If you look at many of the diagrams in "Shattered Sword" which look at the whole picture of the KB at Midway at various points during the day, rather than a photograph of a single isolated carrier, it is striking to note that the initial movement with regard to a single attack is for the whole fleet to make a similar turn in the same direction. However as the attack develops individual ships may have to continue a turn or reverse a turn according to its own individual circumstances. The result seems to be that it becomes even further spread out than normal with individual ships unable to support each other.
I came across this document, a product of someone's research using Japanese records showing their carrier formations at various times in 1942. There is one given for Operation C noted as a "cruising formation", which is evidenced by some photos.
Akagi leaving Celebes Island for the attack on Colombo, 26 March 1942. In the background are other carriers and battleships of the carrier striking force.
The Japanese strike force advancing to the Indian Ocean. Ships shown from left to right are: Akagi, Soryu, Hiryu, Hiei, Kirishima, Haruna, and Kongo.
However these relate to the early part of the Operation where the KB was leaving Staring Bay, Celebes (at 0700 26th March at 12 knots) and then heading south to negotiate Ombai Strait between the Lesser Sunda Islands and Timor. Captions on others taken at this time refer to the ships being "on the route to the Indian Ocean".
On the day of departure Soryu deployed 5 pairs of Kate TB in succession to scout the route to Ombai Strait and another 16 Vals and 4 Kates in 5 consecutive AS patrols. The next day it was Hiryu's turn with 5 Kate pairs in succession to scout the route and 17 Vals and 3 Kates on consecutive AS patrols.
See "The Darkest Hour Vol 1" by Michal A. Piegzik. This author has researched his books in Japan using Japanese records. I'm still on Vol 2 but haven't come across any further details of KB formations.
These seem to be tricky waters to negotiate with the Strait narrowing to about 26km. If it is anything like some of the other straits leading out through the Outer Malay Barrier to the IO, currents could be strong flowing out into the IO. Given the deployment of scouts on those two days would suggest that the Japanese were not entirely sure of what lay ahead of them from their charts.
The KB then turned south west to enter the IO. It was south of Bali by the 28th at which point it turned West before turning north west to head for Ceylon.
A few years back ISTR an internet debate about exactly what formation the KB used in the IO. I certainly wasn't aware at that time of the route they used to enter the IO and I'm not sure how many participants were. But it now seems to me that the formation chosen in those early stages makes sense in potentially restricted waters, with an arc of destroyers up front for AS cover and the main body in line with searoom to avoid any navigational hazards that might be encountered.
The unanswered question then and now is did they then revert to a more box like formation as used at PH & Midway once open ocean was reached. My guess would be that they did.
It is worthy of note however just how spread out these KB formations were and how few escorts there were to cover that large amount of ocean in between. That makes it difficult for one ship to support another when under air attack. It does however give each carrier a large amount of free airspace around it to launch, assemble and land its aircraft from. This was one of the things the IJN changed in 1944.
By the way, in June 1944 at Philippine Sea Taiho was recorded as travelling at 26 knots when torpedoed iimmediately following the launch of her big strike. (By implication so were Shokaku & Zuikaku in her group). She was able to maintain that speed after being torpedoed. She was still running at over 20 knots when the fatal internal explosion occurred some six and a half hours later.