Roobet Crash

About Roobet Crash

Roobet Crash is the in-house crash game offered by Roobet, a crypto-native casino that launched its crash variant in 2019. The game uses an HMAC-SHA-256 derivation comparable to Stake's, with a server seed committed before play and revealed on rotation.

The notable difference from its peers is on Column A. Public sources report Roobet Crash's declared RTP in the 96.5–97% range, corresponding to a 3.0–3.5% house edge — between 2.5× and 3.5× higher than the 1% house edge declared by Stake and BC.Game. We treat this as a stated parameter, not a finding, and our null hypothesis for Column A will be the operator's own published value rather than the category-wide 99% benchmark.

Why this game matters

A higher declared house edge is not a fairness concern in itself — it is a price disclosure. What we audit is whether the observed RTP matches the declared RTP within statistical tolerance. Roobet matters in our coverage because:

1. Pricing transparency. Roobet's declared figure makes it the highest-priced crash game in our initial coverage. Players choosing between platforms can read this fact directly off the methodology and form their own preference.
2. Tolerance calibration. Our Whitelist tolerance bands are absolute percentage points, not relative. A 1.5pp deviation means very different things at 99% and at 96.5%, and our published reports will make this explicit.

What we are watching for

• Column A (RTP): observed RTP tested against the operator's stated value (currently 96.5–97% pending operator confirmation). Any drift below the stated value at scale would matter regardless of the absolute number.
• Column B (Distribution): goodness-of-fit against truncated-geometric with house_edge ≈ 0.03. Distribution shape is independent of price level.
• Column C (Provably-fair): standard hash-chain reveal verification. Where Roobet's published documentation diverges from peers, we will note it explicitly in the audit report.
• Disclosure consistency: Source 1 (Roobet's own pages) is cross-checked against Source 3 (our proxy account observations) for any drift between the published declared RTP and what the live game communicates round-by-round.

Public data sources we are using

• Source 1 (Official): Roobet's /fair page and per-round verification interface.
• Source 3 (Self-operated proxy): a Clash Watchdog AI proxy account at minimum stake.
• Source 2 (Community): opens in Phase 2.

Cross-validation rules: see Three Data Sources and our methodology §2.

Audit status

A first Tier 1 (Provisional) audit is in preparation. Because Roobet's declared house edge is materially different from its peers, we are prioritising baseline Column A collection so that any future statement we publish is anchored to a sample large enough to distinguish a 0.5pp drift from sampling noise. If you have observed something unusual on Roobet Crash, tell us.