Graphene has excellent electrical, optical, thermal and mechanical properties, so it has been used in high-performance field effect transistors, sensors, optoelectronic devices, and quantized devices. It is crucial to realize a high-quality junction between metal electrode and graphene. For example, in the field of electrical measurement, due only to the contact resistance in a proper order of magnitude, the quantum Hall effect can be realized. The lower the contact resistance, the higher the measurement accuracy of Hall resistance is. In order to reveal the factors affecting the contact resistance we propose an effective method to reduce it, and a physical model is established in this paper. The carrier transport between the metal electrode and graphene is divided into two cascaded processes. Carriers first transport from the metal electrode to the graphene underneath it, then transport between the graphene underneath metal and the adjacent graphene. The transport probability of first step is considered through the effective coupling length and the mean free path. The transport probability of second step is considered through the effective length of potential step change between the graphene under the metal and the adjacent graphene. The contact resistance is analyzed by combining the distribution of carriers. In order to verify the correctness of the theoretical results, an experimental sample with gold as the metal electrode is fabricated. The transport line model is used to measure the contact resistance. The length of contact area is 4 μm. The lengths of graphene channel are set to be 2, 4, 6, 8, and 10 μm, respectively. The current values are set to be 10, 20, 40, 60, and 80 μA, respectively. The results show that the relationship between current and voltage is almost linear. The total resistance can be obtained with different lengths of graphene. According to the transmission line model, the resistance value can be estimated as (160±30) Ω when the graphene length is zero. Considering that the measured result is obtained under two metal electrodes contacting the graphene, the contact resistance of experimental result is (320±30) Ω·μm which agrees well with the theoretical result. From the analysis of theoretical process, the factors that affect the contact resistance is determined by material, drain-source voltage, gate voltage, doping concentration, distance between metal electrode and graphene atoms, distance between graphene and gate. Finally, in order to reduce the contact resistance between graphene and metal electrode, we propose some corresponding solutions for choosing the metal material whose work function is close to graphene's, reducing the thickness of the silicon dioxide layer, increasing carrier mean free path, improving the surface morphology of the metal material, and reducing the coupling length between metal and graphene.