Lisa Savage seems to active in the comments of the article this was posted in if anyone here has any questions.
They've polled second choices for the Senate race but not compiled results with RCV and not even compiled crosstabs for second choice by first choice preference. Thankfully, they wrote a bit on it in this article:
https://www.colby.edu/news/2020/10/28/poll-shows-statistical-dead-heat-between-collins-and-gideon/Colby College
October 21-25
Changes with September 17-23
Statewide879 likely voters
MoE: 3.3%
First RoundGideon +3% pre-rounding
Gideon 47% (+2)
Collins 43% (+2)
Savage 5% (n/c)
Linn 2% (-3)
Unsure 4% (-2)
Second Round (only Savage votes reallocated)Their September 17-23 release had this to say on RCV:
However, their second choice was mostly scattered, with Gideon and Collins each netting about two percentage points.
6% remained undecided.
This release actually specifies the way Savage voters break (and includes voters who didn't bother to rank a second choice), so I'll extrapolate below.
Gideon +5
Gideon 49%
Collins 44%
Linn 3%
Undecided 4%
On other voters, it says this:
Using a ranked-choice process, the poll found that Lisa Savage voters selected Gideon as their second choice roughly 51 percent of the time, Max Linn 29 percent of the time, and Susan Collins 13 percent of the time. Linn voters, a much smaller cohort, cast their votes in a less consistent way.
“Given the closeness of the race, and the fact that minor-party candidates will probably net about seven percent of the overall vote, it’s likely the ranked-choice process will matter,” said Shea. “In our opinion, this may help Gideon more than Collins.”
ME-01426 likely voters
Pingree 58%
Allen 31%
Unsure 11%
ME-02453 likely voters
Golden 56% (n/c)
Crafts 31% (-2)
Unsure 13% (+2)