The RATE project though does provide SOME evidence for a young-earth. The piece I am the most familiar with is the helium diffusion in zircon prediction - this was a blind study that resulted in an accurate prediction matching a 6,000 year old earth with accelerated decay at or near the beginning of that timespan. Here's a link to a more recent article Dr. Humphreys discussing/rebutting some of the criticisms of this experiment
Note that there's precious little discussion and no rebuttal. See below.
The biggest detractor to this experiment seems to be Dr. Gary H. Loechelt, whose most recent critical review of this experiment (that I could find) is here. Reading through Dr. Loechelt's response it appears he validly shows how the RATE experiment results could be off - but the percentage is miniscule. From my understanding we're talking 30% of the surface helium (which is 1-2%) remaining - so the results may be off by less than 1/2 a percent - well within tolerable error limits, and nowhere near the hundred thousand orders of magnitude required for the data to fit an old-earth model. I admit I'm not following the second half of his article very well - so perhaps I'm missing something.
Sorry, you're missing something. Dr. Loecholt demonstrated that correct use of mainstream diffusion theory, the real temperature history of Fenton Hill, and the RATE group's data, that the inescapable conclusion is an old Earth. See section 3.5 and 4 of the technical paper listed above. See also the thread linked to above, where he derives the same conclusion from data published by Reiners in 2005.
In addition, in re Humphreys' "rebuttal", I've checked the primary sources and there's no question that Humphreys (or whoever actually did it) read the temperature graphs in his references backward. Humphreys thinks that Fenton Hill was hotter in the immediate past than it is today, whereas it really was cooler. If anything Dr. Loechelt understated the case for Humphreys' making an error. I'm astonished, and I don't have a high opinion of YECs to start with. Incredibly enough this blunder doesn't affect Humphreys' analysis, but it does highlight the amateurish nature of the RATE project and the inadequacy (to say the least) of the "peer review".
The major issue is whether one extrapolates the volume diffusion line to the temperatures of interest (as Loechelt does) or extrapolates the defect diffusion line to the temperatures of interest. Loechelt makes a good case based on references to noble gas diffusion studies from the 1960s up to 2005. Humphreys makes no case at all, he just assumes his method is correct.
Is the RATE project's conclusion of accelerated decay a reasonable one?
Not a chance. See, for example, the thread linked to above. I'm somewhat proud of digging up the dose data showing that the proposed accelerated decay would have killed Noah and the gang from radiation poisoning from the
40K in their bodies. And another poster pointed out that the uranium decay series would not be in secular equilibriium anywhere, which it is in many places. The list of show-stopping problems with accelerated decay is virtually endless. See, for example,
The Constancy of Constants and
The Constancy of Constants, Part 2.