Kentucky and Oregon hold their presidential primaries tomorrow. We'd normally publish our polling projections but there has been just one poll released in each state, so perhaps preview is more appropriate. We'll start with Kentucky's lone poll:

The poll is from June and it means absolutely nothing some 11 months later.

From a geographic perspective, each of Kentucky's seven neighboring states have already voted; Indiana and West Virginia were the most recent and also the only two to vote for Sanders. Kentucky has a closed primary, which means only registered Democrats can vote; the deadline was April 18th to register or change affiliation. From a delegates perspective, Kentucky will allocate 55 pledged delegates [1] with 37 coming from its six congressional districts. The assignment breakdown and allocation thresholds are presented below:

Five of 8 allocation groups are numerically even which means the delegate margin will be minimal if the results are close. There are only three congressional districts that matter; CD-1 is about 35 counties in the west side of the state, CD-3 which is Louisville in Jefferson County, and CD-6 includes portions of 19 counties surrounding Lexington. The only threshold realistically worth watching is the 8.33% statewide margin for the at-large pool. The winner of Kentucky will likely finish with either 29 or 30 delegates, which is a +3 or +5 delegate margin.

Let's move onto Oregon which is slightly more interesting numbers wise despite also having just one poll:

This poll is fundamentally flawed in that it assumes that just 21.05% of the electorate will be under the age of 45. Here are its results by age group:

The lowest turnout rate of the under 45 age bracket, in any nominating contest this cycle with exit polling was Oklahoma at 33% [2]. In Oregon's 2008 Democratic Primary, the turnout of those under 45 years of age, according to exit polling, was 42% [3]. DHM Research's turnout model is predicting that Oregon's youth turnout (<45) will be more than 10% lower than any other state so far and half the turnout of 2008.

DHM Research also released a high turnout model [4], which slightly addresses the problem by increasing the under 45 turnout to a still deficient 28.95%:

If the age results are re-adjusted to more closely align with the 2008 exit poll, Bernie is leading by about 4% at 44%-40%. The age groups don't exactly align as the age groupings in the exit poll use 60+ rather than the 65+ used in the DHM Research poll. This simple extrapolation used a simple multiplication across the 4 age groups despite the fact they don't quite match up; this nuance actually favors Clinton as she performs better among the oldest demographics.

Oregon will allocate 61 pledged delegates [5], in its five congressional districts. They split the 2nd district into an A and B pair to reduce the travel time of participating delegates given the large geographic area; each sub-district has 3 delegates. The assignment breakdown and allocation thresholds are presented below:

Oregon, like Kentucky, has eight allocations groups, but unlike Kentucky, all but one group is numerically odd. The thresholds in Oregon are all greater than 12.5%, so the winner is likely to walk away with something close to 34 delegates, or a +7 margin.