Anthropic holds a strong 63% implied probability as the second-best lab on the Text Arena Math leaderboard by end of October, driven by multiple Claude Opus 5, Fable 5.1, and Opus 4.6 variants consistently ranking in the top eight positions behind Google's leading Gemini 4 Argon models. Recent arena voting data through October 22 shows these Anthropic systems maintaining tight score gaps of under 10 points on math-specific prompts, reflecting robust reasoning capabilities in algebra, geometry, and multi-step problems. Google sits at 27% as the clear frontrunner, while OpenAI and others remain below 5% due to fewer top-placed models in the current snapshot. With resolution days away, traders are weighting the latest live arena results heavily, though minor leaderboard shifts from additional votes could still influence final lab attribution.