Your thoughts can be heard: "But what about the engine? Why not overhead valves?" By 1919, Indian Powerplus-based side-valve (flathead or L-head) racers were out-speeding both Indian and Harley-Davidson eight-valve overheads. The Harley eight-valves were the more advanced of the two competing overheads, since they benefited from World War I aviation engineering and thus featured hemispherical combustion chambers and inlet and exhaust valve pairs spread 90 degrees apart. Yet the deceptively simple side-valve Indian racers had more oomph. It was a brief and quirky time in the history of engine development before the advent of tetraethyl lead additives to gasoline (petrol). In opting for flathead engine layout, Indian was simply choosing the engine form that dominated the automobile industry and that on both sides of the Atlantic was winning the premier races.
You ask: "So what about tetraethyl lead? How did its absence promote the flathead engine layout in both motorcycles and cars?" The answer is that in the pre-tetraethyl lead era, fuel simply burned too fast.
Since the beginning of the internal combustion engine field in the late-nineteenth century, it had been appreciated that power was maximized by keeping the initial combustion volume small in relation to the volume opened up by the descending piston (the swept volume). Turning this statement upside-down, power benefited by increasing the ratio of the swept volume to the initial (unswept) volume. With slight modification, this idea becomes the compression ratio, or the ratio of the sum of the swept and unswept volumes divided by the unswept volume.
For decades, theoreticians had forecasted cylinder heads with both inlet and exhaust valves upstairs, and with compact combustion chambers that produced high compression ratios. Separating theory from practice was the lack of a fuel that would burn smoothly in a compact overhead-valve cylinder head. Explosion is simply very rapid burning. What happens in a combustion chamber is analogous to a forest fire. Once a forest fire becomes sufficiently hot, fire will break out ahead of the flame front, and trees untouched by the advancing fire will nevertheless burst into flames.
The air itself has become so hot that it ignites all ahead of it. Likewise, in a combustion chamber there is an expanding sphere of explosive burning. As in a forest fire, this is termed the flame front. Outside or ahead of the combustion chamber flame front, the as-yet-unburned fuel/air mixture is being rapidly compressed, and with this rapid compression comes a rapid temperature climb, a heat "spike." Thus, "forest fires" start ahead of the flame front, and the one big explosion starts bumping into one or more smaller explosions. In common terms, this uncontrolled explosive burning is variously called "knock," "detonation," or "pre-ignition," according to sub-level details that aren't important enough to this story to explain.
What is important is that without a suitable fuel a typical road-going overhead-valve engine of 1919 produced only a little more power than the
counterpart flathead engine. Both engine types were limited in the sense that neither operated reliably (no knocking) with more than about a six-to-one compression ratio. In racing applications, as proven by Indian, the complexities of flow dynamics could completely level the playing field or even work to the advantage of the flathead.
[page 6]