In the OP,
@hkrommel said that the two main problems with the naval game were (1) the presence of
Doomstacks and (2) the absence of the
Commerce War. I broadly agree with the points made there. In my view, there are three more major gameplay problems:
(3) The excess number of
Inconsequential Alerts. If you're playing as a naval power, you are deluged by endless reports of inconclusive convoy battles. I usually play on speed 1 and 2 in mid-game, and even then I can't keep up.
(4) The shortage of
Interesting Decisions. You assign your ships to fleets and your fleets to strategic regions and that's it. Once a naval battle has started, almost the only thing you can do is sit and watch - and since they last for days or weeks, that's not something you can afford to do if you're also fighting on land (i.e. as all of the most-played countries). You can send further fleets to intervene, but of course that's the thing that was almost impossible to do IRL: naval commanders could very rarely radio for further reinforcements from a home port.
(5)
Convoy Routing is very pretty, but players have no control over it. Not only does this remove one of the major strategic decisions of the naval war, but it also means you tend to end up with the Battle of Biscay Bay, not the Battle of the Atlantic, because that's where all the convoys go. In addition, the game does not store data on the strength of individual convoy ships. I have no problem with aggregation (as you will see), but that means that attacks by NAVs and SS are either inconsequential or devastating, because there's no room for 'damaged but not sunk'.
To fix all of these problems, I think naval warfare needs to be divided into two levels: a Corbettian level using an aggregated totals (like the air war) and a Mahanian level with individual ships (like current naval warfare, but more so). Convoy Routing (5) stands more on its own and will be addressed in a third part.
Corbettian Control of the Sea Should Be Aggregated and Interactive
One level of naval warfare, representing the Commerce War (2) and control of the seas, should take place at a strategic region level and you should interact with it through an expanded version of the air warfare screen. Here is a mock-up:
As at present, fleets would be assigned to strategic regions. If you want to, you can see the 3D models sailing around that region. But they wouldn't just be waiting for a big Battle (capital B) to start. Just like as aircraft do, the game would be carrying out background calculations for small battles (small b) all the time without giving you alerts. In fact, it might well make sense to simply integrate the Commerce War into the air warfare code. A Naval Bomber conducting Naval Strike might find itself in battle with a Light Fighter on an Air Superiority mission, or it might find itself in battle with a CL on a Protect Convoy mission or a Convoy on a Transport Troops mission.
All the fleets present in the region would be aggregated, just as air wings are. In most cases, the game would calculate the result of the battle and add it to the totals seen on the above screen. You wouldn't normally get an alert. But if the battle had serious consequences (a division is destroyed; an island no longer has enough supply), either by itself or as a tipping point, then the existing alerts should usually pick it up.
Sea control shouldn't be a simple yes/no, but a relative concept: an aggregated number, which would indicate the probability of naval invasions and troop transports getting through, and would also control the efficiency of relevant Convoys on Import/Export/Lend-Lease missions. If you have more ships than the enemy, and you have won more recent battles, then your sea control goes up. If all your ships are in Doomstacks (1), then you're going to cede sea control of many other regions, and that is going to hurt.
While the naval war could use the same functions as the air war, there would be some important differences in the maths. In the naval war, the number of battles is much smaller. One Light Fighter has a good chance of encountering another Light Fighter when there are 1,000 enemy planes in the air. When the enemy only has 40 SS chasing 50 Convoys, there will be far few interactions, and detection numbers will reflect this. But, conversely, the size of the battles will be much bigger. I think the air battles are limited to 1v2 (for the air element of naval battles) and 1v3 (for purely air battles). Naval warfare would need bigger numbers, and
the size of battle should vary according to the types of mission involved and the naval doctrines chosen: so Convoys on an Import mission with a Big Convoys doctrine v. Subs with an Intercept Convoys mission with a Wolf Pack doctrine might be 30 (20 Convoys, 9 DDs, 1 LC) v 20 (all SS). But the player won't see any of this directly: it will all be aggregated, just as with the air war.
Note also that there would be an
overall detection number, just as in the air war, so detection would be far more aggregated than at present. (EDIT: Others have cross-posted about the importance of signals intelligence and naval radar. This was all about building up naval commanders' overall intelligence picture. You should be able to boost the Strategic-Regional detection number by having an encryption advantage, by fielding ships with radar sets and perhaps building Radio Listening Posts nearby).
However, the player should be given Interesting Decisions. I've already suggested that doctrines should make a big difference. Players who buy the naval DLC (since this patch needs to be paid for!) should be able to use Command Power to add more detailed strategies, just as you can Add Extra Ground Crews to air warfare. These might include 'Avoid Capital Ship Combat', 'Deploy DD as SS Tenders' (to help subs to rebuild strength more quickly), instructing crews to 'Protect Convoys at all Costs', or choosing between 'Close Blockade' and 'Intercept at Sea'. One important option would be a 'Fleet in Being' strategy, available only in strategic regions where you have a friendly port. Fleets with this setting wouldn't be involved in the day to day battles, but they would contribute (at a reduced level) towards your sea control score, and they would be involved in the big set-piece Battles.
Mahanian Battles Should Be Rare But Engaging
We have said that the size of battle will vary depending on the number of aircraft and ships involved. If the normal calculations throw up an encounter between sizes of sufficient strength (probably anything involving capital ships, as well as large convoy battles), then and only then is the Battle simulated in more detail and in a specific province.
