Too simplistic way of looking at it i think.
Yes, the Pz I and Pz II were light tanks (and already outdated before the war), but small tanks costs less to make. Look at it like this.
It is early 1939, and as Germany you are preparing for war. You have 6 Panzer divisions that are "paper divisions" at present, since they all lack equipment. During your gearing up and preparation process, you have researched and built :
1200 Pz I's
800 Pz II's
200 Pz III's
Lets just for arguments sake say that each panzer division has 2 tank battalions, and each battalion takes 200 tanks (as well as excluding model variants for the purpose of this example).
This means that in total, you can fill 11 battalions with tanks of various types, which still leaves you an entire battalion short. In addition to this, your plans call for forming another 6 panzer divisions before the end of 1940, which is yet another 12 battalions, and you must also replace any tank lost between the outbreak of war and the end of 1940 when your next 6 panzer divisions are formed.
Now, your factories are going at it full tilt, and you produce 100 Pz I's, 150 Pz II's and 50 Pz III's a month.
Switching production over from the 100 completely obsolete Pz I's to the "modern" Pz III only lets you produce another 15 Pz III's a month, which is a "loss" of 85 tanks (gearing bonus, the Pz I being smaller and needing less work / materials, etc). So what do you do.. do you keep cranking out Pz I's so you can fill your units on time, and THEN switch over, or do you switch over immediately, and get more modern tanks, but having a severe shortage of them for a very long time ?
That is how i understand it to work at least, so it'll require planning and compromise. There wont be a way of simply chucking all obsolete tanks on the scrap heap and replace them with modern ones, as your industry wont be able to keep up.
This is a good point, but overlooks the purpose of the thread - the Division Designer. Takng the original situation - 6 Divisions with a mix of Pz I and II, plus a few III apparantly filling 11 battalions, I can't see how this is possible. The Armoured Divisions each comprise two battalions, and since we have been creating light tanks (Pz I and II), they will be light tank battalions. You can't put those 200 Pz III's in a light tank battalion.
So there is only 2000 light tanks to go in the 12 light battalions. I'm not clear whether those tanks would be spread equally between the units, or five Divs would be filled out and one would have none. But let's assume it is a daft idea and that we actually have 6 Panzer Divisions, each with 2 battalions of light tanks which we produced in 1936 using a Panzer Division template, and by 1938 many of the Pz I tanks have been upgraded to Pz II.
We have started producing some Pz III tanks and we want to use these to further upgrade our Divs.
It seems to me we have three choices:
Firstly, we can change the Panzer Division template to include a third battalion in the Armoured Brigade, and have this as a medium tank battalion. All of our Divisions now suddenly have an additional battalion in them, and slowly the 200 Pz III's start flowing into these 6 battalions. As above, possibly spread across the battalions equally each getting 33, or filling one whole battalion first and leaving the other five empty for now.
I suspect the latter, and that means there is a random element to which Div gets the benefit of the upgrade first, but you could micro this by setting priorities on each of the six Divs. But we have a situation where we have 5 or 6 Divs which are significantly understrength of our aims. We want them to each have 600 tanks, and we may only have one Div with that number, and five Divs with 400 tanks
Secondly, we can change the Panzer Division template to replace one of the light battalions with a medium battalion. All of our Divisions now adopt the new design. Let's assume that the medium tank battalions can include light tanks, even though the opposite is not true. So we now still have six "full strength" Divs in the sense that they each have the 400 tanks that we want them to have. Now the system starts swapping out Pz I tanks in the medium battalions for Pz III. We keep the overall total of tanks in our Pz force, but have stronger units. Once all of the Pz I have gone then I would imagine it will start on the Pz II, but again it could swap out all of the tanks in the medium battalion in one Div first. Either way Pz I and II tanks will get replaced and flow back into a central equipment tool to be reused in some way later.
Thirdly, we can fork our Panzer Division template into two templates, keeping the original template and creating a copy (which costs zero Land Combat Experience) which we will name "Panzer Division 39". Then we change the 39 template into our new ToE replacing one light battalion with medium (which does cost LCE). This new template could be used to build a completely new Div from scratch, and use the 200 Pz III tanks our factories already produced. But we have decided we don't want that - we want to keep only 6 Divs for now. But we can now choose when to apply the new 39 template to each existing Div. This gives us complete control over the upgrade process - choose one Div, apply the template and it gets the 200 Pz III's. The other five remain on the original ToE until we decide that we have the production in place to provide them with sufficient Pz III's.
I love the idea of the templates. It seems to me to be perfect to mimic the way that changes actually took place. It gives plenty of scope for fine control over the composition of Divs, if you want. Or to just apply changes across the board. And in our example you can have six Panzer Divisions, but none might be identical, with slight variations in the numbers and types of tanks. and therefore in their combat stats. Wow!!
Yet some are still not satisfied that it "dumbs down" the game? I'm completely lost on that - the use of LCE to limit the use of templates seems wholly correct. As Germany in 1936 I can be given some LCE to start with and get more from events, such as the occupation of the Rhineland or Austria. Anything where large scale movements of units occurred, we can learn from, even if there was no actual combat. This allows the game to give more choices to Germany when war breaks out, while leaving France for instance largely locked into the ToE that it had in 1936. Once war breaks out than I don't supposed that LCE is "precious" at all for Germany, and the people who want full "immersion" and to be able to exactly mimic each historical variation in Divs should be able to do it.
The real limitation is only, as I see it, pre-War. I can't start in 1936 as Japan and overhaul all of my Divs ToE's before Marco Polo in 1937. I could be stuck with the largely useless Mixed Brigades, which experience shows (and here I'm not talking of historical experience, but my own personal experience of playing the game) have too few combat brigades to be effective in the frontline if they encounter full sized enemy Divs, and this even includes when they are in battle alongside of full sized JAP "square" Divs. Normal gameplay then becomes to break up the Mixed Brigades and send their individual brigades to strengthen other units. Since it didn't happen historically, and the China AI can't do the same with all of it's units, then it doesn't appear to be a game breaker, rather a game maker that it provides more of a challenge. You get the LCE after Marco Polo to start re-organising, but once war is taking place it becomes difficult to upgrade/redesign the units on the frontline in China. You are more likely to use the new templates to create Divs for other theatres as the war expands.
And the AI can be taught to use the templates in an historical way. So China can get the new style Divs on the German model as a separate template, either provided in the game set-up in 1936, or later generated in-game by an event (I can't remember when China began the process), and the AI can start to reorganise existing Divs or build new Divs according to the new template, providing it has the production of artillery etc. If it doesn't then it should continue largely using the old template.