[NFP] No civs in the Americas

[Linklite]

I've played twice with Earth TSL I the last couple of days, one on Standard, the other Huge, and neither time have any civs spawned in the Americas.

To my remembrance, there are 9 American civs:

1. Aztec.
2. Brazil.
3. Canada.
4. Cree.
5. Gran Columbia.
6. Inca.
7. Mapuche.
8. Maya.
9. US.

For the first games, there were 7 other civs, and then I my current game there are 11 other civs but I'm Gran Columbia, so altogether that I worked out that I should have had a 96.5% chance that I'd get at least one American civ not including me between the two games (not counting personas or double leaders).

Did I beat the odds or has anyone else seen problems?

 

[Alexander's Hetaroi]

I'm not surprised considering over half of the civs are either in Europe or Asia (I'm including the Middle East).

 

[Aurelesk]

I tried to do the maths, but I end up with a odd of 95.3%. How did you end up with 96.5%? I am not able to find my mistake.

Spoiler This is hidden for the mental sanity of many :

There are 12 slots on the regular Huge TS Earth map.

They are 50 civilizations in the game, and 9 are from the New World (America, Aztecs, Brazil, Canada, Crees, Gran Colombia, Incas, Mapuches, Mayas).

You are playing Gran Colombia, so there are 8 remaining New World civilizations out of 49. The odd of not having any other New World civilizations are:
Odds = 41/49 × 40/48 × [...] × 31/39 = ~10.8%, roughly 1 game every 9 games. This case scenario will appear from time to time.

If you didn't took Gran Colombia or any New World civilization, the odds are ~7.9%, roughly 1 game every 13 games. Uunsual, but still a pretty high probability.

On the Standard TS Earth map, they are 8 slots, and the odds are way higher with ~26% (1/4) and ~22% (1/5) if you took or not a New World civilization.

If I combine both probabilities, I have a 95.3% odd of this not happening.

Not that huge difference, but between 95.3% (→ 4.7% → roughly 1 out 21) and 96.5% (→ 3.5% → roughly 1 out 29), we have quite a little gap here.

 

[Yzman]

If I remember my stats at all I think the easy math on this would be there is a 1 -9/50 chance of having at least 1 NW civ. That ends up being 82% chance. Since you were Gran Colombia though, it is actually 1-8/49 = 84%.

Someone can feel free to correct my simplistic math on that, but only an 84% chance of having at least one of those is not great odds to be honest.

I am getting different numbers than the others, so I am sure mine is off somewhere.

 

[Yzman]

I tried to do the maths, but I end up with a odd of 95.3%. How did you end up with 96.5%? I am not able to find my mistake.

Spoiler This is hidden for the mental sanity of many :

There are 12 slots on the regular Huge TS Earth map.

They are 50 civilizations in the game, and 9 are from the New World (America, Aztecs, Brazil, Canada, Crees, Gran Colombia, Incas, Mapuches, Mayas).

You are playing Gran Colombia, so there are 8 remaining New World civilizations out of 49. The odd of not having any other New World civilizations are:
Odds = 41/49 × 40/48 × [...] × 31/39 = ~10.8%, roughly 1 game every 9 games. This case scenario will appear from time to time.

If you didn't took Gran Colombia or any New World civilization, the odds are ~7.9%, roughly 1 game every 13 games. Uunsual, but still a pretty high probability.

On the Standard TS Earth map, they are 8 slots, and the odds are way higher with ~26% (1/4) and ~22% (1/5) if you took or not a New World civilization.

If I combine both probabilities, I have a 95.3% odd of this not happening.

Not that huge difference, but between 95.3% (→ 4.7% → roughly 1 out 21) and 96.5% (→ 3.5% → roughly 1 out 29), we have quite a little gap here.

I've been out of school a bit too long to know the intricacies of why mine and yours differ ( I think it has to do with mine doesn't account for there being 12 slots but with your way I would say that at best it is a 89.2% chance. Your 10.8% is the chance of not selecting a NW civ for any of the remaining 11 slots. So the probability of at least 1 would be 1 - 10.8% or 89.2%. I lose you when you say you combined the probabilities, as you can't just add probabilities together like that for different things.

I am pretty confident that the true answer is between the 84% I got and the 89.2% I got from yours. Either way, it is not unheard of. Around 1 in 10 games won't feature one. And intuitively, this makes sense. 80% of the civs in the game are not NW civs.

 

[Banazir864]

This is one of the reasons why I wish you could have more than two leader pools. For TSL Earth, you could have one pool for each region and guarantee a relatively even distribution.

 

[kylegrove]

Had a Mapuche Huge TSL game. First one had Brazil. Rerolled. On the second, no one in South America but America and Cree were in the game.

 

[Linklite]

I tried to do the maths, but I end up with a odd of 95.3%. How did you end up with 96.5%? I am not able to find my mistake.

