This paper tries to re-explore the reasons why the prices fluctuate, in microscopic and emergent perspectives, more than expected in an agricultural market, using complex system modeling and agent-based simulation. The model mimic a real world agriculture market by establishing a price determination mechanism. Resorting to the chaotic analysis, averting from accurate price prediction, we find that, even for the simple case where farmers can only choose between two different predictors, with initial equal distribution, smaller entropy, i.e., more aggregative cultivation, usually gives rise to larger price fluctuation.