The bent move of the wide receiver would be a subset of [W?F?qR]. The q is essential here (and because F and R belong to different symmetries describes a 45-degree bent) to distinguish it from the 'delayed Griffon' [W?F?zR] with the zig-zag path. The moves you need would be fs[W?F?qR], since the N moves to the second square on the path (the first to break 4-fold symmetry) would be fsN.
With this system it would not be needed to assign (rather arbitrary) handedness to initial steps. So one could declare a z or q that appears before any bending as meaning both ways (s). That also has advantages, as in a notation with parentheses repeating the z or q step you would not have to prefix an extra s leg. E.g. the Crooked Bishop could be (az)7F instead of Fas(az)6F. (Of course zB is even better, but for more complex crooked paths the equivalent might not exist, or be ambiguous.)
The bent move of the wide receiver would be a subset of [W?F?qR]. The q is essential here (and because F and R belong to different symmetries describes a 45-degree bent) to distinguish it from the 'delayed Griffon' [W?F?zR] with the zig-zag path. The moves you need would be fs[W?F?qR], since the N moves to the second square on the path (the first to break 4-fold symmetry) would be fsN.
With this system it would not be needed to assign (rather arbitrary) handedness to initial steps. So one could declare a z or q that appears before any bending as meaning both ways (s). That also has advantages, as in a notation with parentheses repeating the z or q step you would not have to prefix an extra s leg. E.g. the Crooked Bishop could be (az)7F instead of Fas(az)6F. (Of course zB is even better, but for more complex crooked paths the equivalent might not exist, or be ambiguous.)