The way to compare the efficiencies of different detect strategies (DSs) in the "ping-pong" protocol is studied. The trade-off between information gain and disturbance is calculated and compared for different DSs. The comparison result primely tallies with our intuitional analysis. It is shown that the analysis of this trade-off is a feasible way to compare the performances of different DSs in theory.quantum key distribution, quantum cryptography, quantum secure direct communicationThe task of cryptography is to make secret messages intelligible only for the two legitimate parties of the secret communication, Alice and Bob, and unreadable for other unauthorized users such as Eve. To this end, Alice and Bob have to encrypt their secret messages using a suitable encryption scheme. In 1926, the one-time pad (OTP) cipher was invented by American AT&T Engineer Vernam [1] . It was later shown, by Shannon [2] , that as long as the key is truly random, has the same length as the message, and is never reused, then OTP is perfectly secure. However, there is a problem, called key distribution. Although the public key cryptography accomplishes this task, it is based on computational security. That is, if and when mathematicians or computer scientists come up with fast and clever procedures for resolving mathematical difficult problems such as factoring large integers, the whole privacy of public-key cryptosystems could vanish overnight. Quantum cryptography, which is based on fundamental physical principles, has been proved to be an effective technique for secure key distribution [3][4][5][6] . It overcomes the drawbacks possessed by conventional cryptography and the public key cryptography, and has the vast developing prospect.Boström and Felbinger [7] presented a "ping-pong" communication protocol which can be used both to distribute a secure key (i.e. quantum dey distribution (QKD)) and to transfer information in a deterministic secure manner (i.e., quantum secure direct communication, or QSDC for short).