It used to be when stages first released their PM's, they were much less expensive than many other PM's at the time. That is certainly no longer the case. SRM's and quarks were 30% more expensive. Many other competitors, SRM, Quark, P2M, have lowered their prices. The notion that stages is so much more economical is simply uninformed.
No offense to DC's reviews, as he really tries to do a great job and for the most part does, but for someone who expects a true reading from a PM that reflects instantaneous acceleration, solid sprinting watts, really dynamic power and cadence variability with data integrity, you are NOT going to get that with a stages. No disrespect to DC, but his max sprint wattage at around 800 watts, what he attains in his sprints in data I've seen from him, is regularly exceeded by cat 3 women in my area.
I have owned several SRM's (nearly 10), quarks, powertaps, and certainly stages. I have ridden with DA SRM right crank/spider and a left side stages with 2 head units for many rides. I currently own 2 stages (for no reason other than the fact that they were nearly free and I wanted them on my CX, and spare bikes), and a FSA P2M. The stages PM's I've had have been consistently as much as 150 watts off compared to my SRM's when exceeding 900 watts. It's been like you count 1 alligator, 2 alligator before the stages would catch up to the SRM readout. DC uses smoothing, somtimes over long periods of time, when he posts comparison graphs. That is so that the reader has an easier time looking at graphs, at the expense of accurately tracking variability, You can reduce variability in testing results to near nil using a basically worthless duration of smoothing. Why even bother with a PM if that's the case. Use HR if you don't mind 30sec plus smoothing in your power files, as it can take over 30 secs for HR to respond to efforts. I have used many power meters and his grouping as equivalent, the many power meters currently available, simply misses the mark and is a disservice when it comes to the accuracy of the data you're going to get at the end of a truly variable and dynamic training session or race.
If you are really interested in seeing what a solid effort is, such as in a fast group ride, accelerating on your mtb on a short steep climb (regularly over 900W) attacking repeatedly at over 1000W to split those behind you, highly variable power outputs such as when going from coasting to 600W+ efforts, then my experience with stages has been that the data just does not show the same results as my SRM/P2M/quark.
If you plan to go out and ride static core endurance, or plan to sit at a tempo pace and not vary much, or ride TT's, and won't be doing hard group rides, stages might suffice. But with the retail price of a P2M near $1000 and just a few hundred dollars more than a stages, as well as quark, why would one bother with a left arm when for the price difference, you get a complete crankset with chainrings. Why not buy a 4iii for $200 less if your expectations are that you would buy a stages?