The cleaner path has an effect for sure. My track varies from the 30s to present day stuff.
The new stuff with it's better pin connections and smooth tarnish free surfaces definitely makes a difference, requiring less wire drops, and having less voltage drop over distances. Both work and I'm not chomping at the bit for new; but I won't deny new is going to be better unless (maybe) I start soldering to improve or jump over the weaker older connections.
A diffence in metal quality, thickness, shape, the number of joints and path connections and quality all play into the puzzle to different degrees.
Here is one for you; pressure also cuts resistance. Same sq. inch contact will have less resistance with more pressure applied. (supprisingly less resistance. If you want to chase the math it's out there)
With less resistance you'll have less sparking as the intended path is more sound. The sparks occur during brief interuption or high resistance from wheel to track, which may fade once the intitial voltage jump of the connection is made if the connection can flow the amps. As amp draw rises the weak connections begin suffering again at some point creating heat based on resistance seen, which adds more resistance... it can be a viscious cycle at some point in connection deterioration.
Wheels moving on the track prevent heat build up there; it's issues are mostly because it does move around. But weak connections at track joints can be another story. E.g., with a heavy draw and some running time, you can spot weak connections with an IR thermometer or even a finger if it's very weak.(Ive had hollow O pins glowing red hot, careful of that finger )
(I expected less ohms on the smoke units 25-30Ω & didn't ever bother with the actual math; pure guesstimate. Thanks for that. Gilbert's approach to electrical is a bit unusual to me, it dives deep into theory early on for even simple things. I had to teach myself far beyond it (like working on TVs, coin op, and electronic controls) for years before I really began understanding the approach and lingo. It's learning approach may have been easier had I known nothing when I encountered it. It seems intent on making an engineer out you vs a knowledgable tech. Not bad; just more intense progression and assumption of student knowledge. I don't care for the technique myself and like my teacher lean towards Plato's advice; to teach well, the teacher must assume the student knows nothing)