Difficulty Index scores are very useful...

[derricknehrenberg]

...But it appears this score varies quite a bit by #points.  When #of points is roughly normalized across a range of rides, the rank of DI scores fit my rank of perceived efforts on a given list of rides.  When they aren't, the scores do not match perceived effort. 

Has anyone noticed this anomaly?

Also, it seems it could be useful to include elevation in the calculation.

d

[ScottMorris]

Correct, it is sensitive to the number of points.  For a true comparison always use the same GPS with same settings for tracklog recording.

Elevation data is in the calculation.. fairly directly.  You'll see more total elevation gain for tracks with higher point density as well... (since it is just a sum, it is prone to error, so the more observations you have, the more error...)

Cool to hear that it matches up with reality for you.  Great feedback!