Allowed Mods.” speedrun.com Guides. Accessed February 1, 2026. https://www.speedrun.com/ahit/guides/9teav
Doka. “List of Community Allowed Mods.” Google Docs. Accessed February 1, 2026. https://docs.google.com/document/d/1rJOhy5440RjWR65VgU7XKRQ_4nTeDpLFpLws4Np1fBA/edit?usp=sharing
kaxymonoxy. “Complete General Game Setup Guide + Lagless Tutorial Playlist.” speedrun.com Guides. Accessed February 1, 2026. https://www.speedrun.com/ahit/guides/wmo46
kaxymonoxy. “Complete General Game Setup Guide + Lagless Tutorial Playlist.” YouTube playlist. Accessed February 1, 2026. https://www.youtube.com/playlist?list=PLsvI5_MjhLIauUM9kkYVSTQCiFYeh56ZO
Doka. “Downpatching to Non-Steam Beta Patches.” speedrun.com Guides. Accessed February 1, 2026. https://www.speedrun.com/ahit/guides/uyeh3
Games Done Quick. “A Hat in Time (Run Listing).” GDQ Tracker. Accessed February 1, 2026. https://tracker.gamesdonequick.com/tracker/run/3042
Games Done Quick. “Schedule (Event Archive Page).” Accessed February 1, 2026. https://gamesdonequick.com/schedule/40