One of the four designs for the Orbital Space Plane, intended to be a kind of Honda Civic of space (a cheap people-mover) compared to the pickup truck that the shuttle was meant to be, is a capsule. The other three designs are closer to what we'd recognize as a mini-shuttle. Here's
a picture of all four designs.
It seems to me that saying that delta-wing spaceplanes are unsafe right now is kind of like saying that all hatchbacks are unsafe because a couple of Pintos that were rear-ended at highway speeds exploded in the 80's.
(I've been in a Pinto that was rear-ended, and I'm still alive.)
Furthermore, I'd say your safety figures are inaccurate.
Consider that the shuttle had more than 100 launches. Of those, there have been two failures, and one of those (Challenger) was a failure of the booster and not the spacecraft itself. 2 failures in more than 100 gives us better than 98% reliability. We get better than 99% if we only look at the shuttle itself.
There have been four failures in American capsule-type spacecraft (two unmanned Mercury missions with booster problems, Apollo 1 and Apollo 13 with capsule problems), with 54 launches (by my count, 14 Mercury, 19 Gemini, 17 Apollo, 3 Apollo-Skylab, 1 Apollo-Soyuz... I guess Apollo 1 wasn't technically a launch, but it was a capsule and 3 men died in it). 50/54 = 92.5% reliability. If we ignore the boosters and only look at the capsules, we get 52/54 = 96%.
~98% > 92.5%; the shuttle system has been more reliable than the capsule & booster system.
~99% > 96%; the shuttle itself has been more reliable than the capsules flown.
CONCLUSION: In the American space program, capsules are LESS reliable.
I haven't counted failures in the Soviet capsule program, because I don't have data for them.