Valorant shorts: format checklist

Short-form platforms (TikTok, YouTube Shorts, Instagram Reels) are different beasts from a long YouTube video. A Valorant ace that would carry a full highlight reel can still fall flat on Shorts if the first second doesn't tell the viewer why to care. This is the format we settled on while building Clutch — if you edit shorts by hand, treat it as a checklist.

Length: 15–60 seconds

Clutch cuts clips with a 15 s floor and a 60 s ceiling. The clip ends a few seconds after the last kill of the round, not at the round border, so the pay-off reads cleanly before the clip cuts.

First second: a hook that is not the kill

The viewer sees the first frame as a thumbnail in the feed. If frame 1 is the lobby, the round-start spawn, or a loading icon, the thumbnail reads "Valorant gameplay" and scrolls past. Hook options that tend to work:

• Kill count overlay ("4K in 7 seconds") burned into the first frame — tells the viewer the payoff immediately.
• Last kill first, then rewind. Chop the clip into: last kill → cut to black → "here's how it started" → full round.
• Ability callout. A Jett dash or a Chamber TP in frame 1 telegraphs the agent and the payoff.

Clutch doesn't burn a hook into frame 1 — it keeps the raw round so you can add your own before posting.

Captions: around the HUD, not over it

Valorant's HUD lives in three zones that captions should avoid:

• Top-center scoreboard (round score, timer)
• Bottom-center weapon / ability bar
• Top-right kill feed

Clutch renders captions in Bebas Neue (bold condensed) with word-by-word highlighting. Captions land in the lower-screen area of the vertical frame — not stamped on top of the weapon bar — so the kill feed and score stay readable. If you add captions by hand, keep them clear of those three zones.

Crop: three variants per clip

Clutch renders each clip in three vertical variants (1080×1920) and lets you pick which one to post:

• Composition. Top panel is a wide game crop; the bottom panel is a crop of the detected facecam (head and shoulders), or — if no facecam is found — a blurred and dimmed slice of the gameplay. Default pick.
• Scoreboard-priority (letterbox). The full 16:9 frame is scaled into the vertical canvas with letterboxing above and below. Keeps every HUD element visible — best when the top-center score or the kill feed is the story.
• Center crop. A straight center crop around the crosshair, for clips where the gunplay fills the frame and the HUD matters less.

Which variant performs best depends on the clip. If you're editing manually, letterbox is the lazy-safe option (nothing lost); center crop maximizes screen use at the cost of the kill feed; composition is the middle ground with your face on camera.

Titles: kill facts only, no win claim

Clutch generates titles from a Valorant-aware template set and refuses to assert round outcomes it can't verify. Representative outputs:

"5 kills in one round 💀"
"4 kills in one round 🔥"
"3 kills in one round 🔥"
"Two down 💀"

Phrases like "clutch", "1v5", "round win", "comeback", "W game", "victory" are filtered out when the event type is a generic multi-kill — the pipeline doesn't know if the round or the match was won, so it doesn't claim it. "Ace" is only used when the kill count actually reaches 5. If you write titles by hand, keep the same rule: state what happened (kill count), nothing you can't prove from the clip itself.

Posting cadence

A $29/mo Clutch plan gives you 10 VODs per month, and each VOD returns up to 8 clips. The quota is enough for a near-daily posting schedule if you're streaming regularly; back it off if your output is lower.

Short checklist

• 15–60 s clip, cut a few seconds after the last kill.
• Hook in frame 1 (kill count overlay or last-kill-first edit).
• Captions clear of the top-right kill feed, top-center score and bottom ability bar.
• 1080×1920 vertical in one of three variants (composition / letterbox / center).
• Title: kill count only — no win claim, no "clutch" unless you know it was.
• Post as regularly as your quota and output allow.