The right answer is they should fund all three. That was the intention of the program from the beginning. This would likely look like two full awards and one "h

The right answer is they should fund all three. That was the intention of the program from the beginning. This would likely look like two full awards and one "half" award that would be more of a technology development effort. The reasons are three-fold. 1) Cost savings. If you look at the joint strike fighter as an example, as long as Boeing and Lockheed were both in the game costs for the program were historically low. Both companies built and flew prototype aircraft for $750 million each. That's crazy cheap. It was only after down-selecting to a single aircraft that the costs of the program ballooned out of control, because that's how defense contractors play the game when they have no competition. NASA's Ares and Orion programs, by contrast, were contracted out to various single winners from the beginning, who quickly ran up the tab. 2) Backups. In case one runs into technical difficulties, NASA will simply order more of one of the other companies. If you have a single winner, any kind of incident with either the launcher or the spacecraft will mean a stand-down of US human access to space for months, not years. 3) Variety of capability. Multiple winners means a variety of designs, differing capabilities, and a larger variety of technologies developed that might be useful for future spacecraft. E.g. SNC's horizontal landing. Source: I had a hand in the initial design of the program.