Changes

The analysis was repeated but using only 1995 to 2017 inclusive (i.e., the modern era). The results were much more stable for N>=10. 20 out of 20 maximum R2 hull counts were also maximum adjusted R2 counts. N>=10 also seemed much less contentious, though N=11 (just under 50%) would have given the same result and N>=12 would have change only one result. Austin now maximized at 11 hulls - which is pretty much in the middle of its set. Boston was the same as before: 17 hulls with N=10 and 11 hulls with N>=11. Given the shape using N=11 or 12 might be preferred. New York now follows a similar pattern to Boston, and is 67 hulls with N>=10 and 27 hulls with N>=11. Again, the higher N result seems preferred. Palo Alto is now 18 hulls (out of 33 with 19 N>=10), which is a shame but hey. SF maxes out at 52 hulls, with N=12.  So, aside from the Palo Alto result, this is clearly a greatly preferred spec. I think N>=12 (just over half of the 23 years) is fair as a cut off. Also Portland, OR, maximizes at 4 hulls on both measures with N=15...