If I recall correctly the "land on land" scenario for dragon has it using parachutes most of the way down, then doing a drop and land at the end.
On the moon, there is no atmosphere, so the capsule would have to handle all the deceleration from lunar orbital velocity (~2,000 m/s) to 0. So, in ballpark figures, you're looking at 20 times more 'delta V' (change in velocity) that the capsule would have to accomplish in order to land on the moon. It takes _way_ more fuel to manage that.
No, FH is advertised [1] as 53 mT to LEO (probably around 200x200 km, 28.5°).
Interestingly, Dragon 2 does have enough fuel to land on Mars, with clever aerobraking. NASA has done some feasibility studies on using a Dragon to land on Mars with a sample return vehicle inside it.
They demonstrated this with their pad abort test:
https://www.youtube.com/watch?v=1_FXVjf46T8
Perhaps later this year we will also see an in-flight abort test, which will be very exciting.
Parachutes cannot act as backup to a rocket-powered landing. Chutes need speed+altitude to deploy. If the rockets fail 25 meters above the ground, chutes won't save anything. If the craft has chutes+people on board, they will be deployed long before touchdown. Better out than in.
The only fully propulsive landing option, with humans on board, would be to have a second set of engines ready to perform what in an airplane would be called a zero-zero ejection maneuver[1]. This would see the craft reverse course, ascending rapidly to a trajectory where the chutes have time/energy enough to properly deploy. It would basically be a launch escape system. To keep that, and the chutes, and the descent engines ... just use the chutes. It isn't saving much money not deploying them as they will probably require repacking for inspection either way.
[1]A zero-zero ejection seat is one that can save a pilot from a craft parked on the runway, at zero-speed and zero altitude. Older seats required altitude and/or speed. But the dragon would be descending, with a negative alt/speed, so I guess this would be a sub-zero-zero ejection. A descent escape maneuver?
From what I've seen, it looks like SpaceX plans on using the rockets to cushion a landing already under chutes, like the russians do today. That allows for much smaller chutes and faster landing. If the rockets fail then the landing with be rough but still survivable, with the re-usability of the craft being written off.
The propulsive landing of booster stages makes sense only because you already have giant engine/fuel reserves, which aren't going to orbit. And since there are no humans around you don't need any sort of abort option.
Also, if you are going to have a potentially rough landing, would you rather do so sitting 6" from a box containing a parachute, or 6" from a tank of rocket fuel?
The former, of course. I agree that a propulsive landing wouldn't be likely any time in the near future. But if it could eventually be shown to be safe (and therefore the quote above doesn't apply), I don't necessarily agree that it would be more expensive. You wouldn't need large engines or a lot of fuel, as the capsule wouldn't be very heavy. Yes, it would definitely still be more expensive than parachutes in terms of equipment and weight. However, it would mean you could land right back at your facility, rather than splashing down in the ocean and dealing with retrieval. When we're looking into the future of reusable, commodity space travel, that would be a significant advantage.
http://spaceflight.nasa.gov/history/station/x38/parafoil.htm...