After studying the classical multi-coin quantum game, aiming at the decoherence phenomenon caused by the coupling between the quantum system and the external environment, using the quantum game theory, the multi-coin quantum game models under the conditions of phase damping channel, amplitude damping channel and depolarization channel are established, and the variation laws of the quantum player’s winning probability under three kinds of noise channels are deduced, The two influencing factors: noise intensity and the number of coins are analyzed, the functional formula is calculated, and it is concluded that there is a significant negative correlation between the quantum player’s winning probability and the noiseintensity and the number of coins. The more coins, the faster the quantum participant’s winning probability changes with the noise intensity. When the noise intensity reaches the maximum, the quantum strategy loses its advantage, It shows that classical participants need to delay time and increase the impact of noise if they want to win when they realize that the other party uses the quantum strategy. And when the system is in a closed state, the number of coins has nothing to do with the game result, the quantum player will always win. In the open environment, the more coins, the faster the winning rate of the quantum player decreases with the increase of noise intensity. In addition, among the three channels mentioned above, the phase damping noise channel has the least impact on the winning rate and the depolarization noise channel has the greatest impact. These laws will not be affected by the number of coins. Therefore, in order to obtain greater advantages in the game, the player using quantum strategy should choose the amplitude damping channel if they can choose, in order to reduce the of quantum player, participants using classical strategies will choose the depolarization channel, which provides a theoretical basis for multi-coin game decision-making under different external environment conditions.