While many studies prove more advanced LLMs perform better on tasks such as math and trading, we notice that in cryptocurrency trading, stronger LLMs work worse than weaker LLMs often. To study how this counter-intuitive phenomenon occurs, we examine the LLM reasoning processes on making trading decisions. We find that separating the reasoning process into factual and subjective components can lead to higher profits. Building on this insight, we introduce a multi-agent framework, FS-ReasoningAgent, which enables LLMs to recognize and learn from both factual and subjective reasoning. Extensive experiments demonstrate that this framework enhances LLM trading performance in cryptocurrency markets. Additionally, an ablation study reveals that relying on subjective news tends to generate higher returns in bull markets, whereas focusing on factual information yields better results in bear markets. Our code and data are available at \url{https://anonymous.4open.science/r/FS-ReasoningAgent-B55F/}.

翻译：尽管许多研究证明更先进的大型语言模型（LLM）在数学和交易等任务上表现更优，但我们注意到在加密货币交易中，更强的LLM往往比更弱的LLM表现更差。为探究这一反直觉现象的产生机制，我们分析了LLM在制定交易决策时的推理过程。研究发现，将推理过程分解为事实性成分与主观性成分能够带来更高的收益。基于这一洞见，我们提出了一个多智能体框架——FS-ReasoningAgent，该框架使LLM能够识别并学习事实性与主观性推理。大量实验表明，该框架显著提升了LLM在加密货币市场的交易性能。此外，消融实验揭示：在牛市行情中，依赖主观性新闻往往能产生更高回报；而在熊市行情中，聚焦于事实性信息则能取得更好效果。我们的代码与数据公开于 \url{https://anonymous.4open.science/r/FS-ReasoningAgent-B55F/}。