This match obviously had wild swings from set to set. Looking at totals for the whole match doesn't tell any individual piece of the story. So let's just have a look at the last set.
Success on serve in the fifth set (all points):
Borg 68% on first serve (15/22) and 60% on second (12/20).
Connors 64% on first serve (18/28 ) and 67% on second (4/6).
And now the fifth set, counting only rallies of 2 or more good shots (the serve being the first shot):
Borg 50% on first serve (7/14) and 61% on second (11/18 ).
Connors 60% on first serve (15/25) and 60% on second (3/5).
None of the figures on first serve are that much different in the fifth set, compared to the figures on 1st serve for the match as a whole.
But the figures on second serve are pretty good, in the fifth set. It's really a little surprising, especially when the second serves were returned: both players are above 60%.
If you think about it, neither player had an imposing second serve. They had sometimes long rallies, and yet their second serve seems to have retained its effectiveness in determining the outcome of points. And both of them were great returners. You can understand if a first serve, even returned, can throw the receiver off balance just enough for the server to end up winning the rally. But a second serve? Against a good returner? Even in rallies of some length? You would expect after 3 or 4 hits in a rally, the effectiveness of the second serve would no longer be there.
Granted, the number of second serves in the fifth set was not that large. Connors in particular started just 5 rallies with his second serve, and he won 3, so he's up at 60%. But Borg winning 11 of 18 rallies started with his second serve seems more significant.