PSA: "Random" civ selection is not actually that random

[Sagax]

So I wanted to give a try to the new Pangaea Plus map. Having already played as every civ released so far, I wanted to play fully by the ear: pick Ibn Battuta and let RNG decide the civ selection for me. I got Persia, which I've already played very recently, so I gave it a restart/reroll. Got Carthage - wasn't really feeling it, restarted again. The game rolled Persia for me again. Then Carthage. Then Persia. Then again Persia...

This is when I got a suspicion that maybe the system is not random after all, and might be tied to the leader's "___ choice" that you see when selecting a civ. So I decided to test this by picking a leader that has only one preferred choice - Confucius.

I ended up rolling Han 12 times out of 12. Give it a try if you want, but I think we're safe to make the conclusion.

The "Random" option during civ selection is not actually that random, but instead rolls solely from the 1-3 civs that are marked as the leader's choice. I wonder if this is tied to how AIs pick their starting civs, which tend to be more or less "historical" (and as reflected in the leader's choice icon), unless players in the earlier seats have taken all the valid options.

I can understand the intention when it comes to AI, but would appreciate it if the Random option for players was actually random and pulled from the entire roster.

 

[Gori the Grey]

People sometimes make a mistake in calculating the probability of random events.

If you flip a one-sided coin nine times and it comes up heads all nine times, that doesn't mean that it's somehow more likely to come up tails on the next flip.

 

[StargazingDog]

I'm pretty sure that this has been known since launch. If you choose a leader, and random Civ, you will get only the recommended civs. If you want true random, you must select random leader and random civ. For whatever reason, that actually works, in my experience.

 

[Krikkit1]

They need options for civ selection
Choose
Random Preferred*(AI default)
Unlinked Random (fully Random, ignores Leader choice)

 

[sixty4half]

I haven't ran random as often as you, only 3 times, but I was coming to the same conclusions as you. Both Xerxes and Ibn got Persia for example

 

[Siptah]

As a player that almost always goes full random on everything in civ, I noticed this quickly after launch. Since then, I use random.org to give me random tokens for leader, civ, and map.

 

[evonannoredars]

Hopefully we'll get a true random option in the advanced game setup options in the next update. Could also be fun I think to have an option for a pre-determined random civ selection for all eras.

 

[Arcaian]

People sometimes make a mistake in calculating the probability of random events.

If you flip a one-sided coin nine times and it comes up heads all nine times, that doesn't mean that it's somehow more likely to come up tails on the next flip.

While that is true, the chances of getting 1 of the 11 Antiquity civs with a purely random chance 12 times in a row is (1/11)^12 , or a 0.000000000032% chance

 

[Sagax]

While that is true, the chances of getting 1 of the 11 Antiquity civs with a purely random chance 12 times in a row is (1/11)^12 , or a 0.000000000032% chance

Clearly I should’ve gone to Vegas.

 

[EpicWestern]

If you want true random, you must select random leader and random civ. For whatever reason, that actually works, in my experience.

It doesn't unfortunately. The starting civ you get is at least very heavily influenced (probability wise) by the leader if not outright restricted. One result of this is you'll be getting Greece and Rome a lot more than what an independent random would give you.

I'm a bit torn on how to handle this. I've manually randomed my starting civ/leader using an app and sometimes just a 20 sided die, but then you're losing out on some flavor and synergy. I *mostly* want my civ and leader to match or at least somewhat match. I think next game I'll manually randomize the civ with a die, then flip a coin to see if I choose the most appropriate matching leader (more ambiguity there), but if I lose the flip I just manually randomize the leader.

 

[Niallism]

People sometimes make a mistake in calculating the probability of random events.

If you flip a one-sided coin nine times and it comes up heads all nine times, that doesn't mean that it's somehow more likely to come up tails on the next flip.

It's very clear that's not the situation - nor is that rare knowledge.

 

[Neirai]

I know this will be unpopular to say, but it is a valid workaround --> use random.org and do it manually. It's what I do for games without randomness.

 

[Leucarum]

There is a mod that removes leader preferences for civs floating around out there. I'd say it's my favourite mod apart from the map tacks one. Keeps games way more fresh.

Personally I think having civ unlocks and preferred civs were some of the biggest mistakes on civ switching.

 

[MutilationWave]

There is a mod that removes leader preferences for civs floating around out there. I'd say it's my favourite mod apart from the map tacks one. Keeps games way more fresh.

Personally I think having civ unlocks and preferred civs were some of the biggest mistakes on civ switching.

Nice, I want to get this. Do you remember the name?

 

[Leucarum]

Nice, I want to get this. Do you remember the name?

It's Civ and leader Anarchy by Warlock - https://forums.civfanatics.com/resources/warlocks-leader-civilization-anarchy.32165/

I don't know if it works with the new patch though! Gonna try it out