I looked through some research articles and I found a few things worth noting:
I think we're looking at what's called "subcritical flow" where the Re is high, but not high enough for the drop off in drag associated with very high Re and a fully turbulent boundary layer.
First off, the near wake mechanism for subcritical spheres from "PIV analysis of near-wake behind a sphere at a subcritical Reynolds number", Young Il Jang, Sang Joon Lee
The flow around a sphere shows steady axi-symmetric flow in the range of Reynolds number 20–210. The axisymmetry
is then broken, and planar-symmetric flow appears until Re = 280. From Re = 280, unsteadiness
starts to occur in the planar-symmetric flow, and hairpin vortices are periodically shed. In the range of Reynolds
number 420–800, asymmetric flow is observed, and unsteadiness continues (Taneda 1956; Nakamura 1976;
Wu and Faeth 1993; Johnson and Patel 1990; Leweke et al. 1999). Many experimental and numerical research
works on sphere wake have been carried out in this Reynolds number range to study laminar flow separation
and laminar wake. On the other hand, from Re = 800, the large-scale low-frequency vortex shedding and small-scale
high-frequency shear layer instabilities become prominent flow phenomena. A large-scale vortex is shed with wavy
shape, and turbulence occurs in the far field. At the critical Reynolds number of Re = 3.7 9 105, the drag
coefficient is rapidly reduced. This results from sequential laminar separation, reattachment, and turbulent separation
from the boundary layer of the sphere. The sphere wake becomes fully turbulent beyond this critical Reynolds
number. In the subcritical Reynolds numbers from 800 to 3.7 x 10^5, the drag coefficient of a sphere has almost
constant values, and the flow separates laminarily from the sphere. In addition, Kelvin–Helmhortz instability
occurs in the separating shear layers, and the wake becomes turbulent.
So the wakes are different on paintballs than they would be at Re ~ 100
I found this article, you'll need to purchase it if you want to read it, I could post figures and quotes but I'm not sure if I can get in trouble for doing that. It Covers the wake mechanism of a sphere at subcritical and critical Re but not spinning spheres. if you are a student ask your library and they can get you a free copy (usually):
"Numerical investigations of flow over a sphere in the subcritical and supercritical regimes", Physics of fluids [1070-6631] Constantinescu, George (2004) volume: 16 issue: 5 page: 1449 -1466
Sub critical drag coefficient fluctuations: Cz and Cy are basically lift coefficients
compared to the supercritical equivalent figures:
This shows the sideways forces on the ball that contribute to "random walk".
I honestly can't find anything about streamwise rotating spheres at high reynolds numbers.
The article Snipez posted though is pretty sweet.
<img src="http://img.photobucket.com/albums/v205/Leftystrikesback/Paintball/Sig4.jpg" border="0" class="linked-sig-image" />
<!--sizeo:1--><span style="font-size:8pt;line-height:100%"><!--/sizeo-->"do the math, save the world"<!--sizec--></span><!--/sizec-->