A 4-LLM ensemble of GPT-5.2, Opus 4.5, Gemini 3 Pro, and Grok 4 work together and cross check each other to produce optimal token analyses with a focus on game theory positioning. Sort by latest score, upside potential, or asymmetry. Vote to have tokens added to the rankings.