High-power, large-aperture (HPLA) radars detect the plasma that forms in the vicinity of a meteoroid and moves approximately at its velocity; reflections from these plasmas are called head echoes. For over a decade, HPLA radars have been detecting head echoes with peak velocity distributions >50 km/s. These results have created some controversy within the field of meteor physics because previous data, including spacecraft impact cratering studies, optical and specular meteor data, indicate that the peak of the velocity distribution to a set limiting mass should be <20 km/s [Love, S.G., Brownlee, D.E., 1993. Science 262, 550-553]. Thus the question of whether HPLA radars are preferentially detecting high-velocity meteors arises. In this paper we attempt to address this question by examining both modeled and measured head echo data using the ALTAIR radar, collected during the Leonid 1998 and 1999 showers. These data comprise meteors originating primarily from the North Apex sporadic meteor source. First, we use our scattering theory to convert measured radar-cross-section (RCS) to electron line density and mass, as well as to convert modeled electron line density and mass to RCS. We subsequently compare the dependence between mass, velocity, mean-free-path, RCS and line density using both the measured and modeled data by performing a multiple, linear regression fit. We find a strong correlation between derived mass and velocity and show that line density is approximately proportional to mass times velocity3.1. Next, we determine the cumulative mass index using subsets of our data and use this mass index, along with the results of our regression fit, to weight the velocity distribution. Our results show that while there does indeed exist a bias in the measured head echo velocity distribution, it is smaller than those calculated using traditional specular trail data due to the different scattering mechanism, and also includes a bias against the low-mass, very high-velocity meteoroids.