Typically semicrystalline conjugated polymers exhibit much higher mobility than their amorphous counterpart. The archetypal example is regio-random P3HT, with mobilities on the order of 10-6cm2/V.s as compared to regio-regular P3HT with mobilities on the order of 0.1 cm2/V.s. The oft-cited reason for these differences is the fact that in polymer crystallites charges delocalize and take a partial 2D character, enabling higher mobility. The amount of delocalization and how it depends on structure and processing is however difficult to measure. I will show that charge modulation spectroscopy in the IR with model materials allows to determine the delocalization of the polaron in P3HT, when aided by theory. In order to extract systematic trends we will use 100% regio-regular P3HT of well-defined molecular weights (PDI~1.1). I will show how delocalization depends on molecular weight and substrate interface. Furthermore, I will compare the delocalization of charge induced by field-effect with that of charges induced by doping. We find that doping strongly suppresses delocalization, which may explain why only 5% of doping-induced charges are actually mobile.