It is because the transmission line carries the wires which give the majority of the load (for normal height towers) unlike EIA communication towers which just have a single instance.
Back in the old days of hand calculations and running Alcoa's Sag10 on a mainframe, it was too laborious to consider an oblique wind so the wind direction was always normal to the wires because it was assumed that this produced the largest leg loads. With software like PLS-CADD, it is possible to have wind in many oblique angles and study their effects on the leg loads. The NESC makes a reference to oblique wind angles in rule 252, IIRC.
Square lattice towers have no problems with oblique winds, but highly rectangular towers can have huge problems (on paper anyway). I have studied some of our towers with a lot of redundants on the wide face where the wind normal to the wires produce a leg load of about 100% capacity. When an oblique wind direction was considered the leg loads went to 150% because the wind on the wide face was so much it overcame the loss of load on the wires due to the direction being skewed.
_____________________________________
I have been called "A storehouse of worthless information" many times.
