Civ 4 Tournament Idea

I was just curious if anyone ever thought of this idea.

Take all 52 leaders in civ, and then randomly put them into groups of 7 or 8, making sure that none of the the same civilizations appear twice in a game. That would necessiate 7 games, 3 of which have 8 players and 4 of which have 7. Then play out those games (human picks 1 civ to play) and then the top 2 civs from each of the "preliminary" round move on to the semifinals, where there will be 14 civs left, and then 2 games of 7 civs. The leader that wins an actual victory condition is the one that advances (whether it's the human's leader or an AI leader) and then the next leader in score also advances. Then, from that, the top 3 in each game move on to the final, which will have 6 civs total. Then ... the winner of that game becomes the grand champion of all leaders.

Of course, a large part of who wins is determined by luck and who the human decides to attack, but I'm sure it's doable. It would take quite a lot of games, though (10 in all). Has anyone ever done this before?

Edit: The following games below are done with as little human interaction as possible.


EDIT:

Game 1 is complete. [Writeup below]


Here are the group of AI's in each game that the randomizer I used selected. I put them in the order the randomizer picked, so in game 1 JC was first in turn order (so he wins on ties), SB was second, etc.

Game 1 (Completed: Winner - Justinian I)

1. Julius Caesar
2. Sitting Bull
3. Hammurabi
4. Stalin
5. Ramesses II
6. Justinian I
7. Elizabeth
8. Tokugawa

Game 2

1. Churchill
2. Montezuma
3. Wang Kon
4. Lincoln
5. Suleiman
6. Peter
7. Pericles
8. Pacal II

Game 3

1. Qin Shi Huang
2. Washington
3. Shaka
4. Ragnar
5. Huayna Capac
6. Louis XIV
7. Willem van Oranje
8. Asoka

Game 4

1. Charlemagne
2. Cyrus
3. De Gaulle
4. Bismarck
5. Roosevelt
6. Isabella
7. Genghis Khan

Game 5

1. Catherine
2. Augustus Caesar
3. Alexander
4. Mansa Musa
5. Frederick
6. Gandhi
7. Darius I

Game 6

1. Napoleon
2. Hatshepsut
3. Gilgamesh
4. Kublai Khan
5. Suryavarman II
6. Brennus
7. Mehmed II

Game 7

1. Saladin
2. Boudica
3. Mao Zedong
4. Hannibal
5. Victoria
6. Joao II
7. Zara Yaqob

I tried something similar once.
I randomly seeded AIs against each other on a mirror map and hid my capital from them to see which would come on top.
I never finished the experiment for no reason that I can remember though.

It might be interesting but it doesn't really have any value as the RNG decisions have way too much impact for 1 tournament to be considered important.(Gandhi will most likely lose against Shaka most of the time, but Gandhi might edge out other peaceful civs more easily.)

Now if someone found a way to simulate games faster or have a season before the post-season, it might get interesting. It could work with enough people dedicated to doing it though
The higher the number of games, the more accurate the results.

Yeah, I know it would be difficult.

Just for curiosity, I randomized all the leaders together and in "pot 6" there was Hatshepsut (and Wang Kon), along with

1. Napoleon
2. Gilgamesh
3. Kublai Khan
4. Suryavarman II
5. Brennus

HAHAH she/WK will not survive past the first round.

Maybe I could do this during the summer.



Is there a way for me to just watch the turns go by (i.e. hiding capital away, and setting a program to continue the turns?). If I play out the experiment with NO human player involvement, what's the best way to set the turns to automatically go by?

You'd probably be able to be a passive observer by turning on require complete kills, then deleting your settler+scout/warrior and giving yourself a Great Spy. Not sure about that though.

Unfortunately, even with that you're kind of skewing the results with the civs near your starting position getting extra land.

Lyoncet said:

You'd probably be able to be a passive observer by turning on require complete kills, then deleting your settler+scout/warrior and giving yourself a Great Spy. Not sure about that though.

Click to expand...

This would be so fun to try on an Earth map.

