Satellites hardly ever fail on launch. Once in a great
while, they miss on their trajectory by a sector, but
that's it.
This is silly.
In reality, early satellites and probes to the Moon and
other planets were fraught with huge difficulties. NASA
launched five probes in a row that failed completely. The
Soviet Union faired worse, losing everything they
launched to interstellar space from 1959 to 1967.
I propose that satellite launching success be scaled
relative to the tech that a given satellite arrives at.
Thus, a tech 305 spy sat available at tech 305 has a
very iffy chance of making it into orbit. Say, 20%
chance for example. As tech rises above 305, then the
chance increases.