Early DOWs on Immortal+

[jorissimo]

In every single game I play on Immortal, I get DOWed around the late BCs or early ADs. Not nearly all of these DOWs end in disaster and occur in both unsuccessful and successful games. They tend to be less of an issue when I plan to go to war in the construction or engineering eras because I'm building units anyway. But my question is whether this is normal? Can early DOWs be avoided consistently through diplo? It saves quite a few hammers, maintenance and happiness if you don't have to whip out an emergency army.

 

[AcaMetis]

On higher difficulties the gap between an AI's early army and a player's military police tends to be the largest, and relative strength is one consideration when AIs decide who to plot on. It means the player is a tempting target, but not to the point one should expect to be declared on every single game.

Being 100% safe from DoWs would mean getting every AI to Friendly before they could start plotting war on you, which is not feasible. That said if you have (for example) Shaka next to you, you can gift him a city that ends up giving him border tension with another neighbour. Even AIs that can plot war at Pleased still have a very significant chance to not plot war on someone they're pleased with, and extra border tension will pull an AI's attention further away from you. It's not a guarantee, but it's better than nothing.

Beyond that you have to play the diplo game correctly - adopt the right religion if any, strategically give in to demands, trade or even give away resources purely to build positive relations, same with maybe not getting the most you can out of a tech trade and/or not trading with someone's worst enemy, etc. It's one of those aspects of the game you really don't have to be particularly effective with on lower difficulty levels to pull off a win, but on higher difficulties, it's life or death.

 

[sampsa]

How often do you city gift? At say 1000BC it's a very cheap way to get some diplo.

 

[jorissimo]

How often do you city gift? At say 1000BC it's a very cheap way to get some diplo.

I tried it in my current game with Shaka, who happened to have a lot of useless desert around him. He hasn't even plotted a single time, let alone on me. Should do it more often.

 

[a pen-dragon]

On higher difficulties the gap between an AI's early army and a player's military police tends to be the largest, and relative strength is one consideration when AIs decide who to plot on. It means the player is a tempting target, but not to the point one should expect to be declared on every single game.

AFAIK the only way that relative strength matters is that the AI will not plot on someone whose strength is above a certain threshold, something like twice an AIs strength.

This will usually not be reached in the early game, even on lower difficulty levels (noble ... emperor), meaning that relative strength does rarely matter to the AI decision at this point in the game. Am I missing something?

What definitely has an influence is that the AI reaches military milestones earlier on higher difficulties and thus starts plotting earlier.

 

[AcaMetis]

I thought AIs were more likely to plot on targets that were perceived as militarily weaker than a target perceived as being militarily equal, but I couldn't give you a source or example of that. One of those bits of knowledge you pick up somewhere and stays stuck forever.

 

[Fippy]

Much depends on which AIs are in the game.
Both Boudi & Brennus don't declare at pleased i.e. despite being fairly aggressive.

Good diplo with those close is most important, while you can usually neglect AIs far away (if needed).
Border overlap matters more than power ratio.
Tech gifts are an easy way to buy security..and giving away something that most others already have doesn't matter (they'd often get that tech from an other AI anyways, if not from you).

 

[drewisfat]

Getting hit that late sounds like a gift city issue. You should have diplomatic safety or an army yourself before the ADs. Dows before like turn 80 aren't avoidable. If next to a psycho or even a neutral AI right in my face I'll tech archery. 4-5 archers can survive most pre-construction attacks, and you don't necessarily need to build all the archers ahead of time. The bigger danger is the attacks after turn 80 that are going to be closer to construction. The priority around this time is gift cities to everyone I can, starting with most dangerous first. Note that if they're already plotting the die is cast and you should gift city to them last. Gift cities are also a great way to get out of a war that's becoming dangerous. Obviously the AI needs to be willing to talk and you need a spot to be able to safely gift.
The worst case scenario is multiple neighbors that plot at pleased in different religions that don't plot early. Sometimes there are tricks, like a wall stall to last long enough to gift a city for peace. I'm also fairly willing to beg the end of a tech (pre-currency but post-alpha) because 10 turn truce buys time for a different plot to happen and for diplo modifiers to build. I usually don't give resources away, but when next to multiple pleased-plotters I'll gift even stuff I need.