Spoiler This is hidden for the mental sanity of many :

There are 12 slots on the regular Huge TS Earth map.

They are 50 civilizations in the game, and 9 are from the New World (America, Aztecs, Brazil, Canada, Crees, Gran Colombia, Incas, Mapuches, Mayas).

You are playing Gran Colombia, so there are 8 remaining New World civilizations out of 49. The odd of not having any other New World civilizations are:
Odds = 41/49 × 40/48 × [...] × 31/39 = ~10.8%, roughly 1 game every 9 games. This case scenario will appear from time to time.

If you didn't took Gran Colombia or any New World civilization, the odds are ~7.9%, roughly 1 game every 13 games. Uunsual, but still a pretty high probability.

On the Standard TS Earth map, they are 8 slots, and the odds are way higher with ~26% (1/4) and ~22% (1/5) if you took or not a New World civilization.

If I combine both probabilities, I have a 95.3% odd of this not happening.

Not that huge difference, but between 95.3% (→ 4.7% → roughly 1 out 21) and 96.5% (→ 3.5% → roughly 1 out 29), we have quite a little gap here.

I did it out of 50 on the second game, my bad. To be fair, it was like 3 o'clock in the morning. Still, low odds, although not unfeasibly low, which is why I wanted to check everyone else to see if they were getting odd things happening.

...I lose you when you say you combined the probabilities, as you can't just add probabilities together like that for different things.

Yes, you can. You just multiply them together. If Event A has a 10% of occurring and Event B has a 20% chance occurring, you just multiply them together, giving you 2%. That's all you were doing previously anyway. If the two games were identical, you could even just lump them together as if they were one big game, it doesn't matter (you couldn't because the second game had more slots but fewer NW civs to choose from).

I am pretty confident that the true answer is between the 84% I got and the 89.2% I got from yours. Either way, it is not unheard of.

I haven't done the correct maths, but that's only true for the second game, you have to combine it with the first game as well.

Around 1 in 10 games won't feature one. And intuitively, this makes sense. 80% of the civs in the game are not NW civs.

It's around 1 in 20 (the chances of the first game were ~43% if I remember rightly). This is reasonable enough for me to accept that I could have just happened to be one of those 1 in 20 people, but unlikely enough from my perspective to warrant asking if I'm alone in experiencing it. If I'm not, then it's probably a bug. If I am, then it's just one of those things.

Speaking of the huge Earth TSL, pictures from the other players (I haven't reached the Old World yet) indicate that Europe is tiny, yet the Americas are gigantic in my game. If it weren't for the fact that I was going for Dom, I'd be really happy because there's plenty of space. If the proportions are genuinely accurate, I wish they'd fudged it to give European civs a chance.

 

[Linklite]

Had a Mapuche Huge TSL game. First one had Brazil. Rerolled. On the second, no one in South America but America and Cree were in the game.

Well, at 8 in 50 civs being NW excluding Mapuche and there being 11 slots (I'm assuming you picked Mapuche and had no duplicates set and left the number of civs as standard), that would mean on average you would expect there to be 1.76 NW civs present in each game. You had 1.5 per game, so I'd say that you had it pretty bang on. Seems that the RNG just screwed me over!

 

[Linklite]

This is one of the reasons why I wish you could have more than two leader pools. For TSL Earth, you could have one pool for each region and guarantee a relatively even distribution.

Personally, I'd have done it without pools. Just have on each random player a dropdown checklist that allows you to disable certain civs from that player. Perhaps have an option to create presets so you can have, say, "New World Civs", "Old World Civs", "Naval Civs", or whatever you want. You can click on those presets, and itnwill select and deselect whatever you have listed under them. And of course, a select/deselect all.

The frustrating part is that you have to have as many options as there are players. You can't just simply say that you want Alex, Cathy, Ghandi and then either Vicky or Engleanor. There is a way round it, but that restricts you in other ways. I prefer my method.

I have to say, the "leaders without wins" option was one that never occurred to me, but is a really good idea. They just need to fix the HoF now...

 

[Vandlys]

Personally, I'd have done it without pools. [...] Perhaps have an option to create presets

You are describing the function of pools : P

 

[Linklite]

You are describing the function of pools : P

That one aspect shares function of the pools, but my method allows for more customisability and I think would be easier to use. To use my example, currently to get that set up, you have to go to the leader pool, set it up so that it includes Alex, Cathy, Ghandi, Vicky and Engleanor. Then set up the game to have Alex, Cathy, Ghandi and a random -leader pool, ensuring that you've got the no-duplicate leaders and no duplicate civs on. Of course, that means that you can't do, say, the same list but with two Alexs, but I can't see any other way.

The idea of leader pools isn't bad, it's how they've implemented it. My way is more flexible, more customisable and, as far as I can tell without actually doing it, more intuitive. It would certainly be easier.

