Yes, there's no need for a hex-grid. In reality, this would just be exchanging shaped provinces for hexes of a similar area. It might make map-drawing easier and simplify processing to an extent but would have no other obvious beneficial effect.
Paradox produces real-time grand strategy games. Let's look at other games in the "real-time" genre for solutions - and what do we see? We see maps that are essentially continuous and which units move around on freely. We see micro-management eliminated by providing a system whereby units are grouped and orders can be given to them en-masse.
Play Cossacks 2 for an example of how this might work in a grand-strategy setting - not the turn-based strategic map, but the the real-time battles: In the battles one strives to control villages which dominate cross-roads and resources by maneouvering regiments in columns and lines, the regiments receive reinforcements from the villages and the resources are sent back to the main stockpile via the road-ways. This leads to the wonderful situation where nail-biting Leipzig-like battles can occur, with your forces holding a village at a cross-roads against strong enemy forces advancing along two (or more) roadways from diferent directions.
Translate this into the kind of game you might see in a WW2 setting - what you would see are armies formed in lines on the map (or columns if they are moving quickly along a road), with the player dragging and clicking to show where the armies should form their lines and in which direction they should face, and along which route they should advance/retreat. Similar to the villages in Cossacks 2, provinces would have capitals that you need to control to get the resources and connected to transport links that you need to dominate to deliver resources and suplies.
Now zoom out and imagine a Barbarossa campaign in Russia on this model. The importance of transport links, bridges, etc. would be immediately apparent to the player. Grand sweeps through enemy territory could be displayed on-map - just select THIS panzer army (or drag to select more than one army) on the map and tell it to go THERE and the initial and final positions and orientation of each division in that army will be shown, as well as arrows showing their route across enemy territory (with flags falling and raising in the cities they pass through), and if you are satisifed with what you see you can let it go ahead, and if not, you can tweak it to your hearts content. There would be no need to micro-manage every single division.
The map could actually be simplified on this model, since areas that are now represented as many provinces could be instead be represented as a single, large province. The continental United States, for example, might only have ~42 provinces in it (the six states of New England being grouped in a single province). Only major cities need be represented on-map. Province-shape need no longer be determined by game-play mechanics - long and thin provinces would no longer cause problems. National and provincial borders and city locations could be simply mapped automatically using existing X/Y data or co-ordinate systems. Noly the major transport links need be represented on-map - no more need to worry about minor roads.
The main problem with this is it takes things in a totally different direction to where Paradox has been before, and would probably require tons of work to create an engine capable of modelling it.
Paradox produces real-time grand strategy games. Let's look at other games in the "real-time" genre for solutions - and what do we see? We see maps that are essentially continuous and which units move around on freely. We see micro-management eliminated by providing a system whereby units are grouped and orders can be given to them en-masse.
Play Cossacks 2 for an example of how this might work in a grand-strategy setting - not the turn-based strategic map, but the the real-time battles: In the battles one strives to control villages which dominate cross-roads and resources by maneouvering regiments in columns and lines, the regiments receive reinforcements from the villages and the resources are sent back to the main stockpile via the road-ways. This leads to the wonderful situation where nail-biting Leipzig-like battles can occur, with your forces holding a village at a cross-roads against strong enemy forces advancing along two (or more) roadways from diferent directions.
Translate this into the kind of game you might see in a WW2 setting - what you would see are armies formed in lines on the map (or columns if they are moving quickly along a road), with the player dragging and clicking to show where the armies should form their lines and in which direction they should face, and along which route they should advance/retreat. Similar to the villages in Cossacks 2, provinces would have capitals that you need to control to get the resources and connected to transport links that you need to dominate to deliver resources and suplies.
Now zoom out and imagine a Barbarossa campaign in Russia on this model. The importance of transport links, bridges, etc. would be immediately apparent to the player. Grand sweeps through enemy territory could be displayed on-map - just select THIS panzer army (or drag to select more than one army) on the map and tell it to go THERE and the initial and final positions and orientation of each division in that army will be shown, as well as arrows showing their route across enemy territory (with flags falling and raising in the cities they pass through), and if you are satisifed with what you see you can let it go ahead, and if not, you can tweak it to your hearts content. There would be no need to micro-manage every single division.
The map could actually be simplified on this model, since areas that are now represented as many provinces could be instead be represented as a single, large province. The continental United States, for example, might only have ~42 provinces in it (the six states of New England being grouped in a single province). Only major cities need be represented on-map. Province-shape need no longer be determined by game-play mechanics - long and thin provinces would no longer cause problems. National and provincial borders and city locations could be simply mapped automatically using existing X/Y data or co-ordinate systems. Noly the major transport links need be represented on-map - no more need to worry about minor roads.
The main problem with this is it takes things in a totally different direction to where Paradox has been before, and would probably require tons of work to create an engine capable of modelling it.
