Charge-transfer plasmons (CTP) in complexes of metal nanoparticles bridged by conductive molecular linkers are theoretically analysed using a statistic approach. The applied model takes into account the kinetic energy of carriers inside the linkers including its dissipation and the Coulomb energy of the charged nanoparticles. The plasmons are statistically investigated for systems containing a large number of complexes of bridged nanoparticles of realistic sizes generated using a simplified molecular dynamics algorithm, where the geometries of the complexes are dependent on the rate of connection of the linkers with the nanoparticles. As illustrated, the distribution of CTP frequencies in the generated nanoparticle complexes is very inhomogeneous. It has a narrow peak, corresponding to CTP plasmons in dimers, and two broad peaks, corresponding mainly to low and high-frequency oscillations in chains of connected nanoparticles. It is found that in general the plasmon frequencies depend inversely on the value of the complex dipole moment of the plasmon oscillation, where the assumption follows that low-frequency plasmons will be more efficiently excited in an external electromagnetic field. To calculate the CTP energy absorption in this field two model modifications are proposed: a system-external electromagnetic field interaction model and a simplified broadening plasmon peak model where the plasmons are calculated at first without damping and where the delta-shaped oscillation peaks are broadened then due to the damping. It is demonstrated that both modifications lead to a wide and almost monotonic absorption in the IR region for all generated systems containing a large number of bridged nanoparticles due to the presence of a large number of CTPs in this region.