Case Studies
Backtests, not decks.
Each case below ran on real match data over a specific window with a specific engine. Numbers are rounded to two significant figures and labelled as model backtest or paper run — no live-PnL claims unless a client has explicitly opted in.
-
IMAGE SLOT Glitch Edge research · Cloudbet sports automation 960×540 · alt: "Cloudbet moneyline strategy: 30-day paper automation replay" Paper strategy · Cloudbet API
Cloudbet moneyline strategy: 30-day paper automation replay
A simulated 30-day Cloudbet moneyline strategy run using fixed rules, paper-only bet records, and hard exposure caps. The study describes automation behavior and rejection reasons, not real customer returns or a forecast of future profit.
Paper scenario — not customer returns · 428 evaluations · 37 simulated bets · 0 live wagers
Read case -
IMAGE SLOT Glitch Edge research · risk controls 960×540 · alt: "Bankroll caps under stress: a tilt-prevention simulation" Risk simulation · Responsible automation
Bankroll caps under stress: a tilt-prevention simulation
A simulated stress test showing how per-bet, per-event, and rolling daily caps change an automated strategy's worst-path exposure. This is not a return claim and does not describe live customer betting.
Simulation only · 1,000 bankroll paths · daily cap stopped 18% of worst-path exposure
Read case -
AI betting tools · Cricket
IPL CRR vs market: a 186-innings ball-by-ball backtest
Backtested the cricket model ball-by-ball over 186 IPL innings against mid-innings book pricing under fixed staking parameters. MAE landed at 0.41 rpo and the simulated paper ledger edged the book by ~1.8% on fair-price-deviation staking. These are tool outputs on historical data — not a prediction of the PnL any operator will see running the software against live matches.
Backtest — not live performance · MAE 0.41 rpo · +1.8% simulated paper edge over 186 innings
Read case