Strong heterogeneity, poor pay reservoir connectivity, and complex micro-pore structure are the reasons behind the high difficulty and low accuracy in the productivity evaluation of tight sandstone gas reservoirs. This paper utilized various well logging and mud logging curve data, summarizes the typical curve type, and made use of a mathematical method to form quantitative characterization. Additionally, a logging curve evaluation method is proposed based on the test of production capacity. The total hydrocarbon curve can be divided into six types as shown by the results: box shape, half box shape, upright triangle, inverted triangle, sawtooth shape, single peak, etc. The above is closely related to the level of oil and gas in the reservoir layer. Six types of total hydrocarbon curve can be quantified by amplitude, relative centroid, relative sawtooth, variance, and the parameter value of the autocorrelation function. The box-shaped total hydrocarbon gas logging curve is full and has a large scale. The half box-shaped total hydrocarbon gas logging curve peak scale is significantly smaller than that of the reservoir thickness. The triangularly-shaped total hydrocarbon gas logging curve may either be an upright or inverted triangle. The sawtooth-shaped total hydrocarbon gas logging curve has high and low tendencies that show a number of spikes. Single peak-shaped total hydrocarbon gas logging curve is small with a single peak. There is a good relationship between the six kinds of total hydrocarbon gas logging curve and the tight sandstone gas reservoir production capacity. The high and middle gas pay layers are mainly box-shaped and triangular, whereas the lower production layer is half box-shaped, triangular, and has a zigzag pattern. However, the low gas pay layers mainly possess a sawtooth pattern, while the non-production layer is mainly single peak in shape. The comprehensive interpretation of the total hydrocarbon gas logging curve and log resistivity curve can more accurately distinguish the production capacity of tight sandstone gas reservoirs. Finally, the quantitative model is used for automatic interpretation. Through data verification and analysis, it is found that the coincidence rate of the production gas pay layer is 80.95%, whereas the coincidence rate of production capacity is 78.05%. It is proven by practice that the gas formation capacity of tight sandstone gas reservoirs can be predicted based on the logging and mud logging well curve shapes. Keywords: Method of total hydrocarbon gas mud logging curve, Tight sandstone gas, Reservoir, Fracturing production well capacity