Thanks for the kind words,
Avindian, but I think it's time to wind this one down. Half the Battle of the Atlantic (against the British) is won but pointlessly so, and the other half (against the Americans) is more or less impossible in the game as it currently stands. Even a comprehensive examination of changes to the game that might help is rather futile, as For the Motherland and Dies Irae: Götterdämmerung are already slated to overhaul U-boats.
I regret entering the competition. I believe this AAR would have turned out better and been easier to write had I done it on my own terms and not cut it to fit the fickle contest.
Furthermore, a strategic bombing campaign has nothing to do with Dönitz or U-boats and seeing it through would add little to the bigger picture. So with these things in mind, I'm going to close with a handful of ideas for refining the HoI3 submarine game (bearing in mind, of course, that FTM/DI:G will do other things and render all this irrelevant and obsolete). These have been rigorously selected for simplicity, abstraction, and ease of comprehension.
1. A dud rate which multiplies submarine Sea Attack and Convoy Attack by a number between 1 and nearly 0. This number is primarily influenced by national level of submarine practical, but the introduction of a new torpedo technology will cause a brief spike in duds. Historically, dud torpedoes were a huge problem for the Germans and the Americans alike, with torpedoes being too complicated and expensive and dangerous to test rigorously in peacetime (even, in Germany's case, when planning to go to war). Angry submarine skippers fought with their navies over the effectiveness of the torpedoes and only got defects corrected well into the war. Although for HoI3 purposes it makes sense to lump all these defects under the word "dud," among the most critical cases were run-unders (torpedoes running below their set depth and missing their targets) and guidance failures (torpedoes failing to stay on course, sometimes even circling around and hitting the sub that fired them). More classic dud situations in which the explosive failed to trigger were also a factor; magnetic triggering was unreliable the world over, and for the Americans, even the impact device was a mess, with specific design problems that disinclined it to explode from direct, 90-degree impacts (which were supposed to be the most reliable attack setup). Given the mechanical unreliability of early-war submarines, spontaneous breakdown damage should also be a possibility, perhaps tied to a rare low-practical-linked event and representing only the worst catastrophes (although total failures of American HOR engines were routine; they tore the teeth off their own gears).
2. An increase to submarine Sea Attack. A general increase has been justified previously in this AAR. More specifically, a Magnetic Exploder yes/no technology ought to dramatically boost Sea Attack. Magnetic torpedo triggers were developed specifically to allow a submarine-fired torpedo to bypass the waterline armor belt present on warships and explode under the fragile, less-protected keel. The precise goal of this project was to make submarines a more viable anti-warship weapon. It was, however, more difficult than expected and should be a challenging research item whose accomplishment reflects not the earliest implementations (which were present at the start of the war but ineffective), but a fully operational magnetic exploder that really does make torpedoes more effective against armored vessels. (It should still spike the dud rate as normal to reflect the usual teething problems of a new piece of torpedo technology.)
3. Submarines should suffer a visibility penalty in coastal waters, and possibly an attack penalty too. This reflects both the dangers of a coastal shelf, which limits submarine submerge depth, and the potential for refloating operations to recover ships lost to submarine attacks in shallow waters.
4. Submarine ranges should be extended, either just by bumping the numbers upward (to about double what they are now) or by throwing in a U-tanker/sub-tender yes/no technology. Such a technology should be easily developed, retrofittable, cheap, and non-secret. Any flotilla with "U-tanker/sub-tender: yes" has its range doubled from stock and its fuel requirements increased slightly. This makes Operation Drumbeat possible (and probably adds a measure of sanity to American submarine activity in the Pacific, but I don't know firsthand what that's like now, so I can't say for sure).
5. I'd like to see more recognition of the importance of radio policies to submarine effectiveness. Possibly there could be doctrines or even doctrine paths that range between seriously compromising Visibility and Convoy Attack in exchange for detailed scouting reports and seriously compromising intelligence-gathering in exchange for serious radio silence and deeply obfuscated radio orders (possibly incorporating an order delay that makes submarines slower to respond). Historical navies were so concerned about these issues that they had, for example, specific policies designed to prevent the enemy from guessing anything about a transmission based on its word lengths. Submarine effectiveness, particularly against convoys, has a major informational component; the Germans didn't even realize how hamstrung they were by Allied radio direction-finding technology, but it hugely cut convoy losses. A submarine would make a contact report before attacking, and just like that, the convoy would turn away and avoid contact.
6. An explicit understanding of a handful of submarine abstractions. There should be official designations of tonnage, if not number of ships (which would have varied historically), per "merchant flotilla," and convoy attack popups should list kills in tons. It should be assumed that any given submarine flotilla in the ocean represents about half of the unit's strength, with the other half cycling to and from port (submarine patrols never lasted longer than three months). This might justify a reduction in stats, but should also prevent a submarine flotilla from being wiped out in a single naval battle, as half the boats in it are simply not there.
