Posted By jonr on 11 Nov 2014 10:45 AM
As I said, "unlikely" - consider my numbers (based on a double jog) a near worse case example serving only to illustrate that "fraction of a percent" isn't always correct when you only know surface distance. On the other hand, as geodean's picture shows, unlikely, worse-case events can happen and it's often good to know what they are.
With two holes, deviation from straight is just as likely to improve performance. Ie, typical/likely case = 0% impact.
I'm pretty sure that Yonder has moved from concerned to interested in the theories.
The numbers are straight from the graph in the reference.
Now keep in mind that the numbers, formulas and algorithms the tables and software programs are using are derived from certain field measures. How did people know that the pipes did not deviated from straight when they originally established those performance numbers?