I agree that highly specific training is what is required to reach world-class level. But for amateur multisport athletes, there are a whole other host of constraints which exert themselves long before they will approach that level. In the case of triathlon, you have the logistical nightmare of training three sports, plus job, children and spouse. "Optimum" training, given that full context, looks a lot different from what a pro has to deal with... the amateur can't elect to take a nap or two extra whenever s/he begins overreaching. Now consider the heavy recovery toll running takes, even for those who don't manage to injure themselves. There has got to be a better way. Here is a real-word example of effective cross-training:
(1) In Oly Tri's, I've run 10k in 37-38 minutes a half dozen times
(2) The only solo 15k I've run in 59m flat
(3) The only solo half marathons I've run -both hilly- in less than 1h28m (even had a glass of champagne on one of those).
These run performances were achieved on:
(1) SWIM: 7.5hrs/week masters swimming
(2) BIKE: 4hrs/week cycling
(3) RUN: ... wait for it... a single moderately hard 6 mile run the week before each race
. Effectively, no running.
Note: the only organized sport I've ever played was 13-14 baseball... where I sucked.
You've asked me to explain WHY cross-training works. I too would like to know WHY a given training protocol/intervention works. I think we are both genuinely curious about "The WHY" of training.
Here's my first, least try at explaining "The WHY": cross training provokes central adaptations, even in quite fit people. Although these adaptations are insufficient to bring an athlete to world class level, they are powerful enough to get you well ahead of MOP. Improvement well beyond that point will eventually require specificity.
...Yet your reference then asserts "... the principles of specificity of training are likely to have greater significance" and "Both scientific evidence and anecdotal reports overwhelmingly indicate that the best way... " but makes no attempt to present or cite such evidence.
Sometime the evidence is self-evident.
I don't need the principal of specificity proved behind reasonable doubt simply because it is played out on very large scales on a daily basis. The trick is to prove that there are other ways to achieve the same result. The answer to date has been no. Yes the research is somewhat lacking but there could be a reason for this ie: training said activity gets you better in that activity is kinda hard to argue, and really, why would you want to?
So it's not to me to disprove crosstraining rather for you to prove WHY it works (or doesn't) as if it were self evident that crosstraining works then more cyclists would be running marathons and vice versa.
Aside from injury, rehab, posterial correction etc I have seen no evidence that cross training is anything other than the usual "fitness industy" mumbo jumbo. Right up there with "core workouts", and "protein shakes".
Simplicity and specificity should be embraced on a more regular basis.