DM’s got a point.
Just because the ratio is good, does not mean you’ll actually have the ratio by the time you get to battle unless you distort the ratio to account for losses along the way. (ie 5/1/1 if you want 3/1/1 at the end, or even 4/2/1 if you think you might have to divert an artillery to trading and want 3/1/1 at the end.)
The amphibious point raised by someone else is also appropriate. 2@2 = 4 punch; 1@1 + 1@3 = 4 Punch. However, there are times that the armor is a better choice.
If you have Infantry + Artillery vs 2 Defending Infantry, you have a 45% chance of getting one hit, or 10% chance of getting two hits in one round of battle. (Average over 10,000 games.) This translates to a chance to win the battle of only 45% over 2.6 rounds.
If you have Infantry + Armor vs 2 Defending Infantry, you have a 50% chance of getting one hit, or 9% chance of getting two hits in one round of battle. (again, average over 10,000 games.) This translates to a chance to win the battle of only 51% over 2.5 rounds.
So the odds are slightly better that your Infantry + Armor will win than your Infantry + Artillery will win.
Obviously you normally bring more to the mix if the defender has two infantry, so we can run it again with only one defender:
Infantry + Artillery (10,000 games, to the death) = 87% odds (with 52% chance of having no losses.)
Infantry + Armor (10,000 games, to the death) = 91% odds (with 55% chance of having no losses.)
These differences might seem negligent to some, but they do add up over time. Furthermore, the armor defending against the counter attack is significantly more powerful than the artillery.
As for me, I find Russia generally has more artillery than 3/1/1. It’s closer to 9/6/2 because I need that artillery to tade territories.
Likewise, because of their massive income; Japan and Germany skew differently with something like 9/1/4. (Germany can only build 10 units guys, 5 Infantry + 5 Armor = 40 IPC. Even that isn’t enough to use up all the cash most of the time!)