● cswatch / dispatches
Long-form writing on CS2 anti-cheat, the cheat ecosystem, and what community oversight is actually for. Updated when we have something worth saying.
We counted the warning signs across every account our detection engine currently flags — 774 of them. Here's the real fingerprint of a suspicious CS2 account, which signals show up most, and the honest caveats about what this data does and doesn't prove.
Configs, binds, and overlays are fine; automation and scripts that play for you aren't. Here's exactly where the line sits so you can customise without fear of a ban.
CS2 never got Valve's Overwatch — the community demo-review system that let players judge suspected cheaters in CS:GO. Here's what happened to it, whether it's coming back, and how cheaters actually get reviewed in CS2 now.
VAC stands for Valve Anti-Cheat. Here's what it actually is in plain English — what it detects, VAC bans vs game bans vs cooldowns, why it misses so many cheaters, and how to check any player's VAC status for free.
Prime separates your matchmaking pool and raises the cost of every account — which thins out cheaters, but doesn't make them disappear. Here's what Prime fixes and what it can't.
Account age, hours, ban history, private profiles, and banned-friend density — how to read a Steam profile like an investigator before you ever load a demo.
Why kernel-level FACEIT AC catches things user-mode VAC never will, the privacy trade-offs, and what still slips past both. The honest comparison.
Downloading, opening, and actually driving demo playback — including the controls that matter for catching a cheater (slow-mo, first-person spectate, x-ray) and why the kill-cam lies.
A plain-English glossary of the cheats you'll actually run into in CS2 — what each one does, and the specific tell that gives it away on a demo.
A VAC ban is permanent and can't be appealed — but people confuse it with cooldowns and game bans constantly. Here's exactly how long each CS2 ban type lasts and what you can actually do.
The match share code is the single most useful piece of evidence against a suspected cheater. Here's how to get one from the CS2 client, how to pull match history automatically, and why it matters.
That player wrecking your lobby might be a smurf, not a cheater — and the tells are genuinely different. Skill leaves human fingerprints; cheats don't. Here's how to separate the two before you report.
Trust Factor is Valve's invisible system that decides who you get matched with. Here's what feeds it, how to actually move it in the right direction, and how it differs from VAC and your public reputation.
Suspect a teammate or opponent is cheating? Here's how to actually verify it — checking account history and ban records, watching the demo, and knowing what aimbot and wallhack really look like (and what's just skill).
The in-game report mostly vanishes into a silent queue. Here's how CS2 reporting really works, how to grab the match share code as evidence, and how to file a report that humans actually review.
VAC catches script-kiddies but misses elite cheaters. Here's the technical breakdown of what Valve's anti-cheat sees, what it misses, and why community oversight still matters.
What experienced Overwatch reviewers actually look for when watching a suspect's demo — pre-aim before sound cue, smooth flick consistency, recoil pattern reset, and the 17 other tells that separate a smurf from a wallhacker.
External, internal, and DMA cheats explained — how today's paid cheats stay one step ahead of VAC, what trust-factor really tracks, and why hardware-level cheats are the new frontier.
Account age vs. CS2 hours, friend-network analysis, badge-collection patterns, premier-rating decay, and the other signals that flag a smurf before they ruin your match.
Analysis of CSWatch's reviewer corpus: which reviewers consistently match the consensus, which over-convict, and the heuristics that distinguish a careful judge from a witch-hunter.
Inside the reputation algorithm: how ban history, account legitimacy, community reports, and behavioral signals get weighted into one number, and why a score of 30 is a stronger signal than you'd think.
Valve's ML systems are getting better, but they optimise for false-positive minimisation, not catch-rate. Here's why human community review remains the only reliable backstop against elite cheats.
Email hello@cswatch.gg with the question. If it's a good one we'll write it up here and link back to you.