7. A secret project to create a faster, more-time-submerged submarine engine that runs on hydrogen peroxide or oxygen fuel. This would be a yes/no with a considerable speed boost and modest improvements to attack and visibility, possibly at the cost of an even more fragile boat (due to volatile fuels). The Germans actually had a working hydrogen peroxide prototype; submerged, it ran half again as fast as a comparable diesel/electric and did not have battery limitations crippling its ability to move underwater, gains which would have allowed it to catch up to most fast convoys and task forces while remaining concealed. Obviously the rollout of hydrogen peroxide or oxygen boats should be accompanied by a hearty pulse of mechanical failure event chance. Given the fundamental differences in design, this technology should be baked into the hull, so that diesel/electrics cannot be retrofitted for hydrogen peroxide or oxygen. Engine upgrades should still function as normal, representing refinement of mechanical engineering rather than different methods of power generation or different types of fuel.
8. A "crew kill" doctrine which has additional effects on enemy manpower from ship sinkings, but carries a steep diplomatic disadvantage. Various navies flirted with the idea of deliberately exterminating the crews of lost enemy ships as a matter of policy, and although none adopted it, they could have. The Americans massacred Japanese crews from time to time. Certain British orders related to capturing intelligence information (Enigma machines, secret papers) from U-boats actually did specifically order sailors to fire on surrendered crews in some circumstances. By the same token, there might be room for rigid rescue doctrines that placed submarines in greater peril but had diplomatic advantages. Dönitz twice ordered U-boats to stop assisting distressed Allied crews at risk to themselves, yet still they sometimes did. One captain actually tried to organize a local truce and joint rescue operation off the coast of Africa (which didn't work). The catch is that these efforts didn't really do that much diplomatically for Germany; the British were established masters of international wartime propaganda and could paint every U-boat action as an atrocity regardless.
9. Convoy loopholes in general should be closed. If lend-lease teleports supplies across the ocean, it shouldn't do that. If a successful convoy-raiding campaign can't contribute to the surrender of a blockaded nation, it ought to be able to. If submarines can't attack neutral-flagged ships carrying war materiel (everything HoI3 models) to the enemy, that absolutely needs to be fixed, and not just for Unrestricted Submarine Warfare. Even the strictest visit-and-search rules permit action against neutrals carrying contraband. That's the whole reason they board and inspect neutrals. Also, licensing technologies should be harder at best if the countries involved have no open lines of trade. It's all very well for the UK to license production of American tank destroyers, but if nothing Allied can make it across the Atlantic, how exactly is that going to work? You can't telegraph blueprints and specimen parts and engineering experts. It would shift from a routine transport operation to a daring (and risky) feat of espionage.
10. It would be nice if submarines had a "support partisans" mission based on their historical roles doing so. American submarines in particular supplied, reinforced, and relieved guerrillas behind Japanese lines. They might also benefit from an "espionage" mission to support spy placement, reflecting the many times German submarines attempted to land spies in Allied territory, but this is less essential as the spies were typically caught immediately. Then again, perhaps with better radio security...
11. I'd like to see better presentation of information related to convoy raiding, perhaps a ledger page which lists convoy sinkings (as is already done for warships). The naval map should pip ports which have lost convoys recently and display further information on mouseover (e.g., "Dover has lost 2 merchant flotillas this month: Dover-Halifax, Boston-Dover").
12. You should be able to find out information about convoys through espionage. This would be among a historical Germany's most urgent espionage priorities. It would also have helped American efforts immensely, revealing a concentration of Japanese shipping routes passing through the Luzon Strait. Intelligence services should provide their best guesses at enemy convoy routes on the naval map, perhaps as dashed lines to distinguish them from your own.
13. Convoy routes should attempt to change to avoid known dangerous sea zones, leading to a war of maneuver on the high seas. Obviously this would have an impact on time taken, supplies transported, fuel used, ships required, etc. This is perhaps the HoI3 equivalent of the cost of convoying, which as a rule of thumb was contemporaneously estimated at a third again the burden of peacetime transport.
14. I am not sure whether ASW is potent enough. ASW aircraft, as technology develops, should be a pretty darn powerful force, to the extent that a submarine power ought to try to deploy into gaps in air coverage (historically, Germans in the mid-Atlantic).
15. Harsh weather should have brutal effects on submarines, particularly navigationally. With poorer surfaced deck stability, and not being surfaced all the time, star-and-compass navigation was difficult for subs in extreme conditions. In practice, it wasn't clear to me that weather was making much of a difference one way or the other.
16. Some of the torpedo tech names could stand to be tweaked. The name of an acoustic torpedo should not be used for a non-acoustic upgrade.
U-11, a Type IIB retired as a museum ship. In this game, it was probably one of Assmann's during his early-war rampage at the mouth of the Thames. My thanks to Soviet Germany for graciously exhibiting it postwar.
I have no current plans for future AARs, but I have been thinking tentatively about an unorthodox defense of France, both historically precedented and unlike any existing AAR I am aware of (no, it does not involve submarines).
PS -
Laurwin, if fuel was simply being transported between owned provinces, it wouldn't be leaving the national stockpile. And it couldn't have been keeping a naval convoy running, because I had manually disabled all naval convoys. Stettin was the last.