AbstractAnthocyanins are phenolic compounds that provide colour to young red wines following extraction from grape skins, and their reaction kinetics are not well understood as they exist within a complex pH-dependent multistate system. For commercial wineries to best control anthocyanin colour expression that is a major determinant for red wine quality, there is a critical need for behaviours of all pH-dependent monomeric and self-associated anthocyanins to be considered together. In response, the current study employed mathematical and experimental techniques to reveal kinetic and steady-state behaviours of all species of a model anthocyanin, malvidin-3-O-β-D-glucopyranoside (M3G), within three Shiraz red wines throughout fermentation. Investigator-developed models represented the following: (i) acid–base reactions that form red, purple, and blue anthocyanin monomers; (ii) hydration reactions that result in monomeric colour loss; and (iii.) physical self-association interactions that temporally stabilise anthocyanin colour. Simulations were validated experimentally using high-performance liquid chromatography with diode array detector (HPLC–DAD) and colourimetric analysis of wine samples. Moreover, a unique predictive modelling framework was developed that takes in measured values for anthocyanin concentration, pH, and temperature at the onset of fermentation, and output data describing corresponding wine colour expression and stability characteristics that would arise after fermentation has progressed. Results of the current study elucidate complex reaction kinetics occurring between anthocyanin species under dynamic red wine conditions and provide meaningful predictions for future wine colour and shelf-life characteristics. Outputs may be used to inform key winemaking decisions influencing red wine quality and to optimise the use of winery resources throughout the fermentation process.