Looking at the maps on the weekend, I had D chosen as my route - NW out of the control past the building to the trail, then towards A.
 

This one surprises me. I ran C, but that was before I realized that the flag was on the west side of the library entrance. I figured A would have been faster by a lot more. I also like mrpither's D choice, too.
 

Its confusing to pick the fastest way.
 

I am quite sure (D) is faster than (A) at least. So if the other three are equivalent, (D) must be the fastest. I would have thought (A), (B) and (C) to be about 5-10 seconds slower.
 

I would have run D.
 

D for sure
 

I didn't test D when I tested routes. I planned to, but I was running a few minutes late and had to meet Mary and the start/finish.

Michael
 

Consider the gap near the "a".

I don't consider this risky if you are running from S->N. You take a quick look on the approach, and if it seems iffy you run the extra 20 meters to the stairs without breaking stride. Cost is 4-5 seconds.

It's a lot riskier if you are going N->S. Now you commit to the gap earlier. If it doesn't work, you lose a couple seconds trying to find the gap. Then you have to run an extra 40 meters around the wall with 2 direction reversals that cost speed. Could cost up to 15 seconds.

My evaluation. I don't even consider the gap if I'm going N->S, but going S->N I keep my eye open for an opportunity to shave seconds.
 

My first choice is a small variation of MrPither´s D route. I think I would have gone straight down to the road/bike-path and then to 'a'. Possibly using the stairs - with the same thoughts as blegg - if the 'gap' looked risky.
This is really a 'micro route choice' leg that may not differ more than a few seconds but a sprint course offers a lot of those. One of the most important things in sprint is consistent orienteering through fast (and correct) decisions and I tend to try to simplify a lot even if that means that I sometimes run a bit longer. I´m not very succesful, so that may be the wrong tactics...