Is it down to the CS prop??
I have a feeling that for the Dakota with the c/s prop, "best-glide" is calculated with the prop lever fully forward. I think this is due to it being the "worst case" scenario, as with a lack of oil pressure to the prop, it will fail to fully-fine (as opposed to most Multi-Engine types where it fails to a feathered position) and cause more drag when windmilling.
You will see this for yourself if you descend power off at a set airspeed with the prop lever forward and note the descent rate, and then pull the prop lever all the way back and compare at the same airspeed. Think of it that you are commanding a prop RPM from the airflow (in a power off descent) by moving the prop lever, so that the governor when trying to set a lower RPM, does this by setting a smaller blade AoA and taking a lesser bite of the air. This in turn creates less drag, and gives you the lower rate of descent.
I have done this in an Arrow, and it makes quite a marked difference. I'm sure that such a configuration was mentioned in the Arrow IV POH, but don't have access to a copy to check.
Regards.