Do that and you'll get about half of the useable chain life. What you're really measuring is a mixture of very little actual elongation and mostly roller wear. A bike chain is designed in English units of measure, with a true .500 inch pitch. When new, the space between any two rollers will measure .200-.205 inch. Using Campy's recommendation, you'd trash a chain when it's slightly more than half-worn. If you alternate the use of three chains, you can change the chain when Campy recommends, but don't toss it. After you've used three chains to the recommended "maximum" wear, you can reuse each chain until the wear measurement is about twice what Campy recommends.
Of course, you could also buy another 3 new chains and get the same amount of use from six chains rather than three.
While the method of rotating chains as you suggest seems to allow one to use 3 chains over a longer period of time, the end result appears to be that you save money in the total number of chains purchased over a specified mileage but at the expense of the certainty of using essentially worn out chains on the chainrings and cassette rendering those components useless once all 3 chains are worn out.
I realize that many posters, including you claim that you are getting upwards of 5,000 miles out of a chain. I would have to say that in my experience, there is no way in hell that I could have a useful service life for a chain of 5,000 miles.
The chain on my girlfriend's bike, on the other hand, still measures inside campy's specs after 4,000 miles. The key difference is that her power at threshold is close to 200 watts and her max sprint is maybe 700 watts, and that's for someone who weighs 130 lbs, so a lot less torque being applied than someone like me who weighs 160 lbs and sprints at a fair bit more than double her wattage. It's not even necessarily the sprints but the continued 1000 plus watt surges that regularly take place as well as lots of time spent above 400 watts that I believe leave my chain worn by 1800 miles.
My experience with my own chains is that once they hit about 1500 (1800 miles max), they are over the .004 elongation figure that I use (about 132.7) which I've adopted from campy's own 132.6 metric with a little added on. I generally measure with calipers, a new campy or CN 7900/6700 chain at approx 132.2 +/- .04 across 10 links. Using campy's 132.6 figure, that works out to about .3% elongation/wear/stretch, whatever you want to call it. I have found that I can push it to .4% and still shift adequately. My experience has shown that, if I forget or get lax about checking chain wear, I start to notice that my shifting becomes more sluggish and imprecise. Measuring at that point confirms my subjective performance observations every time. Take 10 links, and multiply the new figure by 1.004, (about 132.7 over 10 links) and I'm there.
If we assume that you are getting on avg 5,000 miles per chain, using 3 chains, and then have to replace chainrings as well as cassette, then your argument for rotating chains makes better sense than my method of going through approx 6 or so chains/yr. I certainly don't get more than 1.5 yrs useful life out of a set of chainrings/cassette, which would be about the length of time it would take me to log 15,000 miles (a little less time than that actually). In that scenario, you would be 3 chains ahead of me.
But an alternative to your method might be to buy and use 6 chains instead of 3. Theoretically, at the end of the same mileage and end use of 6 chains, you might not have to buy new chainrings or a cassette. DA7900 chainrings and cassette certainly cost more than 3 chains. Cheap FSA rings and 105 cassette, maybe not.
It seems to me that without being able to compare riding styles, conditions, and power outputs, it's difficult to compare apples to apples and dangerous to suggest to others that 5,000 miles or more is acceptable chain life. There are a lot of strong non racing, recreational riders who I could not see safely using a chain more than say, 2,500 miles-of course, that's my opinion.
What I can say that I observe, is that 9/10 people whose bikes I observe or check out, (outside of the more serious racer crowd) are riding around on bikes that shift like yak dung with worn out drivetrains. The degradation has just been so gradual that they no longer remember what a new drivetrain feels like. Funny thing is that they are either squeaking like a bird or grungy beyond belief, not much in between.
I've mentioned this before as another risk of maximizing chain life- someone on a ride just 2 weeks ago went flying over the bars and landed on his chin in a sprint. Cause was a broken chain. SRAM btw.
He lay on the ground for 10 minutes writhing in pain without moving because of the impact on his chin and chest. He's ok, but just something to remember.