Actually, it is rather controversial. Normally, when you add heat to a large hydrocarbon molecule, it cracks (breaks down into smaller pieces plus more heat). This is why hydrocarbons can catch fire and burn. Going the other direction -- from large molecules to even larger ones -- requires very specific heat ranges and very delicate conditions involving unusual catalysts. Add too much heat, and it cracks faster than it merges. Add too little heat, and it cracks instead of merging (if anything changes at all).

There are indeed specific conditions under which merging occurs, but in geology they are extremely rare. If abiogenic oil theory were true, it would predict abundant oil in Washington state (close, but not too close, to vulcanism) and zero oil in Texas (where there are no volcanoes). The facts are, of course, otherwise. The amount of naturally-occurring abiogenic oil is quite possibly non-zero, but clearly accounts for at most a tiny fraction of all naturally-occurring oil.

Even if abiogenic oil theory were true, we would still be experiencing peak oil. Assuming it were true, the earth would have been producing all our current global oil reserves over the course of billions of years, and our current rate of extraction would be depleting the reserves faster than abiogenic production could refill them. (Two arguments why abiogenic oil must be slow, even if it's significant: one, the annual rate of carbonate rock subduction is not sufficient to sustain the current annual rate of oil extraction, even with 100% conversion; two, the oil produced would have to go somewhere, e.g. break to the surface and feed an ecosystem of oil-digesting bacteria, if it were being produced at such a rapid rate.)

The theory is predictive. It can be used with 67% accuracy to find new oil deposits. That is both falsifiable and HUGELY practical. The bulk of expense in oil extraction is finding the oil in the first place, which is why the Middle east is so attractive, it's all at the surface, shallow drilling and easy testing.

If this theory pans out, and as he says that oil is distributed all over the world; it means that oil gets MUCH cheaper and unstable regions can go F@#$ themselves, we'll just drill somewhere else.

Where did you get that 67% number fromß