@Bison21: Thanks for those links.
As it happens, I've seen both the first two LP's and have been watching @Moriarte's.
One of the things that crossed my mind when I posted was: I wonder what @Bison21 means by “very fine liberty play” and whether what I'm looking for in an LP reflects that description. In this case, I think the answer is that I'm looking for something a little different from what I've seen in those links. What's more, I think the OP might also be looking for something a little different too.
To be precise, I'm looking for an LP which shows a gamer take liberty and REx as much as they can – certainly well beyond the standard four city tradition opening. Now, as good as those LPs are, (and they're terrific gamers of course), RExxing on that scale is actually not something that I saw in any of those LPs. Perhaps I've missed something?
As I recall, @MadDjinn for instance talked about taking liberty and then RExing hard – but actually only settled four cities (plus his capital) before running into happiness issues. Going back to the topic of the thread, this is only one more city than the four city empire that tradition is designed for. Now to be fair, the deity AI managed to settle a couple cities which I'm sure Mad would love to have settled – which may have given a better illustration of the power of liberty (which is what Mad took as his opening social policy tree).
As it happened, @Elcee (known as @Light Cleric here on cfc) also took liberty – but then again settled just four cities for a five city (including the capital) empire. As I'm sure you noticed, @Elcee then went on and conquered a sixth city from Gandhi before T100 – which is more like what I was hoping to see. @Moriarte meanwhile also took liberty – but then only expanded out to four cities in total (for which tradition is designed) before running into happiness issues, and that on pangaea when all AI's can be met for trade pre-astronomy. In other words, whilst all those gamers took liberty, only @Elcee / @Light Cleric actually expanded significantly beyond the standard 4 city opening.
The mention of pangaea in the last paragraph meanwhile takes me to another point. One thing that all these LPs have in common of course is that they're all being (or were) played on a pangaea, which provides access to all AI's for trading pre-astronomy. As such, it goes without saying of course that these maps each provided the ideal scenario to REx: all of the game's luxuries were available to settle and trade for on one continent. Despite that though, whilst each gamer took liberty, they actually only managed to settle (or conquer in @Elcee's game) an initial 4-6 cities before running into happiness issues, and / or running out of room, courtesy of the deity AI. This is why, as an aside, I've often wondered if there's an LP out there illustrating a REx on a standard continents (or continents plus) map, where maybe half of Civ 5's twenty luxuries won't be available to trade or settle pre-astronomy.
Talking of happiness, I still think that remains the number one handicap facing the RExxer. You see the key point to note is that, as mentioned, there are only 20 luxuries in a game of Civ 5 (assuming we ignore mercantile CS and Indonesia's unique resources for a moment.) Now assuming a player settles roughly one city per luxury, that caps them at settling 20 pop one cities pre-ideologies. Each luxury they fail to secure (eg. in trade) is one city less that they can settle. Going back to my point re: maps, this might actually mean that on something like a continents map, as few as 10 luxuries – and therefore 10 pop 1 cities - might be available to settle before hitting happiness caps. And again, even this best case assumes that the player can settle or trade for each luxury. This of course is very unlikely (if not impossible), given both that other AI will settle land (and luxuries) and that the number of luxuries the human can trade for is capped by the number of surplus luxuries they have and / or the amount of gold they can offer in trade.
Of course, there are other ways for the gamer to secure happiness (which they can, for instance, use to grow their cities), but the gamer can only secure four happiness from buildings (via a colosseum and a zoo) in each city before the renaissance finishes. Assuming that each city has been settled for a luxury, this caps each city settled at 5 pop until then. They can of course build wonders or use certain follower beliefs to boost happiness (and therefore population) further, but can of course get beaten to them too. Certain (pre-ideology) social policies can help a little too - see more re: liberty below. Befriending or allying Mercantile CS and constructing other buildings like circuses, stoneworks and UBs can also raise the happy cap further – if those CS have spawned on that landmass (which is another reason why pangaea helps) and the gamer is playing a Civ such as Egypt. Absent those things though, it would seem to me that the gamer playing a standard continents (or continents plus) map is currently capped at settling something like 10 cities at 5 population (or perhaps 6, given the spare happiness they start with), at least until astronomy allows contact with unmet AI (or until ideologies make more happiness available). And doing even this requires building a zoo (and colosseum) in each city. Of course, it also assumes once again that a gamer can trade for each luxury that spawns on their landmass...each luxury they cannot at least secure in trade means that one less city can be settled. Have I missed something? Assuming I haven't it's why I agree wholeheartedly with gamers who've been asking for another luxury or two (such as tea and / or coffee) to be added to Civ 5.
There is of course one other thing that gamer can do to boost early game happiness: take the liberty social policy. As advertised, this social policy adds one happiness for each city connected to the capital (from meritocracy). Now this sounds fine in theory, but it competes with monarchy in tradition as a happiness booster, which provides one happiness (or technically one less unhappiness) for each two citizens in the capital. As a result, unless I'm missing something, liberty is only going to provide more happiness for the RExxer when the number of cities they connect to the capital exceeds half the population of the capital. If the gamer for instance has a pop 10 capital – which all the youtubers mentioned in this long post do quite early to boost research – then liberty (specifically meritocracy) requires connecting six cities to the capital to provide more happiness for the RExxer than the 5 happy cap boost the gamer who goes tradition will get from monarchy. Now given that the gamer might only be able to settle at most 10 pop 1 cities on a continents (plus) map without happiness penalty (depending on how resources have spawned), that's actually quite a task. This is precisely why I also agree completely with those suggesting that Firaxis needs at the very least to boost the amount of happiness coming from meritocracy (and / or lower the happiness boost from monarchy) to change that break even and at least allow the gamer trying to play wide sufficient happiness to do so. As it stands, tradition is a far stronger social policy and can actually provide more early-mid game happiness than liberty, assuming the gamer attempts to grow their capital to boost research.
TL, DR: Thanks for the links but, @Elcee's apart, all the gamers actually take liberty and then simply sit on four or five cities. Some very simple maths meanwhile shows that happiness remains a huge obstacle to the RExxer, especially if they play on maps where meeting AI requires astronomy. Firaxis really need to offer the gamer additional sources of happiness, whether via more luxuries, boosting meritocracy or perhaps taking other steps, if going wide is to be considered a genuine alternative to simply settling four cities and growing them.