There's an opinion piece on SportNet's web site by Donnovan Bennett about the CFL's rule changes. To me the most interesting part of it is that it has stats on what the completion rate was for field goals from the new starting point:
"During the 2014 season, 99.4 percent of extra-point kicks were executed
successfully while field-goal attempts from between 31 and 33 yards went
through the uprights 81 percent of the time."
Assuming naively that would result in an average of 0.81 points for single point conversions versus the previous 0.994 points, that means that if the chance of making a two point convert from the three yard line is higher than 40.5% a team should always go for two barring end game situations.
Bennett does point out though that the field goal completion percentage for point afters will likely be higher than 81%:
"Add to that the fact that 81 percent isn’t a true metric when predicting
“new extra point” success rates. Extra points will be kicked from the
middle—not the wide hashmark the ball could be spotted on for a FG.
Also, since there is no threat for a return, kickers don’t have to worry
about drilling the ball through the end zone in case of a miss.
Depending on which kicker you ask, that should improve the percentage of
long extra points that are completed by another five to 10 percent."
Of course the completion percentage is probably less outside and highly weather dependent (and kicker dependent considering some of the mediocre kickers lately). For the Cats this year with Medlock, 32 yard field goals aren't really a big deal for him.
I'm not sure what the league expects in terms of going for two. Teams going for two all the time? Humans psychologically don't really like losing what they have, even when the odds are in their favour which going for two may be. Plus going for two and not making it will likely bring opprobrium from television commentators and newspaper columnists who aren't really known for their sophisticated statistical skills.