Like it or not the market will compare EV-DO against every other widely available technology.
Sprint still has this economic issue to address spending many billions on a network that potentially delivers lower revenue per customer than what they already have.
Consumers have consistently shown they prefer one device for communications and WiMAX will not satisfy both the voice & data need. For Sprint to overcome the voiceless WiMAX limitation they have to rely on more costly and complex dual radio solutions that don't yet exist and will take more time to develop. By the time the handsets might be available in volume WiMAX would have lost virtually all the time advantage it claims it has.
The case for Sprint broadband is strong, the case for voiceless WiMAX is weak.
Motorola already had a prototype idEN/WiDEN/WiMAX handset designed for Nextel before the merger with Sprint was announced. So there's the convergence idea. Now, if you want to address CDMA, then just replace the iDEN/WiDEN core of Harmony with a CDMA EVDO core. It can be that simple. Thus, you have voice, high speed data, and wickedly high speed data.
And, because it's WiMAX enabled, Sprint could opt to NOT subsidize the handset, allowing the consumer to make the choice of how to utilize the device as they see fit. Nifty concept that is actually.
BTW, I don't buy your theory that WiMAX will result in delivering lower margins (lower revenues, perhaps...but that won't matter since the margins will be so much better).