Exactly!
So far in this thread the greatest threat, to your early civilisation, has been presented as the barb horseman outpost, and don't misunderstand me, that can be a serious threat to your units, but there is a far greater one, particularly at higher levels of difficulty, and on longer speed games, that I've not seen mentioned yet.....
The AI rush.
Specifically, the early intrusion of your land, using 3-5 warrior units, with the express intention of capturing your capitol and thus, removing you from the game. This tactic carries a far more existential threat to you, even though it utilises units that are far weaker than horseman, for the following reasons:-
1) It can happen very quickly. The AI places it's second settler around turn 6 and turn 15 (marathon), at this point, the AI has a large excess of units, with few reasons now to remain in it's territory, and very many reasons to defeat a weak neighbour.... you!
2) It can happen without warning. The AI is "omniscient"... it doesn't need to "discover" your start position, nor does it need to assess relative army strengths.... therefore, very often, your first contact with your neighbour is also your first indication that your are about to be dog-piled.
3) Your ability to react, at this stage of the game, is very limited. At marathon speed, with a pop 2 city, it can take 10-12 turns to build a slinger, and with not enough gold to buy one outright, you don't have the luxury of time, or gold, to build a defence against it. You'll be fighting with what you have.
Having said all this, the greatest danger to your existence is NOT the sudden influx of units, hit-points or relative strengths. The greatest danger to you is siege and your ability to fight sustained, for the maybe 10-20 turns, it can take for you to defeat them. If your units, and most importantly, your city cannot heal, you will die. Therefore, in higher level games, with longer speeds, larger numbers of civs and maps that promote early contact, it is incumbent upon you to begin with a build that, combined with your initial warrior, can break successive sieges upon your capitol. The only build that I know that can
consistently do this is slinger/*/*/*, by using the poor AI against itself, with the following method.
With the slinger fortified in the city, for extra city def, and the warrior, fortified, on an adjacent hex (with def value) that ideally overlooks two, ring 1, zero def value hexes. With this formation the AI HAS to place ONE of the besieging units in strike range of both of yours. Use the slinger, with no risk of damage to himself, to take the shine of the enemy's armour, overcoming the difficulty penalty, and your warrior to score a "minor" or "major" victory against him. Do NOT go for "decisive" victory as this may result in relinquishing your defensive position and can also expose you to counter attack. The AI tries to save it's own unit by running to your borders for faster healing, thus breaking it's own siege. It may well replace this unit with another, similarly, this first unit may well heal and return, but by utilising this tactic, you expose yourself to minimum damage for the greatest effect, i.e. breaking the siege.
One last thing, about and against the scout/scout/*/*/* build. If your objective is to map and acquire opportunity, whether acting as individual scouts or hunting as pairs, this tends to lead to a "divergent" position, one that dilutes the limited resources available to you, where after 15-20 turns of scouting, your scouts are now 10-15 turns from your capitol, and effectively useless when considering it's defence. You may very well have created opportunistic advantage, but that will be of no value (excepting maybe discovery of a warrior, or +2 cog from an IND CS) when it comes to defending your city, or building an early rush army. Also, the last 10 turns of movement by those scouts, and the information gained, is of less relevancy to your IMMEDIATE decision making, and very often provides info that will not be acted upon for another 50 turns, and thus being out-dated.
Earlier, I posted my opening moves, detailing the scouting pattern of my initial warrior, I finished the post quite quickly.... It had been a long day. I didn't make it clear in that post, so I will do so now. That 15-20 turn circular pattern, allows you to map your immediate area, providing all the information you require to formulate your moves and goals for the next 20-30 turns. It leaves the warrior back where he started, alongside and able to support, your new slinger in the possible defence of your capitol, and/or forming the nucleus of a rush army, with all the relevant map to plan your first expansion, or identify and plan the approach route to your first target....a convergent position, concentrating your resources, optimising your information and providing flexibility in your response.