You can't get consisten readings using pastigage on the sides of bearings. The bearing bore is not a perfect circle. It is oval and bells out at the joint of the bearing shells to lessen the chance of snagging or drag at the critical bearing joint. If you check the big end of a rod, for example, using an inside mic or bore gage, you can see how much wider it is from joint to joint compared to center of bearing shell top to bottom.

As for the colored bearing scheme, the japanese engineering philosophy is to create equal clearances on all bearing surfaces. The idea, or at least I have read, is to improve harmonics and balance at high rpm and to spread and equalize friction loads under extreme rotation. The only way to accomplish this is to have every crank journal and bearing bores identical in size (not practical) or a range of bearing sizes to mix and match to ensure equal clearance across the board.
Keep in mind Honda engines will run 200,000 miles as they come from the factory, even under frequent hard driving. How many guys slapping in std bearings in their garage see nearly that much life out of their bearings?
It is common to see people challenge what the factory does. Sometimes people come up with a better way of doing things, which is the heart of performance tuning.

That said, a build I am working on now just got std sized ACL bearings on rods and mains. This is because of four cranks I had, only one had journals consistent enough in size to keep them all within a couple 10,000ths of each other with the ACLs. These bearings were sent out for dry film lube coating and further massaged with scotchbrite to bring them down to the exact tolerances I was looking for. This engine goes together with every bearing within 1 10,000th of an inch of each other.
It's just a matter of whether you are a meticulous builder or a what-the-Hell type. When the engine is going to see hard use or maximum performance is the goal, I go meticulous.

If I had to worry about the pocketbook or keeping the parts guy happy, I might come up with solutions that involved compromise as well, so I am not being critical, just saying my preference is for precision when possible. Throwing engines together with standard bearings was something myself and buddies always did with V8s. We got a wake up call rebuilding a Buick V6 when we discovered it not only called for closer tolerances, it demanded them. When I got into Hondas, it was crazy critical. There is a reason these little four bangers are so efficient and capable and close tolerances is one reason.

I don't question Honda engineers. They spend the time and money and see it all in torture testing and teardown. Costs and meeting regulations are limiting factors for car makers and whenever you see a maker go an extreme on something like bearing sizes and tolerances, you can bet your ass they are doing it for a reason.

Stocking and engineering bearings in a wide range of sizes is more expensive than just cranking out std bearings for a one-size fits all solution, so I have to figure there is a good reason for it. I put my trust in Honda engineers. Their solid work is why were are all here doing what we do.

It bears repeating: When it comes to the bearing scheme Japanese makers use, they opted for the more complicated, costly way of putting their motors together and that tells you something. Ignore it at your own risk. Lots of guys talk about their workarounds and shortcuts and how they suffered no ill effects. Many times it goes something like: that motor ran for another ......... you fill in the blank but that number usually is nowhere near what it should have been when you think of what marvelous pieces of engineering these Honda engines are - at least when put together they way their creators intended.