Defending your opponent to death....

[Lawrence of Arabia]

I like to wage an early war. Gimme 10 archers and a civ who doesnt have an early UU, and by 1000 Bc at the latest, Ill be feasting on their ruined capital. Defense is good, but only to give you time to build units to attacks.

[Txurce]

I agree that the "defend to death" strategy is least useful in the ancient era, given that the primary goal is to expand as much as possible before the AI hits the Middle Ages. It works much better as a textbook defensive strategy, most of which tend to leave opponents vulnerable to a counterattack, rather than as the spearpoint of your offensive strategy.

One benefit of the strategy that hasn't been cited is that, while it sacrifices the advantages of a blitzkrieg, it saves you shields in the long run, because you won't need as big a counterstrike force when you do finally hit back. The shields you've saved in building a smaller military could be applied to building an early wonder, or just faster infrastructure, so as to hit republic sooner. As such, it might appeal to a builder playing at higher levels, where early expansion is essential.

That said, the best civ with which to try Vel's strategy is probably Greece, since you could use this tactic without any chokepoints or mountains. Declare war (if they don't), let them come to you and destroy themselves against your hoplites, and only then counter-attack with a smaller force than you otherwise would have required.

Another interesting option is Babylon, since bowmen defend as well as spearmen, and can double as the counter-attacking force, saving you about as many shields as the hoplites would, in terms of building defensive and offensive units.

[RaveScotts]

However in some cases, "Best defense is the good offense", may be taken into consideration. The AI uses many offensive units but too few defensive units when it tries to capture your city. When the enemy is at the gate, attack them as hard as you can and when you kill its defensive units which is mostly 5 attackers to one defenders, the rest of the attacker units will be sitting ducks with their low defense ratings in front of your walls. Your attack will not only gain your offensive units experience for future invasions but also will reduce the enemy opposition strength before they have a chance to attack your defenders behind the city walls. Especially if you're keeping a few horsemen at hand you can hunt the battle-tired and retreated enemy units in a turn or two. For the use of this strategy you may focus on building offensive units without worrying about the incoming threats to your cities. This will also saves much time to counter attack with more efficency pursuing the enemy back to their city
while defending yourself. If you are already waging war on an other front or you have other plans than invasion it will also quicken a peace treaty with the odds on your side as well. The concluding benefit of this strategy is the ability to focus on building offensive units and to use them with more efficiency on both defense and offense.

[Curiosity]

Sounds like it would work well if either:

You have no horses or iron, and you need to hold the AI off with those very limited resources, esp as Greeks, or;

After the AI has been beaten down to only a few cities, so that you can set up 'kill-zones' for the purposes of GL creation.

I dunno, though, as I've only ever played OCC. I can say it's definitely a good idea for that.

I'll see if I can try it in a 'normal' game.