Hoi4 massively rewards stat concentration. If you don't have enough divisions to mostly fill your lines, don't make tanks yet; but once you do, then it's far better to have 1 actual tank which can run circles around your enemies than 5 "space marine"-type armor-meme divisions, in almost any scenario.
Still, even if you are aiming to have a single battalion for armor value, you are screwing yourself by picking cast armor over welded. With the cheapest possible 1938 max-armor heavy I - so just max armor, light turret 1, light machine gun (and caveat of christie and 14 speed ticks, so speed is a realistic 7.4kmph) - the welded tank costs 17.7, while the cast tank costs 21.2, or ~20% more. This disparity only increases if you make a reasonable tank design and add real turrets/guns/modules to the tanks, since the cast cost is multiplicative.
Given that, as your average minor, you're not going to be putting more than 16 or so factories on armor meme tanks regardless, you'd have to agree that 1-2 civs for 20% more tanks; or, more fairly, 2 civs for (with 16 50% cap, +25% output factories) 35 more heavies a year (effectively an additional battalion) is pretty worthwhile, no? Playing as a minor the "meta" is to conquer, and spending resources to get more/better equipment NOW as opposed to building mils for later is how you achieve that. It's like how buying equipment from the market is often worthwhile as a minor early on too.
Additionally, chromium is even easier than steel or oil to come by. If you're in the Americas you basically always have access to Cuba, if you're in Europe you basically always have access to Turkey or the USSR (or one of the Asian chromium holders), if you're in Asia you likely are neutral/friends with one of the Raj or Japan at any given time. And if you're just using these tanks for single battalion armor boosting, you don't need all that much chromium anyway, relative to the rubber requirements of any airforce or the tungsten requirements of arty/most cannons.
Chromium is easier than steel or oil depending on the scenario case. What if Turkey is allied to you? What if Turkey is on an alliance you do not want to fight yet? What if you are fighting Turkey but need to go around Crimea / Caucasus to reach it, because the strait is a no-go? What if due to lack of divisions you decided to just place 2 or so divs on Instanbul to shorten your already huge frontline? What if you are democratic and Turkey did not produce any tension? There are many cases to be taken in consideration. In many cases, you are fighitng all alliances at once, or at least 2. You do not want one alliance to get all the provinces you control at the peace table all because they lost a bazillion more men than you, after all. Or you cannot spare the extra men until you take care of other bits of Europe.
On a historical scenario, of course, I can understand what you are saying. On a relatively historical scenario, perhaps as well. On a lot of cases, you won't be touching the Turkey yet.
In most of the cases you have unrestrained acess to Turkey, you already had acess to the french island in the pacific, because you already capitulated france one way or another, after all, pre-1939 paradrop on historical paris and a few other french places is easy as damn. This (french) island is more than enough usually, and so are the southern bits of Yugoslavia. Cuba is irrelevant for most nations. In many cases, unless you are a fan of constant convoy raids and can spare the PP to counter the effects, cuba is a bad choice, specially since in many cases you won't get anything out of it. You are only losing the factories required for the trade.
For early game cast armor is a no-go usually. There is no point on going for it. In my early posts you will note that I used the "Riveted" word. Specially for minors. You want a mass produced tank that still retains a lot of armor. This is only possible with Riveted. You want something that costs 5 or so production. Anything else and it will cost 10 or well over it. If it costs too much, you won't be able to outfit most of your divisions with it. Perhaps not even 24, in order to maximize its early advantage. But my point with @MeinTeam is that he said that a tank with armor and without soft attack is useless. Couldn't be more wrong. You are literally saying that a 50% less dmg is irrelevant, and there are few bonus toe-to-toe that outperform this one. Most of the tanks I produce nowadays just have a crappy machinegun or an early AA turret and never cost over 1 steel. With one or another superpower, of course, I might go with more refined templates, but for minors? PASS.
Hoi4 massively reward stat concentration? Its highly subjective. Your stats can be meaningless in a lot of situations, specially if you are limited by width either due to enemy tactics (guerrila for instance), terrain, or both. Now, imagine if on top of those, the enemy manages to add a -50% dmg buff to your great stats? In many cases, your big divisions full of meat are just that. Divisions full of meat. A smaller div in many cases is stronger than a much bigger one. Also many of those great divisions you speak of usually consume a lot of supply and are not viable for large fronts specially when supply or even manpower is of concern. Which is something to be taken in consideration early game, as you do not want to tie down civs for too long into railroad production. Minors rarely can afford 20 width divisions because the frontline will grow as time passes, and you won't be able to have 1 div per tile if you do it like this. Unless of course, you rely on the big AI buddy to cover for you. I rarely have AI buddies on my alliance, and usually only to farm them for 1 or 2 extra spies, or puppets, to neutralize one front or another while tying a lot of AI divs in the process. Like Italian Ethiopia, or a far eastern Russia (while I grabbed the western part entirely for myself).
As for getting a proper tank that can run circles around the enemy... Unless you can spare the mot, or mecs, you aren't going around anything. Cav + tank template is probably the best you can afford (in terms of speed) or get your hands on. Few countries can afford or make it logically plausible to afford mass mot production in 1938 or something. And with cav + tank template you aren't going to create many encirclements (due to speed at least), and if you are sticking with cav, chances are, you can't spare to go for crazy with tanks either. I suppose you "could" make 3 or 4 divs for offensive roles and with 12 km/h of speed, but at the rate you will be chewing mots its not viable either. Specially not viable if you just form a frontline and tell the AI to handle your invasion plan. Not everyone plays Germany or something like that, you know. And even as Germany, you can abuse the AI in many cases with just cav + 1 tank with a ton of armor. In that screenshot I showed, I did an early rush on netherlands for the rubber so that I could mass produce airplanes without issue and without bothering with refineries. The end result was what I showed. I also produced heavies pratically since game started.
