As I pointed out, the exhaustion rate for all ballots was 15%, but I should have excluded the ballots of Adams and Garcia from that calculation, because they were the final two and their ballots were not subject to exhaustion.
So the 'real' exhaustion rate in this election was 31%, which is to say that on average 31% of the first-order votes for candidates other than Garcia and Adams did not flow to either of them. Wiley's exhaustion rate on the other hand was 29%.
But yes, these exhaustion rates clearly may have affected the outcome. It should be noted however that exhaustion can happen because of incomplete ranking - which itself may be deliberate, due to misunderstanding of the process, or lack of knowledge/willingness to rank all slots - or just because the final candidates happen not to show up on complete lists. For any candidate's voters, including Wiley's, there could have been an element of a conscious rejection of both Adams and Garcia (i.e. voters not taking tactical voting into consideration). We'll never know the answer I suppose.
I entertain that the exhaustion rate ought to be lower than 1/3 in good practice, as a matter of robust political process. Someone look up typical exhaustion rates in ranked systems internationally.
Bookmarks