In this paper, an approach for links reliability estimation in bacterial nanonetworks is presented. In particular, we focus on the usage of replication coding to increase the network performance (i.e., the transmitter nanomachine sends a set of bacteria with equal plasmid messages towards the receiver nanomachine). Unfortunately, the analysis of accurate bacteria mobility model in both 2D and 3D is quite complicated. Therefore, a 2D lattice grid is proposed as a reasonable simplification of the bacterial mobility model, while the success delivery estimation technique is based on the First Passage Time approach. This step allows us to derive the rough and scalable estimation for network characteristics. When being compared to results obtained by simulation of the network with realistic bacteria mobility model, achieved approximation seams being a lower bound, accurate enough for the distance of few millimetres between nanomachines --- distance bacterial nano networks were originally proposed for. The presented analysis leads to a method that is capable of predicting the amount of bacteria to transmit, which is dependent on the environment characteristics in order to satisfy the link quality requirements. By developing an analytical model, this provides a scalable approach for studying bacterial nanonetworks that is far more efficient compared to current simulation techniques.