Sticking to the strict rule inf and art only 7/2 is very close to the sweet point. You can appraise this by using the basic Lanchester square law. Efficiency for casualties can be appraised for a specific combat width by maximising HP per manpower times soft attack per manpower. The sweet point for this is very close to the 7/2 but. with modifiers, may shift to something like a 6/2 (ie a slightly higher artillery ratio). However, this does assume that whoever is attacking you has plenty of breakthrough. Raising the artillery content tends to undermine the casualty efficiency due to the reduction of hit points which starts having more impact than the increased firepower. In most real world situations there are so many other factors that it all becomes a bit moot.
this is why i say it's not necessarily true. for it to be true, you have to make assumptions that usually don't track to game situations
this was using 3/3, for example
the above was also using grand battleplan, not superior firepower, because using line arty with integrated support is still nonsense.
"whoever is attacking you has plenty of breakthrough" is not a practical assumption for single player. the vast majority of ai divisions do not (even its tank divisions usually take crits). we also have to make assumptions about attacker coordination and their own soft attack values, both of which will influence whether smaller divs with support companies or 7/2 takes more vs less damage to attacker. 5/0s at width have more total defense than 7/2 at width, but are at risk of being targeted down individually.
regardless, since more damage = attacker taking more crits = they de-org faster, more damage is really valuable for this hypothetical. in fact, there are plenty of cases vs ai where adding damage will hold when otherwise you wouldn't, because at high enough damage output, ai can't org-cycle you anymore (you out-damage its org recovery at max supplied divisions).
although you can certainly look at values like sa/width or others as a proxy for how good a division is, i'm also speaking from tons of practical experience. the parrot theme on reddit is that "attacking with infantry" and "attacking with small widths" will "bleed". when i micro with infantry, i achieve casualty ratios that those same players can't using tanks. this is in large part due to the following:
- most combat vs ai in most wars is the ai battleplan smashing divisions into you. "offensive" divisions are a tiny fraction of combat hours in hoi 4 sp
- regardless of using tanks or infantry, you want to deal a ton of damage of the type that matters vs target (almost always soft in unmodded sp)
- both infantry and tanks can be set up to deal 3000-5000 or more soft attack damage onto a target province, which de-orgs defenders there in < 1 day.
- casualties closing a pocket are minimal.
tanks confer a small advantage during the 3rd point above. people claim this is the difference of hundreds of thousands of manpower, but it's usually more like 5-25k vs ai, which has no high-hardness countermeasures. unless offensive tank or infantry div are a meme, your casualties are > 90% from ai hitting your line regardless, and nearly always with insufficient breakthrough to avoid crits. thus we really care about damage/width, and 7/2s are a poor choice for that using superior firepower.
