The major target of this research is the classification of pyrite types using fractal and stepwise factor analyses in the Chah Zard ore deposit, Central Iran. The ore occurs within a breccia/vein type and the major ore mineral hosting gold mineralization is pyrite. In this study, data were selected using optical, scanning electron microscopy (SEM) and backscattered electron observations as well as laser ablation-inductively coupled plasma–mass spectrometry (LA-ICP–MS) analysis. Conventional interpretations represent four gold-bearing pyrite types of various textures including fractured and porous Py1, oscillatory-rimmed and simple-zoned Py2, colloform Py3 and inclusion-rich Py4. The stepwise factor process was performed on the centred log ratio (clr) transformed data in two phases and Au was grouped with As, Te, Ni and Co in the second factor from the second stage (F2-2). Also, C-N fractal modelling was performed on the As, Au, Te and F2-2 values, all of which demonstrate multifractal nature. Four populations were separated based on F2-2 values and the C-N log–log plot. The main gold mineralization starts from 32 ppm, 2.8%, 7.94 ppm and 1.26 for Au, As, Te concentrations and F2-2, respectively, based on the C-N fractal modelling. These values are correlated with inclusion-rich Py4 and simple-zoned and oscillatory-rimmed Py2. The results obtained in this study show that fractal interpretation of LA-ICP–MS data by stepwise factor analysis may provide a suitable tool for the recognition of ore mineralization in epithermal gold deposits.