I wish I had known you were interested, and we'd have just ordered more parts. Other than the mounting location, the kits at Pinball Life will work the same.

Looking at the Stern docs, I'd agree that I'd expect to see about 2.7 ohms for your unwound coils. For your purpose, trying to make sure you don't blow anything up, using a low estimate would be better anyway.

The 20V that powers the coils is actually higher than 20V, though I don't remember what I measured. The 50V ones actually measured above 70V on my Tron!

Anyway, the rating on a TIP122 is 5A continuous / 8A pulsed. The slingshots are a pulsed operation. If the 20V was really 20V and your coil was really 2.7 ohms, you could calculate the instantaneous draw using:
V = I * R
20 = I * 2.7
I = 7.4A

If I were designing the circuit, I'd want more margin in there, and I'd prefer to use a TIP102, whose current ratings are 8A continuous / 15A pulsed.

I'm a "play it like it was designed to be played" guy myself, but in this case, the change is not permanent. If Dan sells the game, or gets tired of the mod, $20 gets it back to factory spec.

Of course 90% of the time I work on someone's game, incorrect coils are installed somewhere anyway.

Owners have had the game for years and never knew until I pointed it out.

If the transistors fail, they'll likely short, and the fuse will blow soon after. If you're set on leaving the coils juiced up, I think I would take my chances with the TIP122 rather than proactively changing to a TIP102. You would run the risk of heat damage by replacing it, and if it does fail, you're basically left with the same job.

BTW, I would highly suspect that the 20V IS actually higher. Both the 50V and 20V are generated using a transformer on line voltage, then running them through a bridge. They are not regulated. I have not measured it, but my estimate would be 28V based on calculations.