I don't know what "efficient" or "rate of payoff" is supposed to mean here. It seems to me that both are based on an assumption that no time spent not climbing can possibly be as effective as climbing. Obviously, Steph Davis doesn't believe this, and if this claim were true, then climbing might be the only athletic activity in the world for which it is true. In every other sport I know of, beginning to advanced participants have found it advantageous to spend some time in strength training targeted to the needs of the sport but not developed through simply practicing the sport.

Here's another way to think about it: if you get hurt and need to rehab, you'll be doing all kinds of supplementary training before you get back to climbing, at least if you ever want to recover. Climbing by itself is neither efficient nor does it offer a high payoff in this case.

The rehab analogy is neither far-fetched nor irrelevant. The advantage of supplementary training is control. First, control of the loads imposed on the body, and second, control of the progress of those loads. By contrast, training via climbing has neither form of control. Individual moves, most especially dynamic ones, will place highly variable and unpredictable loads on the body, and although one can go from easier to harder problems, the level of control over the resistances applied to the body isn't even remotely comparable to the fine-tuned ability to adjust supplementary training loads.

Training isn't an either-or situation in any other sport and I don't see any reason why it should be an either-or situation in climbing. Climbing training, involving technique, movement, strength, endurance, and coordination drills, supplemented by weight and/or gymnastic training that targets weaknesses relevant to either climbing moves or injury prevention makes more sense than trying to do it all through climbing. That approach doesn't work for other sports, you can't rehab injuries that way, and there is no reason I can see why climbing should be radically different.

Shotwell's sample of one is contradicted by the considerably more experienced Steph Davis sample of one, and in any case tells us nothing about what the best, safest, and most efficient approach might be, because we can't run Shotwell's wife through an alternate program to see how she would have done, and we can't run clones of her through the same experiences to see what injury potential she may have dodged by building strength via relatively uncontrollable means.

By the way, campusing does not meet the kind of control criteria I mentioned above, which makes it a relatively risky training procedure that is only appropriate for advanced climbers for whom the potential gains outweigh the injury risks.

Many years ago, I had some friendly arguments with a famous climber who insisted that the best way to climbing was just to climb. He developed into a fantastic climber, but I think the lack of supplementary training together with a steady diet of very strenuous climbing resulted in the shoulder instabilities that essentially ended his career.

Getting back to the topic at hand, there are controllable progressive pulling exercises Shannon could do that would double or triple her pullup ability. (Most definitely not the Armstrong program, however. Once she can do, say, six to ten pullups, she'd be far better off building higher pulling strength rather than shooting for pointless high repetition numbers.) Add to the pullups a few simple exercises to strengthen and protect the shoulder girdle and she will be a better and more injury-resistant climber with a better shot at a long career. I'd guess she could make significant progress in an hour a week (split into three 20-minute sessions). That is hardly going to take a big bite out of her climbing opportunities or decrease the "efficiency" of her climbing training.