A Living Twin of Pollard's operational depth — paper pull-tabs, MN e-tabs, NPi iLottery, Catalyst RGS architecture, GLI math, USA + Canada markets, and TSX:PBL quant — debating your questions in real time, with full provenance.
The Council picks 3–4 specialists. Each plans their own research, gathers fresh evidence, and replies in their own voice. Click any pill to pose another question — agents work in parallel, you watch.
Each one carries thousands of words of Pollard-specific context: real lottery directors, contract cycles, jurisdictional nuance, family history, math models, certification paths. They cite specifics — not generic answers.
A sample of recent questions from Pollard leadership, with the specialists who answered. Every answer cites real numbers + sources. Click any question to open the Living Twin with it pre-loaded..
We took Arctic Polar Bear from theme card to playable + exportable spec JSON in under 6 working hours inside this Studio. Math model imported from the PA Connect family (127 reference titles, verified), art v15 generated, device frame previewed, spec.json exported, engine fed, calibrated to 70% RTP, stress-tested at 69.1% over 30 million Monte Carlo plays. Once we standardize a family template, the spec→playable path drops to under 2 hours per skin.
GLI-19 (Internet Gambling Systems) with the e-instant addendum. Our pipeline is GLI-aware from step one: math model → paytable PDF → Monte Carlo seed transcript → outcome table (the RGS consumes this) → device build with provenance hash baked in. Variants inside a certified family: weeks. Brand-new mechanic family: 8–14 weeks depending on lab queue.
A licensed third-party iLottery title carries a per-spin rev-share — industry band 8–15% of NGR — plus a per-title integration fee. Over a 10-year contract on NPi-scale, that compounds materially against margin. A first-party catalog: Pollard owns the IP, the math model, the art. On NPi alone, a conservative model (40 titles, mid-band performance) closes the cost-to-build gap inside 24 months; the remaining 8 years are pure margin. Plus speed under Pollard's name — IWG ships when their roadmap allows. We ship when your calendar needs.
PA, MI and every modern iLottery use variable-wager e-instants: player picks a stake inside an allowed range, prizes scale linearly with stake; overall odds (1-in-X) stay constant — that's a regulatory anchor. We calibrate a 7-tier ladder from chosen stake up to published top prize. Binary-search a geometric decay so expected payout per hit hits 70% RTP, accounting for multiplier bonuses (modeled as occasional, p≈18%). Verification across 100 PA titles: analytical RTP exactly 70% on 99, one at 77% (low-max-win clamp, disclosed). Monte Carlo of 30M plays → 69.1% portfolio. Real, not decorated.
Three swim-lanes. Regulatory: State A certifies via its lab, State B re-runs against its own standard (2–4 weeks overlap if math unchanged), NPi-tier states often defer with addendum. Technical: same outcome table, same paytable, same RGS contract — skins on top of an identical math seed. Commercial: Pollard sells state-by-state, Kobow ships skins on demand. We hold verified catalogs for PA 228, MI 268, MN 41, KY 58, and three Canadian provinces. Multi-state is data, not pitch.
Type your question in plain English (or French, or Portuguese). The Council auto-routes to the 3–4 specialists whose domain matches.
Each specialist responds in their own voice — Erik's Minnesota cadence, Étienne's French inflection, Victor's quant rigor — with real numbers, real titles, real contracts.
A moderator condenses the debate into consensus + disagreements + a 2-3 line actionable takeaway. Sources cited inline.
Sessions remember context. Aggregate intel sharpens over time. Pollard-mode keeps data isolated from the public pool.
Click below to open the Council in demo mode. Ask Erik about MN GCB, Étienne about California Loteria heritage, or Victor for the TSX:PBL EV/EBITDA picture.
Open Living Twin →