Exports - Imports
I have been playing around with AI Cities a little bit and looking at Exports - Imports values.
Probably the best thing to do would be to load up some old saved games where you can dip into the World Builder and look at various values. then use Great Spies plus the Infiltrate Mission as well as the "Reveal Tile Mode" -> "Reveal All Tiles" option to check out what is going on in individual Cities. Since old BOTM games won't let you mess around with the World Builder, older games, such as our test games, or test games from other teams from SGOTM 16 or earlier (not SGOTM 17!) might be the ticket to getting more info.
But, from some rudimentary playing around, it looks like Jastrow is right such that if there are three or more AIs that have Trade Routes with each other, or at least one AI that has with Trade Routes with 2 or more AIs, you're far more likely to see a greater difference in the numbers (i.e. you'll see bigger or smaller numbers than 1 and -1).
So, it is certainly possible that there are two groups of 2 AIs, Mansa, and 1 isolated AI. Or, there could be 5 isolated AIs, or some combination thereof of groups of 1 or 2 AIs who have Trade Routes with each other (note that who has Trade Routes with whom says nothing about who knows whom and thus the tech-trading game could be a completely different story).
But, it is also possible that there are 5 AIs out there who know each other but have Barb Cities blocking Trade Routes between them.
So, the AIs could be tech-trading and could all be reached without needing Paratroopers or Astronomy. We can't really know given the small amount of data available to us.
Settler for Mansa?
I mentioned this idea earlier... do we want to build a 4th City for Mansa? One good part is that we could pick where it goes. Another good part is that we can build Settlers relatively cheaply. Another good part is that if he settles a 4th City all by himself, we can just re-use the Settler for a City on the west coast. A final good part is that we can gift this City to Mansa shortly before starting the war (say, 1 turn sooner) so that he'll be tempted to send 2 Military Units to defend the City (or maybe on the turn of declaring war, so that he doesn't feel tempted to build too many new units).
Horse Archer Army
Note that "22 Horse Archers" are a renewable Resource. The survivors can be used to take down Mansa's Gems City and could be used against the western Barb City (or, if our one of our two southern Cultural Bridge attempts works, against a southern AI). That's the whole point of Military Units... their initial cost when weighed against 1 single battle is often questionable, but those that can survive to fight 2 or more battles are where you see the real gains.
Southern City
Like Jastrow, I'm not convinced that a City settled to the south would earn us Intercontinental Trade Routes... that was not the case in the test game, which was supposedly designed to roughly mimic the amount of water in the real game... you need a lot of water to set up an Intercontinental area. The fact that there is Ocean next to the south-western Peaks instead of just wide open Ocean also makes it seem that we'll see Peaks connecting "our" landmass to any southern landmass.
That said, when we finally do get a path to the south (via Culture Bridges in 2 turns' time or via a Great Artist or other means later), it might be a reasonable idea to build such a City... as it would give us a "staging point" into which to dump a bunch of Military Units that are shuttled via Galleys, without incurring Unit Supply Costs and without needing a massive Galley army.
More Water in the Real Game
Actually, looking closer, 1405 = the Land denominator of our Score in the real game, but 1551 = the Land denominator of our Score in the test game.
So, maybe we should be expecting more water on the map. That fact may point to Astronomy, granted.
Still, I'm not an Oxford-beeline expert, but I'm pretty certain that even if we do manage to move our capital to Prague in the short term (because otherwise, Bureaucracy won't be powerful), it will take us a lot longer to get to Astronomy via Education than via ignoring Meditation + Civil Service + Theology and dual-Lightbulbing Astronomy.
That said, I bring us back to an earlier point: if Astronomy is required, on a small map like this one, a Great Artist Bomb actually becomes a very powerful tool, allowing us to skip Astronomy altogether, whether you would have teched it via Education or via Lightbulbing.
But, we'll know in just over 2 turns' time what we're up against to the south... a deep Ocean or a chance to reach the next AI.
If we do need a Great Artist or Astronomy, we can grab Code of Laws in trade by grabbing Alphabet and then can use Caste System to generate a Great Artist.
Military Techs
Otherwise, if we do meet the next AI, the discussion will first turn to maximizing our tech trading and then deciding whether we will need Horse Archers, Knights, or something more powerful to take on the next AI.
If Mansa limps his way to Feudalism sometime after we take his best Cities, that'll be ideal, as we can trade for it when the time is right, saving us 1 tech on the path to Guilds, leaving us only needing to research Metal Casting -> Machinery -> Guilds.
Machinery + Construction also opens up Engineering for faster Road movement and Trebs (and unlocks our junky Unique Unit that is a Pikeman with +100% bonus against Melee units, meaning only that they don't get totally pwned by Macemen, but are still no better against Longbowmen or Musketmen... not much of a valuable unique bonus for your Unique Unit).
Basically, we have a lot of options pre-Cavalry. In fact, Cavalry do NOT get immunity from First Strikes, so they might not be any stronger against Skirms than Cuirassiers, and if we wait for either of those units, we'll be facing Longbowmen out of Mansa anyway.
If we then decide that we need to tech deeply, knowing both Code of Laws and Feudalism will make Civil Service slightly cheaper.
Clicking on the F6 screen on Military Tradition + Rifling (the two techs required for Cavalry) shows us needing to learn 23 techs before we get there. Personally, I'd much prefer to use earlier Military Units against Mansa and at least a couple of AIs (via Cultural Bridges, Great Artist Bombs, going through Barb Cities, or whatever it takes to reach them) then use Rathauses + our large amount of land... which mostly means ignoring the negative impacts of Maintenance by expanding... to power our way through any extra techs needed to deal with the final AIs.
