Well - those 'few' dollars, to design, develop, test and build a more advanced EDL system for the capsule would probably run into $10M's as a minimum. They would also add, in a big way, to the mass of the capsule, and thus the mass of the spacecraft and thus you use more RCS fuel to manouver around slimming your margins and you bust your weight budget for the launch vehicle and you dont launch. But you've missed the launch anyway because you needed another 18 months to design and test the system, even if you've not been cancelled for going over budget.

Even sans-'chute, the samples will be safe, burried as they are deep inside aerogel, the moment of impact causing a spike of heat that wraps each sample in a little silica ball.

Genesis crashed, and the samples were not scattered across Utah. They were scattered across the sample canister, and the most critical ones were not scattered at all. The science return is still good.

If Stardust crashes, the samples will do much better than Genesis, and the science will still be good.

Lest we forget, adding another system adds another level of complexity, another potential failure, and indirectly adds to the risk. Design as simply as possible to get the job done. Better is the enemy of Good Enough.

If you go to the point of mitigating risk to the n'th degree, you end up with a 50 ton, $25B, 50 year mission that never happens.

Doug