For armor meme battalion tanks, even ultra-lategame you'll never have more than ~200 factories on them, which are easily supported by every continent's chromium reserves (besides South America lol). The rest of your tanks should be riveted armor anyway so it doesn't matter. And again, chromium is all around the Hoi4 globe, often in countries which are usually neutral and/or easy to conquer. Non-strategic materials usage is great too, but aluminum is far more scarce comparatively.
Did I say you would have +200 factories allocated to tank production? I even spoke of SHBBs. You can also be mass producing railway guns, they consume a lot (specially the one unlocked by special project), specially if you are churning them out non stop. On multiple lines at the same time. Late game you have so many factories that you don't even know where to waste them. I have had cases where I just did a crappy lvl 1 or lvl 2 armored car with 150 factories and distributed them across every goddamn territory I held because why not, less manpower losses from garrison duty are always nice and at least they won't go to waste. As I said previous, you can afford the factories but cannot afford the resources. This is why in many cases you will be converting tanks or airplanes, with 1 line being dedicated to produce outdated tanks / planes and another at converting them (this reduces resource comsumption). Or why you are going with Vertical Integration, in order to reduce costs as well. Or going with an industrial concern / political advisor that increase resource gain. LACK OF RESOURCES ARE A REALITY!!!
In these cases (of resource constraints) I will go without a problem with Cast Armor. Who cares about 20% extra costs by this time? You will care about chromium for SHBBs (and other projects) as I pointed out several times. SHBB for instance cost 20 chromium each with only 2 heavy turrets, more if you decide to go nuts with them. And in countless cases you cannot go closed economy, so you are still selling a lot of your chromium to outsiders. We could of course argue on the viability of SHBBs, with 1, 2 or many turrets. Or on the viability of other projects. But that would be a very lenghty conversation you see. And far outside the scope of this thread.
Lack of rubber has no bearing on using cast vs welded armor. It doesn't even have much bearing on what you're arguing, whether or not to use tanks or single-armor-battalion space marines, as cavalry or even special forces/infantry are totally good substitutes for motorized/mech in a division with any number of tanks - i.e. a 9/9 tank/mountaineer division.
9/9 mountaineer+tank division? If you can afford those 9 battalion of tanks why aren't you producing mecs instead? Cheaper, fast, and with 1 strong armor battalion virtually unpierceable. Even mec lvl 1 is. And can be obtained in 1940, probably even before if you use spies or one or another national focus that gives a boost to research.
I also never said that lack of rubber is related to using cast opposed to welded. You either go with tanks for space marine-esk divisions (1 battalion per div), or for fast divisions. If you are going for 1 battalion per army, you want to be on the cheap side and invest on armor - and other stats like soft atk in many cases are therefore irrelevant, but you could go for SP AA if you feel like it. If you are going for full mec + tank, you can afford more refined templates. There is little point on going mountaineer + 9 tanks. If you can afford the 9 tank batallions, you can afford the mec. 8km/h mecs will blow any enemy frontline and with 5 prod levels for being cheaper and 2 or so on reliability cost something like 5.2. And 8 mec battalions + 1 medium/heavy tank already work wonders. You will be obliterating entire enemy frontlines and this on a "create battleplan and press go" automode. The level of hardness of mecs will also spare manpower by a great deal. I suppose there are many possibilities and you can of course reduce mecs and increase tank battalions, specially if you got the org to do it.
Space marines are a good option for when you are starved in terms of resources. As stated, 1 tank will in many cases cost just 1 steel. Mecs will cost 2 steel and 1 rubber, at the very best, and in many cases you cannot afford the rubber. And when you cannot afford the rubber, in many cases you cannot also afford the steel, because all potentially sellers are probably already at war vs you or because you simply want to reduce resource consumption and there is no point in wasting more if this tank that cost 1 steel already does the job. This is true with many countries that can only declare war later on, like unaligned ones poltically speaking.
In short:
- The "space marine vs tank" discussion doesn't convincingly impact if cast armor is good
- A few civilian factories paid now to build many more tanks and faster/earlier conquests are doing a lot more than a few civilian factories building military factories that won't have efficiency for a year
- The ideal application for cast AND welded armor is in a single high-armor tank battalion to take advantage of the armor formula, the bulk of a tank division's tanks should be made with riveted armor for cost
- Because of that, you need relatively few factories on the tanks with cast/welded armor, so the impact of the chromium requirement is minimal
1º You are right.
2º Earlier conquests isn't even possible when you need over 50% tension to declare war or have to face similar restrictions.
3º You are right, but I don't recall saying otherwise. Riveted is possible even for high-armor tanks, but it will depend on how much armor you need and the tank template we are using as the base. And I guess everyone heard about the so called tank meme by now.
4º Will depend on the amount of divs you are fielding with tanks and if you are using them on frontline offensive duty or not. Also on the template itself. Heavy tanks are more costly so you will have to dedicate more factories. For prolongued offensives sometimes 30 factories for heavies isn't enough for 60 space marine divs. Unless you have been producing them for years and have a huge stockpile to last a good while. 30 factories on 1940/1943 heavies will mean 30 chromium and if you are using welded it will mean 60 chromium. And if you are using a heavier turret for SP arty for example, you will be paying even more.
I will also repeat myself one more time and write my:
tldr: Cast has its uses specially when you want more armor and cannot pay the chromium / steel that higher armor levels require or that welded armor needs. The other stats are irrelevant. Its all about the production/resource costs (and/or extra armor) baby.