Innovation has attracted attention of researches in last 20 years, while networks and clusters are relatively new research subjects. In our paper we made an attempt to find the relationship between network centrality indexes and innovation performance. Each index represents different features of being in the network. To find the network indexes we have constructed adjacency matrixes based on alliance data. For our research we have chosen China’s automobile industry network as an example, for the reason that Chinese automobile industry showed tremendous growth in recent decade and is fit to research scope which we are conducting. We have collected the data on innovation performance for 59 firms in China’s automobile industry. We used UCINET software program to get the data regarding network properties. After we ran the negative binomial regression model on Gretl software program and constructed 5 models, with total of 7 variables. We have analyzed the relationship between innovation performance and three network centrality measures. According to our new findings firms in the network with more total number of connections and firms with more connections with well-connected firms have better innovation performance. We found that there is no effect on innovation performance when firms have capability to pass information fast.