"The league’s last 25 champions had starting point guards with an average salary accounting for only 11.3 percent of that year’s cap. That’s the equivalent of $10.6 million (or, nearly, Austin Rivers’s 2016–17 salary) under this season’s enormous $94.1 million salary cap. Only two of the championship point guards have accounted for slightly over "20 percent of a cap (Tony Parker in 2014 and Irving in 2016), and only Parker was the highest-paid player on his team. Instead of splurging on star free-agent point guards, general managers of Finals winners have typically leaned on young 1s (Rajon Rondo, B.J. Armstrong, early-’00s Parker) or PGs on inexpensive deals (Derek Fisher, Mario Chalmers, Avery Johnson), paying the spared funds to stars at different positions and key role players."
Kevin O'Connor wrote this recently in Ringer as he discusses the advantages of grabbing a rookie point guard in 2017 draft. It seems that paying IT a large chunk of salary cap, considering his age is not that logical - if history holds.
Full article: https://theringer.com/2017-nba-draft-point-guards-markelle-fultz-dennis-smith-jr-ef66c2bf3653#.ey58ayic0
Hmm... "the last 25 champions" amounts to ... how many distinct _teams_? Probably closer to a little less than half that number? So that's not really much of a sample size.
If all positions were equal, one would expect, 'on average' that each team would devote roughly 20% of it's cap to each position, with the bulk of THAT number going to the starter at that position. That means that on average one should expect the starter at each position to have a little less than 20%, except on the occasional team .... say 1 in 5 ... where the particular position might be the slot of their best player. Again, if all things were equal, that would be a different position on each team. So out of 10 teams, the PG would be expected to be the top player on 2 of them.
Things aren't equal, of course, because size is a biological rarity, so there is a premium that has to be paid for the positions of size. So that puts a little downward pressure on the 'small' position salary.
All told, I don't think Kevin's numbers are all that new or compelling. If we are really talking about 10-12 distinct championship teams across all those titles, then yeah, I'd expect the PG to be the top paid guy on just 1 or 2 of them. If you sampled across some other span of 25 years it might be 2 or 3.