The new coronavirus pneumonia data of Tianjin city is complete and data link is clear. Based on the data, the data characteristics of new coronavirus cases are studied by statistical analysis. The statistical characteristics of latent period are thoroughly explored by lognormal distribution model. The digital characteristics of lognormal distribution are obtained by maximum likelihood estimation. The mean of the latent period of the case is 12.6 days, and the standard deviation is about 6.1 days. The latent period of 95% patients was within 24 days. It can be obtained from the function of excess that the average latent period was about 30 days for patients with latent period more than 24 days. These results are reasonable explanations for the phenomenon of extra long latent period in some patients. At the same time, single factor analysis of variance showed that the virus had no relationship with age and gender, but was suspected to be related to the way of contact.