Such major, Mahanian Battles should be rare, but will have a major influence on the outcome of the naval war. As the UK or USA, you might see this once a month during the peak of the Pacific War; if you play the USSR historically, you might never see it.
At the very least, such Battles should throw up an alert. In fact, it would probably make sense for players to have the option for the game to automatically pop up the minimap and slow the game down to speed 1. That's unhelpful for high-speed MP or people playing land-locked Switzerland, but if you're a major naval power in SP, you ought to be paying close attention when your equivalent of Midway or D-Day happens. And even in MP, the fact that Japan may or may not lose most of its CVs in the next ten game hours is surely going to affect
everybody's strategy. It's going to be worth 2 minutes of everybody's (real-life) time, with every player watching on the edge of their seats! I know that PDX has historically avoided tactical minigames (outside of
Svea Rike), but actually, the current (1.5) naval war is already based on a minigame display. It's just a frustrating one because you can't really interact with it, an unrealistic one because the Battles drag on for weeks, and an unappealing one because individual Battles rarely change your overall strategy.
Rare but engaging Battles would be great for generating the stories that make PDX games great.
To be engaging, these major Battles would function the same as current Battles, but with three changes.
Firstly,
the fleets involved and the timing and size of any reinforcements are set at the start of the Battle based on what's available in the strategic region. Given the size of HoI4 strategic regions, you're not going to be able to rush in any reinforcements in the short period of time (half a day to a couple of days) that a 1940s naval Battle takes. So the game will generate & store the reinforcements at the start of the Battle ("in the fourth hour, add 4 DD I and 172 NAV IIs"), but the players and AI will only see them added to the Battle at the appropriate time.
Secondly,
the location and purpose of the Battles should matter more than it does right now. So when a Battle starts, the game will note the purpose of the fleets involved and (probabilistically) assign a particular province within the Strategic Region to it. The number of land-based aircraft that are involved in the Battle should depend on the precise location (so range really matters). The purpose of the respective fleets should affect how they behave. If you have a Convoy on a Troop Transport mission about to land in enemy/your own territory, then the fleets on Convoy Protection and Convoy Interception respectively are going to engage aggressively. If it's core territory, then they will tend to go for all-out attack on the convoys. If you have a Convoy on a Import mission, then they are more likely to just run away. There are currently four different missions, which gives 16 combinations: that's enough to program a discrete AI aggressive setting for each. I think Convoys should also be considered as having as missions (Troop Transport/Import/Lend-Lease/Supply), and as stated would add some modifiers based on location, but I still think that there are sufficiently few combinations that the AI programmers could pre-assign sensible strategies (DDs attack aggresively; SS attack underwater; SS attack above water; Convoys flee, etc.) to each discrete possibility in a lookup table.
Thirdly,
if you've bought the naval DLC, then you should be able to use Command Power to issue such tactical instructions (probably the same ones issued at the aggregated level and used by the discrete AI, such as 'Avoid Capital Ship Combat' or 'Convoys Flee') based on the situation in this particular Battle. If your Grand Fleet is facing an unprotected convoy, you can be more aggressive. If the enemy's 5 carrier fleet then appears, you can order 'All ships flee' or sacrifice some DDs in an aggressive attack. So you won't just be watching the battle on the edge of your seat, but you'll be able to engage and influence the outcome (within the resources and reinforcements predetermined at the start).
Convoy Routing Should Be Based on Larger, Nodal Sea Regions
My third suggestion is a more tentative one to deal with the issues in Convoy Routing (5) and the other aspects can be implemented without it.
At the moment, convoys are set by the game, which removes Interesting Decisions (4) from the player (should I take the risky route through the Med or the safe route around Africa?). But I realize why this has been done: path-finding is extremely computationally intensive and so it's very difficult for the AI to make decisions between different alternatives without slowing the game down to a crawl. IMHO the best way to solve this is to steal the solution used by EUIV and make convoy routing nodal.
Instead of being routed from province to province, convoys should be routed through a limited number of nodes, each of which would be identified with a single sea Strategic Region (for deploying ships and aircraft). To keep the numbers to a computationally feasible level, these would be much larger than at present. Here is another mock-up (red lines indicate straits and canals):
That map is probably still too Eurocentric, but it gives you the idea.
The efficiency of trade routes would be based on the level of sea control at the various nodes that it passed through. The network is still large enough to make AI calculations hard, but hopefully it would be manageable for each tag's AI to periodically (monthly?) calculate the efficiency of its routes (given current sea control levels) and assess whether it might be worthwhile to change the convoy routing and/or redeploy its fleets and aircraft.
The efficiency calculation could be based on the mean of sea control, or on the lowest level of sea control on the route. But in my view, the best way to make this function would be to actually simulate convoys as ships, with missions, and then (a) assign them to strategic regions or (b) simulate the journey. Either approach would allow them to engage in battles and take damage, so each convoy battle is no longer sink-or-survive. Using regional assignment (a), then efficiency would depend upon the minimum convoy capacity of the nodes that the route uses. Using journeys (b), when the convoy arrives at its destination, then the trade efficiency for that route is updated according to the number of convoy ships that made it. Either way, the outcome of a big convoy Battle might really matter to players. And the size of each convoy (whether you assign/send 8 ships every month or 24 ships every 3 months) is something that could be affected by naval doctrines (and if you have the DLC, naval tactics).
I do indeed have many thoughts on things naval (and have made numerous suggestion threads). I'm a bit crook now, so this won't be (too much of) a mega-post (and there are already some great thoughts on the naval game here - I highly recommend 
