While I hope that everything I'm writing is an educated guess and not just baseless speculation, we're obviously missing enough information to take everything I say and make it utterly meaningless. So this theorycraft is subject to revision or abandonment depending on just how incorrect it is; and in any case will be outstripped by players who are better than I am.
We've seen enough early gameplay to start crafting general strategies for the early game. The limit of our knowledge, and therefore how much theorizing we can do is the lack of late-game play and the lack of hard numbers for some areas of the early game.
The Rise and Fall of Wide and Tall
In order to provide some context to the strategies I'm going to lay out, I want to say that I think the developers have abandoned the Tall versus Wide dichotomy that was forced onto Civ 5. In all prior versions of Civ, the best strategies were ones that let you go both wide and tall - you built as many cities as possible and then you tried to make them as large as possible, in order to bring in more money and hammers. Civ 3 introduced the concept of Great People; and Civ 4 took a deep delve into that concept. By the end of Beyond The Sword's development, there were two clear strategies for building a civilization: The Specialist Economy and the Cottage Economy.
In Civ 5, for a variety of reasons, Firaxis greatly decreased the economic/governmental complexity from 4. They got rid of the slider & made government choices (policies) static. Those two decisions by themselves shifted Civ from being a game where you primarily simulate the economy of an empire into one where you primarily simulate the military of an empire. However, in order to provide decision points and make the game interesting, they provided an option for pursuing tall empires - which have a crude resemblance to the Specialist Economies from 4, or wide empires - which vaguely resemble Cottage Economies.
The problem was that the restrictions they put on Wide were linear, but the benefit from a new city - especially in science - were exponential. So at every difficulty, there was a break-out point where you wanted to go wide - or infinite city sprawl, as its detractors named it. By Brave New World, they'd done everything possible with the limited tools they'd allowed themselves to invalidate wide as a strategy - exponential penalties, nerfed policy trees, even increasing the distance required between cities to physically limit you; with the result that Tall has become the primary strategy at high difficulties.
In Civ 6, Firaxis seems to have made an about face. They've reintroduced economic complexity in three interesting ways: Districts that partially rely on geography for their output, Policies that can be switched, and the Divided Tech Tree. I don't know that this is enough for the game to have economic complexity comparable to Civ 4, but it does provide the tools for economic strategies that aren't as elementary as 5's were.
The other thing they've done, that's very important, is remove the exponential penalties for building a city - replacing it with a one-time cost increase to Settlers. It's unclear if this is exponential or linear, but since it's one time it's surmountable. This means that adding a city is always an option, and your empire will find its size naturally.
With all that in mind, I think I've found 3 basic strategies that a player should align their burgeoning empire towards:
The Specialist Strategy - The Project Economy
I've only seen one video where someone completed a project. It's the Cayinator Aztec video, which is basically uncommented, and I almost missed it. Then he did it again. It made me realize something that had been eating at me during the other post-comment playthroughs - Great People are competed for globally in Civ 6. They have unique attributes and are limited in number. They have timing concerns, with buffs that are only useful in certain eras. In fact, they may be more competed for than Wonders are, since Wonders have geographical restrictions.
There are three ways to acquire great people. First, you can buy them with money. Second, you can buy them with Faith. Third, you can choose to recruit (or pass) on them with the appropriate type of Great People Points.
Cayinator was building for war, and built an encampment fairly early on. He also completed the encampment project. We don't have numbers, but we do know that the 1GGP per turn + the Project gave him enough points to recruit a Great General, in about 7 turns. He did the same thing again, several turns later.
The Project Economy is a build-out focused on having as many districts as possible, and using projects in those districts to pursue great people, in order to either catch up or accelerate ahead as the case may be. It's a synergy based strategy. Again, it's too soon to tell, but this may be the best strategy to unfold into a culture victory in the late game.
Empire configuration: Densely packed cities with shared clusters of districts in the middle. Since most districts have some form of adjacency bonus, you build them in the middle of your cities to maximize their per-turn impact. This will also alleviate the population restriction on districts.
Civics: Start with Urban Planning. Move into Meritocracy, Insulae.
Research: Balanced between tech and culture. The first governments unlock wildcard policies - which provide targeted GP points - and Political Philosophy is probably the first beeline.
Best Civs revealed so far: Japan, Brazil
Best Wonder: The Oracle, Forbidden City
The Generalist Strategy - The Trade Economy
Where the Project economy is going to look for synergies to leapfrog ahead, the generalist strategy is an efficiency strategy, looking for the largest marginal gain from each new citizen or tile. So far, it looks like in Civ 6 this strategy will probably create excess gold which can be used to buy catch-up units and upgrade units and buildings. If a particularly compelling great person comes along, it will be possible to buy that great person for cash.
The Trade Economy build-out looks to place a large number of cities with each one having some breathing room to optimize tile yields. They'll build most districts opportunistically - instead of a campus in the best spot in every city, there'll be a campus in those cities with the best spots. Since money will be a limiting factor for this build, cities will be placed with an eye to two things - harbor placement and commerce hub placement. Most cities will be on a river, within one tile of the coast. In the late game I expect empires that start on this footing will be full of neighborhoods.
Empire Configuration: Widely Scattered cities with lots of land between, as many trade routes as possible.
Civics: Start with God King. Move into Surveyors, Colonization, Ilkum. Edit: Caravanseries, duh.
Research: Prioritizes technology over civics. Plays the Eureka game instead of beelining.
Best Civs revealed so far: Aztecs, China
Best Wonders: Pyramids, Colossus
Hybrid Strategy - The Faith Economy
While it's going to be possible, and desirable, for both of the other two strategies to pursue religions, there also looks to be a strategy that prioritizes religion over other economic concerns. I wasn't sure if this was a full-fledged strategy with its own constraints and goals - I thought the ability to buy Great People with faith might just be a sink for late game production - some of the design choices of India and Spain are making me think it must be.
For this strategy, you pursue synergistic returns from faith improvements - like you would in the Project economy - and accept incremental gains from other areas - like you would in the Trade Economy.
The build-out for the initial three or four cities is probably to cluster them close together, with holy sites adjacent in the center and other districts surrounding those. India's Stepwell becomes real cool here, because it allows for bonus food and faith in those cities. We don't know enough about how religion functions after establishment to say how this plays past that point yet; but we do know that a significant number of Wonders are religious in nature. Very likely, your second wave of cities are settled primarily for geographic features that allows you to compete for Medieval and Renaissance Wonders.
Empire Configuration: Dense core with additional cities more scattered.
Civics: Start with God-King. Scripture, Corvee.
Best Civs revealed so far: India
Research: Unclear. It's possible this strategy might prioritize culture MORE highly than Tech.
Best Wonders: Unclear. Not Stonehenge, which is there to allow other strategies to pursue a religion.
So I'm sure that all this is, at best, 25% accurate, but I do think it allows for a framework for further theorycrafting. I can edit in some links and illustrations if that'd be appreciated.
Theorycrafting - Early General Strategies - Bonus Content: The Death of Tall v Wide
Theorycrafting - Early General Strategies - Bonus Content: The Death of Tall v Wide