You're still ultimately at the mercy of RNG if everything goes wrong, but you can do quite a bit to minimize it. Gift cities should be part of your regular build order. IMP is a great trait provided you have neighbors.

Note that for the extra +1 liberation bonus the gift city needs to be closer to the target capital than any other capitals. Ties aren't good enough, and diagonals count as 1.5 tiles. Ideally we gift very bad cities that stoke border conflicts with other parties, but most aren't going to be so perfect.

*Edit* Another thing to note is that land target status greatly increases the chance of being a target. This is different than the normal modifiers from culture tension. Land target status is when either party has 8+ land tiles that border the other's culture. Against pleased-plotters avoid this like the plague, even taking worse cities or delaying border pops.

 

[BornInCantaloup]

my question is whether this is normal?

Border overlap matters more than power ratio.

Yes, and yes, the latter explaining the former.
AIs follow grand strategies in phases, much like the player should, and their focus towards war is inversely proportionnal to the amount of land they can settle.

- An AI will prioritize settling peacefully vs invading a neighbour ;
- They will even prioritize capturing barb cities vs declaring war vs a neighbour.

Once there is no more material to act as buffer on the map, as would most often be the case in the late BCs, then many AIs will happily shift to making war plans.
It can be very dangerous when a barb city falling triggers that shift, because it could mean those AIs have already amassed serious forces and do not really need a build up phase. I've seen Monty DoW like 1 turn after that last barb city is gone...

How you choose to deal with those neighbours is up to you. You can work on diplo or prepare an invasion of your own.
I will reiterate that leaving a buffer of land for the AIs to expand into is a very safe way to keep them in an expansion phase. It also spares you some money on city maintenance while you research Horseback Riding. Everybody wins !

 

[Tecumseh1]

What helped me improved in this area is start thinking about the diplo setup from beginning. When you first meet AIs you can try gifting a tech or cash. The +4 trade relations thing decays but you can regift later. Look at favorite civics. You can get 4-5+ from some AIs for being in their civic. So you know ahead of time that some AIs will definitely not be a big issue once you hit certain civics like monarchy or civil service. Religion obviously has a big impact as well. And avoid trading with people who are worst enemy of more dangerous AIs. It's actually relatively easy to get guys like Shaka and Ragnar to pleased and even friendly compared to say Hannibal. Shared military battles can be helpful if your neighbor is warring and his enemy is farther away. This is usually a phony war for you with nice diplo boost. Finally, it's OK to give into certain demands. For example, backward Monty demands some tech I'm happy to keep giving into those demands since he's so far behind.

 

[BornInCantaloup]

What helped me improved in this area is start thinking about the diplo setup from beginning

This is very good practice You should make plans for AIs as early as possible.

To add on my initial post, while I will maintain that you can monitor how aggressive the AIs are likely to be depending on how much empty space there is,
I do not recommend to systematically plan for a HBR rush.
I think it's one of the strongest rushes in the game BUT
Most of the early game is about getting edges and, especially, a tech lead. A tech lead amounts to a gigantic discount on military units.

And it doesn't always happen that a rush is profitable vs teching up + invading easier.
A properly set defensive war comes on the cheap. This is an entirely valid option. Repelling an 8 units stack with 5 and then suing for peace while you're pursuing Lib is probably gg. This does require some awareness of the danger, though, and I would think everybody's been caught before
In the cost-effectiveness sense, I do feel a little uneasy with the Construction wars (unless 100% boxed in) and acknowledge not all targets are adequate for Horse Archers.
A costly offensive war in the BCs, I would generally not consider as desirable.

If we push the argument forward that plan N°1 for the player is to get a tech lead, I mean,
Bribes...
They're a lot more efficient if you have military of your own but they're a very good way to completely own the warmongers. Provided you can afford the bribe, bribes can be very expensive if you are not at war yourself.
There are the Gwaja's all warmongers maps to testify for this one. Absolute Zero also loved to do that in his days. I do not diplo much and that is my fault but diplo is wide.