1. I think the use of hover craft would be seen in the intercoastal traffic. That would be critical for the Americans, since MOST of the casualties (as opposed to British accounts which gets DRUMBEAT wrong.) is American shipping traffic from South America and American intercoastal traffic, NOT trans-atlantic routed stuff. The Americans really need that coastal kind of transport and it is less an effective British option for their own coastwise operations where air power negates it.
Hovercraft are no good for this either. Most coastal shipping is, and was, heavy, bulky goods that were not time-sensitive - coal and oil in particular. Hovercraft, which burn a lot more fuel and carry a lot less cargo, but carry it much faster, are hugely inefficient for carrying those sorts of goods. The correct solution was to establish convoys along the intercoastal routes. Even poorly (or even unescorted) convoys were much less likely to encounter a hunting U-boat than a constant stream of independent sailers.
2. The RN got these things wrong.
a. Hunter Killer groups.;
b. Convoy control and traffic management. Absolute chaos for the first year at least resulting in unbelievable port jam ups, delayed shippings and incompetent cargo flow management.
c. Communications. This was such a crypto catastrophe, that it alone in my personal opinion accounts for 60% of Doenitz's early success.
d.
Misunderstanding convoy mathematics. That is the most damning thing of 1939-1941.
Going through your points step-by-step:
a) There were two distinct phases to the British use of hunter-killer groups. The first, and the most often criticised, were the use of offensive patrols in the first few months of the war. This was always intended as a temporary measure, to cover the arrivals of independents that had departed before the establishment of the convoy system. Once the first two months of the war were over, the hunting groups were mostly disbanded. During these months, they sank three U-boats (one of which was in the defence of
Ark Royal). Meanwhile, there were only five losses to ships in convoy. Most of the losses suffered in this period were suffered by independents, and there was no easy way to protect them bar hunting. The second phase of hunting was in 1943 and later. This involved the use of hunting groups, supported by escort carriers and keyed in by Ultra decrypts, on offensive patrols against U-tankers and U-boats in transit in the Bay of Biscay. Support groups, meanwhile, were used on distant offensive patrols in the Mid-Atlantic, and could be detached to add to the escorts of beleagured convoys. Attacks around convoys may have killed more total U-boats, but the hunting groups were instrumental in breaking the U-boat force as a whole.
b) This was mostly the fault of German action, rather than RN incompetence. The German mine offensive in 1939 led to major delays and jams in British harbours, until countermeasures could be put into place. Similarly, in 1940, the Fall of France (and Norway), made moving trans-Atlantic shipping through the Channel and down Britain's east coast much more dangerous, due to the risk of air attack. This essentially closed London and Southampton to overseas shipping, and made it more challenging to use less threatened ports like Leith and Hull. By doing so, it ruined the RN's careful planning. Redoing everything from railway schedules to the allocations of longshoremen and stevedores led to the confusion. There's certainly an argument that the RN should have forseen the German use of mines, but the Fall of France was fundamentally unforseeable.
c) The RN should absolutely be criticised for the failures of its coding systems. The choice to use the less-secure book codes made sense; Britain could not produce enough Typex machines, nor enough trained Typex operators, to equip the vast numbers of British merchants. Even so, they should have been changed more often. That said, codebreaking was more important in 1942-3 than in 1939-40, as the Germans could not read Naval Cipher 1 messages in real-time, while they saw much more success with Naval Cipher 3.
d) There was no misunderstanding of convoy mathematics. The typical British convoy in 1918 had 30-40 merchants protected by 1-2 escorts. The typical convoy in 1939 looked much the same. While Rollo Appleyard did, in 1918, produce a mathematical argument for larger convoys, this was far from conclusive, and relied on a number of assumptions. Blackett's argument, which put the topic to sleep, was based on statistical analysis of convoy actions in 1942, and was much more conclusive.
3. As for the hunter killer group arguments? Later the USN tried this nonsense when there was enough air cover and escorts to go around, it turned out that it was more efficient to use convoys as "bait" to bring enemy subs to battle than to go roving for them. The RN admiralty, who had WW I experience to show this exact same thing happened TO THEM in 1917 before the USN discovered it in late 1943 should be triply damned for
e. not protecting what convoys they could by close escort with what they had.
f. risking their few fleet carriers as targets without understanding 2 at all.
g. not paying attention to in war immediate lessons learned when what they did at the start resulted in unexpected U-boat successes and RN disasters when e. and f. yielded numbers of merchantmen and flattops going down beyond what mathematical predictions showed should be the game theory loss results.
The hunting groups in 1943 were more effective than you suggest. In June-August 1943, the USN's hunter-killer CVE groups sank 15 U-boats. RN hunter groups in the Bay of Biscay, meanwhile, added three more, plus one shared with RAF aircraft. The RN's tally represented ~20% of the total sunk by surface ships in the period, while the USN's tally was a little under a quarter of those scored by aircraft. The
Black Swan class sloops, typically deployed on hunting operations, scored a total of 28 kills, more than any single class of ship bar the
'Flower's, impressive considering there were only 37
Black Swans (six of which served mainly in the Indian Ocean), compared to 294 'Flower's.
e) As noted above, the initial hunting groups were only set up as a temporary measure, to cover the transit of independent sailers which could not be protected any other way. They were only intended to be in place for the first two months of the war. No ships were sunk in convoy in September 1939, and only five in October 1939. Meanwhile, 44 independent sailers were sunk by U-boat torpedoes in September 1939, and 22 in October. Adding more escorts to the convoys would have prevented none of these sinkings. In this time-frame, ships on hunting patrols sank three U-boats, plus a fourth while covering a straggler from a convoy. Even after the end of the hunting groups, the main source of losses in the early period were still independent sailers, with only 9 ships being sunk in convoys between the start of October 1939 and the Fall of France.
f) The RN was well aware that these operations was a risk, but judged it one that it was willing to take. It was well aware that aircraft greatly extended the reach and effectiveness of anti-submarine operations. The independent sailers had to be covered until convoys could be formed, and hunting groups were the only way to do this. Putting carriers into the hunting groups would make them much more effective. Unfortunately, the RN did not have effective air ASW weaponry (and to a lesser extent, sensors) at the time. This meant that destroyers had to be detached from the carrier screen to hunt down submarine contacts. This left
Courageous vulnerable.
Ark Royal was similarly vulnerable because her destroyer screen was similarly too weak to be effective. Strengthening the hunting groups in the period when convoy was being set up might well have been more effective.
g) There were no immediate, unexpected successes, barring the loss of
Royal Oak. It wasn't until June-July 1940, after the Fall of France, that the mid-Atlantic convoys started to take heavy casualties.
Courageous' loss was understood as being the result of taking a necessary risk, while losses in 1939 and the first half of 1940 were manageable.