SideGuy Betting Lab

01Why this matters

1. Hard sizing rules — max % of bankroll per chaos position vs per core anchor. Without rules, "structure over gambling" becomes "structured tilt" the first time PJ catches a hot streak. 2. Stop-loss convention — at what loss does a position auto-exit? Live markets can move against you fast (a leadoff HR followed by another). 3. Slippage assumptions — Polymarket and Kalshi have wide bid-ask spreads on event markets. The "+240 → +120" example assumes you can actually fill at those prices. Often you can't. 4. Sample-size requirements — how many cycles before the strategy is validated vs lucky? (50 trades minimum? 200?) 5. Vig + fee math — Kalshi takes a fee per contract; Polymarket has gas + spread. Need to bake into expected value. 6. Hedging mechanics — when the core rail is hit and the volatility rail is also up, when do you take profit on one to lock the other? (Bracket orders aren't standard on these venues.) 7. Bankroll segregation — chaos rail bankroll separate from core rail bankroll? Or one pool? The doctrine doesn't say. 8. Information edge — what's PJ's actual edge over the market? "Humans overreact" is true, but the market includes ALL the operators who also know that. Need to identify the specific informational/timing edge.

02How to apply

Read the doctrine. Apply it on the next ship cycle.

03The closer

PJ-shipped 2026-05-07 afternoon. A two-rail trading framework for live sports markets (Kalshi + Polymarket/Underdog) that explicitly treats volatility as the product, not predictions. Core thesis: humans + sportsbooks + prediction markets overreact to short-term emotional events (HR, error, single inning) — the edge is the disconnect between emotional repricing and true remaining win probability. Sister positioning to operator-translation doctrine: translating market chaos into structured trading, not betting on outcomes. Apply when PJ surfaces sports-market discussion or wants to build the Betting Lab as a SideGuy vertical.

04Text PJ — no Calendly, no meeting