Here's some of the actual physics related to this phenomena, for anyone interested:
Let \$v_v\$ be the vehicle speed with respect to ground, and \$v_w\$ be the wind speed with respect to ground.
For a vehicle that is in continuous contact with ground (through e.g. wheels), the amount of power available from the wind is
$$P_\text{in} = \frac{\kappa}{2} \rho A_\text{in} v_w^3 \tag{1}\label{1}$$
where \$\kappa\$ is the extraction efficiency, up to about 0.593 in theory (Betz's coefficient), and up to about 0.475 in existing devices; \$\rho\$ is the density of air; and \$A_\text{in}\$ is the cross-sectional area of the extraction device.
In addition to the ground friction losses, the amount of power lost via drag is
$$P_\text{out} = \frac{C_D}{2} \rho A_\text{out} \lvert v_w - v_v \rvert^3 \tag{2}\label{2}$$
where \$C_D\$ is the drag coefficient depending on the shape of the vehicle (varies between approx. \$0.02\$ for airfoils to \$0.5\$ for golfballs and similar), and can be different for \$v_w \gt v_v\$ than for \$v_w \lt v_v\$ (i.e. front-to-back drag coefficient can be and often is different than back-to-front drag coefficient); and \$A_\text{out}\$ is the cross-sectional area of the vehicle.
The only situation where using \$P_\text{out}\$ to model the amount of energy available from airflow to a vehicle makes any sense at all, is when that vehicle is in flight and only uses airflow propulsion, and \$C_D\$ is used to denote the energy extraction efficiency \$\kappa\$, noting that there is absolutely no physical reason or mathematical relation tying \$\kappa\$ to \$D_C\$.
For example, existing airfoil-type wind turbines, ignoring the tower, have \$A_\text{in} = A_\text{out}\$, \$v_v = 0\$ (stationary), and \$\kappa \gt C_D\$. In current commercial wind turbines, \$C_D \lt 0.1\$, with \$\kappa \gt 0.4\$, i.e. \$\kappa \approx 4 C_D\$. If they had \$P_\text{out} = P_\text{in}\$, they could not produce any power at all. It is the difference between \$\kappa\$ and \$C_D\$ that allows a stationary object to extract power from wind.
The reason there is no \$v_v\$ term (typo fixed) in \$\eqref{1}\$ is that the fact that the contact to ground, even if rolling, allows the mechanism to balance static forces with forces against the ground. A similar effect with the keel is extensively used in sailing; the keel having very little drag in the direction of travel, but enormous drag in the perpendicular direction, allowing the keel to be used to balance the forces exerted by the wind to the vehicle. In recent decades, hydrofoils are being used more and more, lifting the vessel upwards, reducing drag (\$C_D\$) due to smaller cross-sectional area \$A_\text{out}\$, without increasing the total drag much at all.
In very rough terms, a sufficiently clever land vehicle can balance the forces from the wind with static forces through the contact with the ground. Such mechanisms will have some small losses, but at least in theory, they can be reduced without any known lower limit.
While we currently believe the upper maximum limit for \$\kappa\$ is indeed about 0.593 as described by Betz's limit (at essentially \$C_D \approx 0\$), there are no known physical limitations for the ratio of \$\kappa\$ to \$C_D\$: that is just an engineering problem.
Indeed, the Mythbusters experiment by dimpling a Ford Taurus with over a thousand golfball-like dimples, increasing the fuel efficiency from 26 to 29 MPG, shows that many widely used vehicles have an unreasonably large \$C_D\$. This affects what humans perceive as normal, and results in quite normal but efficient mechanisms obeying Newtonian physics to defy human intuition. Those interested in such quirks, may find the
misconceptions related to Bumblebee flight informative.