The numbers are there in
that linked post, straight from the official archives. Just saying Pistol averaging 110 mph on 2nds wasn't anywhere near as rare as some Fed boosters make it out to be.
I can go with that, as long as we acknowledge that the rest of his service game wasn't up to par. I mean the guy hadn't won a single title for a year, for gawd's sake.
That ranking wasn't meant to be taken literally, LOL. More to show that simply looking at the #s ain't enough.
But my point re: the '97 and '01 Sampras serves stands. It's a little hard to see the ball in that grainy clip so compare his serves in these '97 matches vs. Korda and Pioline (focus on DTT bombs rather than wide sliders):
With those in his 5-setter vs. Fed:
I still see a little more kick in his '97 matches, hence his high 1st-serve %s if admittedly at the expense of ace counts. So let's take a close look at his '91-02 service stats - 1st-serve %s, SGW%s, aces/match and DFs/match in that order (ignoring the double counting, obviously):
1991 - 59.5%, 87.3%, 7.7, 2.3
1992 - 60.3%, 85.5%, 9.6, 3.0
1993 - 57.7%, 89.6%, 10.5, 3.6
1994 - 59.0%, 88.4%, 10.3, 3.3
1995 - 56.1%, 88.9%, 12.1, 3.8
1996 - 59.8%, 90.8%, 12.0, 3.0
1997 - 63.7%, 90.4%, 10.6, 2.3
1998 - 58.7%, 88.7%, 11.6, 4.2
1999 - 59.9%, 89.9%, 10.0, 4.4
2000 - 62.3%, 91.4%, 12.6, 5.5
2001 - 60.2%, 86.4%, 12.3, 5.4
2002 - 58.9%, 87.2%, 14.5, 5.4
Pay attention to that big boost in 1st-serve % and the drop-off in aces (and DFs)/match in '97. That's something I noticed well before the ATP made its leaderboards available on its site, indeed without the double counting Pete probably averages close to 65% of 1st serves in for the whole year.
And I bet his URS%s if anything went up or at least stayed about the same throughout that summer. Now imagine trying to return THAT Sampras serve with the extra topspin on the old grass with unpredictable bounces. Korda did well to push that Pistol to 5 sets, ditto Becker keeping his nemesis' URS% under 50%.
As to your Q, I actually had the URS%s for that Pete-Boris match and several others from '97 Wimby, so I was able to confirm that the "service winners" on that edition's scoreboards referred to all unreturned serves sans the aces. AFAIK that's the last time Wimbledon made such detailed scoreboards on its site, so '96* and '97 are the only editions with official stats you can pull URS%s from.
*For the '96 scoreboards you need to add up the "Forehand Unreturned" and "Backhand Unreturned" totals under "Service Winners," not the SWs themselves which include aces and whatever additional serves were judged to be winners. SW is probably the trickiest stat in tennis to compare across eras due to its diverging criteria.