A couple things I would suggest. First, run a subtotal for only people who are in the weighted pool, because effectively they are the only ones in the draw. Those who have 3 years or less in the system have no effect on the drawing.
Also when I use the term "point creep" it is usually in reference to a preference type system or preference type variation (like Utah has) to the draw. The "creep" is how the points required to be in the tag draw pool goes up each year. "Point creep" in the way I have usually used it has to do with not being able to draw (or having extremely low odds like in UT) because you can't catch the main draw pool because the draw pool keeps going up each year.
What I find interesting from your spreadsheet is how many people apparently "give up" with only 1 or 2 weighted points. Look at the "1 weighted point" level it has not had more than 123 applicants, despite there being 448 to 668 applicants in the "0 weighted point level"
Here is a specific example of how applicants dropped out:
2010: 719 applicants (3 regular points)
2011: 668 of that pool applied (0 weighted)
2012: 123 of that pool applied (1 weighted)
It happens repeatedly, here is another:
2010: 807 applicants (3 regular points)
2011: 656 of that pool applied (0 weighted)
2012: 121 of that pool applied (1 weighted)
I'm referencing your data in the spreadsheet and haven't verified it. If the data is good, it means that the vast majority of applicants that start out, give up around the 4th or 5th year.
A question for others, is there actually a "0 weighted points" level? I'm not sure on that one. Do you start at 0 weighted points or 1 weighted point in your 4th year applying?