 

[mrwho]

It's honestly absurd that they added these TSL maps but couldn't be bothered to put the effort in to actually make the maps remotely balanced. In the stream they talked at length about all the hoops you have to jump through and significant effort needed to get a remotely balanced spawn. How about they just made TSL maps do this automatically? Boggles my mind they released TSL in its completely broken state.

That said, TSL is actually really fun for the novelty just because of how insanely broken and unbalanced it is. I started in close proximity to two other civs in my most recent game. Grabbed one settler right of the bat - nice, free settler. The other one I managed to almost completely block off, so he was easy to conquest. Then in all of Africa there was only Mansu Musa, who for some reason couldn't be bothered to settle much, so I got the lion's share of the best bits of Africa for free too. Then after exploring I realised South America was also almost completely unoccupied, and thanks to a bunch of forest fires over the centuries, the base rainforest yields in SA were the most bonkers I've ever seen in a civ game by far. It was a deity game but almost like I was playing on Warlord or something...

It was some of the most fun I've had in Civ for a good while, to be fair. I won a culture victory super early, and then once I conolised SA I won a science victory from one city I settled there just for the fun (turns out being surrounded by 8 food 7 production tiles will make a city grow quite quickly). But TSL is completely unviable to play an even remotely balanced game unless you invest half an hour to set it up properly, and even then it won't be very balanced.

 

[Yzman]

It's honestly absurd that they added these TSL maps but couldn't be bothered to put the effort in to actually make the maps remotely balanced. In the stream they talked at length about all the hoops you have to jump through and significant effort needed to get a remotely balanced spawn. How about they just made TSL maps do this automatically? Boggles my mind they released TSL in its completely broken state.

That said, TSL is actually really fun for the novelty just because of how insanely broken and unbalanced it is. I started in close proximity to two other civs in my most recent game. Grabbed one settler right of the bat - nice, free settler. The other one I managed to almost completely block off, so he was easy to conquest. Then in all of Africa there was only Mansu Musa, who for some reason couldn't be bothered to settle much, so I got the lion's share of the best bits of Africa for free too. Then after exploring I realised South America was also almost completely unoccupied, and thanks to a bunch of forest fires over the centuries, the base rainforest yields in SA were the most bonkers I've ever seen in a civ game by far. It was a deity game but almost like I was playing on Warlord or something...
View attachment 594786

It was some of the most fun I've had in Civ for a good while, to be fair. I won a culture victory super early, and then once I conolised SA I won a science victory from one city I settled there just for the fun (turns out being surrounded by 8 food 7 production tiles will make a city grow quite quickly). But TSL is completely unviable to play an even remotely balanced game unless you invest half an hour to set it up properly, and even then it won't be very balanced.

I made a topic about the yields on that map a while ago. I had amazon tiles giving me 15 production and 14 food if I remember correctly. My cities were bonkers.

 

[Yzman]

Yes, you can. You just multiply them together. If Event A has a 10% of occurring and Event B has a 20% chance occurring, you just multiply them together, giving you 2%. That's all you were doing previously anyway. If the two games were identical, you could even just lump them together as if they were one big game, it doesn't matter (you couldn't because the second game had more slots but fewer NW civs to choose from).

I misread the first post. I thought you were just trying to arrive at the odds for a single game, not for both of his games and I thought you were just adding together different maps sizes even though he had played one of them. I attribute it to posting too late.

 

[Eagle Pursuit]

This is why the devs have that trick where they put all the Americas civs in a leader pool and then having a different leader pool for the rest. Then they apply the Americas leader pool to a certain number of civs to guarantee there will always be a civ in the Americas.

 

[Alexander's Hetaroi]

This is why the devs have that trick where they put all the Americas civs in a leader pool and then having a different leader pool for the rest. Then they apply the Americas leader pool to a certain number of civs to guarantee there will always be a civ in the Americas.

Since there are two leader pools I guess you technically do it with African civs too. That way you are at least guaranteed to get one from the Americas and Africa, because most likely one from Europe and Asia will always be present with random leaders enabled.

 

[Linklite]

This is why the devs have that trick where they put all the Americas civs in a leader pool and then having a different leader pool for the rest. Then they apply the Americas leader pool to a certain number of civs to guarantee there will always be a civ in the Americas.

I think it was the Eurasian civs, at least that was what I was told. Europe tends to get a bit crazy.

 

[kryat]

For those trying to do the math here, treating starts as Bernoulli trials is the right way to compute this probability. I’ll try to have it in a bit.

 

[kryat]

Also, as I was thinking about it... this is more complicated than it first appears! My assumption is that the leaders each have equal probability of appearing (mostly), so 54 should be in the denominator. The inclusion of multiple leaders for a single civ or multiple civs for a leaders definitely complicates it though.

My simplistic assumption is that the leaders are picked first, but in any invalid combos, one is thrown out, then rolled again (ie, Gorgo and Pericles). Then, for any leaders who could represent multiple civs, or who have multiple personae, a coin flip happens to decide.