Starch is the main source of energy in commonly used pig diets. Besides effects related to the extent of starch digestion, also several effects related to variation in digestion rate have recently been demonstrated in non-ruminants. Different rates of starch digestion in animals and in in vitro models have been reported, depending on the botanic origin of starch. Starches from different botanic sources differ widely in structural and molecular properties. Predicting the effect of starch properties on in vitro digestion kinetics based on existing literature is hampered by incomplete characterization of the starches, or by a selective choice of starches from a limited number of botanic sources. This research aimed to analyse the relationships between starch properties and in vitro digestion kinetics of pure starches isolated from a broad range of botanic origins, which are used in non-ruminant diets or have a potential to be used in the future. Therefore we studied starch digestion kinetics of potato, pea, corn, rice, barley, and wheat starches, and analysed the granule diameter, number of pores, type and amount of crystalline structure, amylose content and amylopectin side-chain length of all starches. Results Multivariate analysis revealed strong correlations among starch properties, leading us to conclude that effects of most starch characteristics are strongly interrelated. Across all analysed botanic sources, crystalline type and amylopectin chain length showed the strongest correlation with in vitro digestion kinetics. Increased percentages of A–type crystalline structure and amylopectin side chains of DP 6–24 both increased the rate of digestion. In addition, within, but not across, (clusters of) botanic sources, a decrease in amylose content and increase in number of pores correlated positively with digestion kinetics. Conclusion The type of crystalline structure and amylopectin chain length distribution of starch correlate significantly with digestion kinetics of starches across botanic sources in an in vitro pig model. Variation in digestion kinetics across botanic sources is not additively explained by other starch properties measured, but appears to be confined within botanical sources.