Quite a few early electrics were rod (and jack shaft) drive. The DD1 is notable for employing it at precisely the time Westinghouse was looking at competing drives, including quill drive, for the 11kV New Haven electrification. There were several ‘odd D’ prototypes built for PRR to test competing drives.
A major issue then was that lateral force from heavy motors mounted low in truck frames caused bad riding and even track damage. So Unusual as this may seem, keeping the motors relatively high up, and using larger drive wheels, improved the track-following characteristics profoundly.
Meanwhile, observe this picture of a pair of DD1 chassis:
The most efficient motors built for the desired hourly/instantaneous rating at that time were both very large and very heavy, and would not fit in the ‘gauge’ between drivers as some later motor designs would. To connect the motor to the tall drivers with gears would have involved expensive bearings, tooth metallurgy, lubrication, etc. — when a simple counterweighted and quartered jackshaft and rods did the business as well as on any contemporary D-class 4-4-0. Articulating two of these back to back gave a good ‘eight-coupled’ express locomotive.
An important issue with rods, though, is that while steam pressure is elastic, the inertia of a big rotating armature sure isn’t. So suspension action or buff/draft can produce much damage (see the Kando drive for one version of dealing with this).
Those wretched ‘20s L5s retained the deflicted features but threw out most of the characteristics that made the DD1 so good.