Unfortunately, even with that you're kind of skewing the results with the civs near your starting position getting extra land.

Click to expand...

Or you could just set the number of civs a little higher.

timmy827 said:

Unfortunately, even with that you're kind of skewing the results with the civs near your starting position getting extra land.

Click to expand...

if he's already going to this much trouble he might as well make sure its a fair mirror map with an even number of civs, evenly spaced.

i might very well be interested in reading a summary of such a tourney (one without human involvement preferably).

I just tried the Great Spy thing, but all of the civs met me even though they technically couldn't see me. Oh well.

Can we make bets? My money's on Shaka.

If we're betting, I choose Charlie and Giggles <3

The Better AI mod has a let the AI play me mode you can stop any time and have the computer play your turns. I'm not sure its specific to better ai.. Though you can give it all the starting techs I assume you'd have to play whatever difficulty has no AI bonuses which the other AIs would get over the "human" AI.

There is an AI autoplay that IIRC is separate from betterAI also, which should allow you to spam games rapidly, although you'd have to find a way (aka worldbuilder) to level the bonuses between human and AI or still stick your capitol off somewhere behind peaks in WB

Well this morning while I ate breakfast I ran through a quick 1.5 hr game while watching from the sidelines (yes I even beat TMIT times ). It was on deity, normal speed. I put the first 8 AI's that the randomizer selected in the game, and had 2 continents, 4 vs 4. In retrospect, that was a mistake as one civ would have EASILY won if this were a pangea.

Anyway, the map was as follows. The first screenshot shows where I (Egypt) started. I deleted my start as well as nearby land. However, everything else I left on the map. In other words, the only alterations I made were deleting my start (and of course the fish resources that were there too) and putting me in the top of the map. I then gave myself Great Spies so I could have contact with everyone and I had visibility into their cities for the whole game.



The Civs that were part of the game:

1. Julius Caesar
2. Sitting Bull
3. Hammurabi
4. Stalin
5. Ramesses II (he's not normal yellow because I foolishly chose egypt )
6. Justinian I
7. Elizabeth
8. Tokugawa

They are in the turn order, so if JC and SB had a tie on something, JC wins. This is purely the randomizer's work, not mine.

I have already finished the game, but here's the preview I had before I started. I like to have previews, kind of like what sports people do when they try to preview matches.



Preview for Round 1, Game 1


A continents map does not lend itself well to a domination or conquest victory, due to how the AI has difficulty managing overseas invasions. That being said, while those military conditions are not impossible, it's more likely that we'll see a space race victory or cultural victory. But that's moving ahead of ourselves - let's look at the civs on the continents and their positions. Both continents seem to have an odd peninsula to the north.

In the first continent, where the Hatshepsut start (mine) was removed, we have Tokugawa, Ramesses II, Hammurabi, and Julius Caesar. Tokugawa starts to the north, and he has the advantage of getting slightly extra land than normal. However, much of his land is desert, so Tokugawa might not gain that much of an advantage. My prediction is that Tokugawa will start strong, but fade by the ADs.

Ramesses II starts smack dab in the middle of the continent. He'll encounter trouble with Tokugawa to the north, but peace with Hammurabi to the south. Ramesses II, who starts inland, will likely focus on several wonders. It will be interesting to see how he can handle wonder-building and protection from Tokugawa (and possibly Julius Caesar).

Hammurabi, as said before, starts south of Ramesses II. He will struggle throughout the game because a large part of his territory is tundra and ice. However, militarily, he should be okay in case Julius or Tokugawa have eyes his way, becuase his unique unit, the bowman, has a bonus vs. both of their unique units. Hammurabi will probably be in the bottom half of civs in terms of score by the time the game ends.

Julius Caesar will be the most unpredictable. Will he be able to take advantage of the space near him? Will he be technologically advanced because of his gold resources? Will he utilize his praetorian unit vs. Egypt? Naturally, Rome has iron near them - two sources in fact!

