I noticed some variance between coverage runs. Looking at the nyc github repo, they recommend enabling caching to only instrument the code once for better results and more performant testing.
Coverage remained the same at 93.578% when pulling a03273ea4de03bcc23a2ed19a389419f4516c5e9 on probablyup:es-enable-nyc-caching into ded45451c856ecd05ed4fc2cf4f6044bd916c323 on iamstarkov:master.
I noticed some variance between coverage runs. Looking at the nyc github repo, they recommend enabling caching to only instrument the code once for better results and more performant testing.