Because of how Julius and Ramesses II appear to have the best land initially, they will probably be the two top contenders in this continent. If Julius survives to the modern era - watch out! He's a beast at that time. My conclusion is that either Ramesses II wins a cultural victory with his religions and wonders, or Julius Caesar wins a space race victory with his production edge.

Meanwhile, in the second continent, we have Elizabeth, Sitting Bull, Justinian I, and Stalin.

Looking at Elizabeth's start, she'll probably be the first one to be eliminated, if any. She starts way off to the coast, which means her expansion will be slowed by the need to build workboats. She has a protective Sitting Bull to her east, which will slow down any military conquests that way. FUrthermore, there is a dangerous Justinian I nearby and an always-hostile Stalin to the north. Maybe Elizabeth can pull her way through with a technological edge, but my guess is no.

Sitting Bull will probably do better than Elizabeth, but not great. He will probably get beaten to city spots by Justinian and have to resort to tundra. Sitting Bull's technological knowledge will also deteriorate by the Renaissance era becuase of his AI personality. At least he probably won't lose cities because of his thorny protective trait and unique building.

Stalin, on the other hand, has a better chance of winning than Elizabeth and Sitting Bull. While his territor is limited, at least he has grasslands rather than plains and tundra. He also has stone, which means he'll likely nab a couple of wonders - possibly the Pyramids. A large part of how well he does will depend on where Stalin founds his first few cities, since he could get boxed in his peninsula by Justinian I.

Justinian I, by the way, is the clear favorite to win this continent. He starts off in the middle of the continent with great land surrounding him and gold in the capital. On top of that, he's got ivory and horses, so he'll be under no shortage of strategic resources. His imperialistic trait means he'll probably gobble up all the land in the middle of the continent, while his religious predilection mean he'll likely get a shrine to help support his expansion. With his land advantage, it will be no surprise if Justinian ends up winning the whole game.

In conclusion, the likely victory condition is Justinian winning a space race victory. Justinian could also win a cultural victory - although Ramesses II might steal some of his religions. Failing that, Julius Caesar has probably the second best shot at victory.

----

If anyone else has a prediction, let me know.

EDIT: By the way, many parts of my prediction were thankfully wrong.

Well this morning while I ate breakfast I ran through a quick 1.5 hr game while watching from the sidelines (yes I even beat TMIT times ).

Click to expand...

Now now, play nice:

http://forums.civfanatics.com/showthread.php?t=352494

What could possibly take 1.5 whole hours ?

Edit: I will rank win predictions by likelihood:

1. Justinian (space)
2. Justinian (culture)
3. Justinian (diplomatic)
4. Julius Caesar (space)
5. Julius Caesar (diplomatic)
6. AP cheese (anybody)

I struggle to place Justin (diplomatic). Maybe I'd put it below the 2 JC wins. Diplo by AIs is essentially a luck box or an AI becoming a reasonably dominant force with his enemies vassaled while his friends like him still.

Reasoning:

Peaceweights + probably religion distribution will make Justinian's continent the favorable techer. Stalin has basically no chance if he winds up a heathen, is unlikely to surpass justinian's REX, and will want to hit liz or sitting bull early. If liz goes heathen, she's toast.

Julius is IMO the most likely candidate for dominant force in the other continent. This continent is more likely to be war torn. Toku has a lot of desert land and techs poorly, so it is likely that JC will overcome him as the dominant military power.

When the continents meet, the warmongers will dislike liz somewhat even if she's alive and in FR. They will hate sitting bull as an absolute certainty, tending to leave justinian alone.

If stalin leaves justinian alone or they go same religion (reasonably likely, justinian is the #1 founder of religion on the map and will have a plot-group connection to stalin the instant one of them has sailing for auto spread), there is a medium-to-high chance that nobody declares on justinian on this map, ever. Depending on what he opts to do (go for other AIs or head to space), he is the most likely victory candidate amongst AIs. If he gets rolling (say cuts down a heathen liz) he could become the dominant military force too ---> he has a 30 unitprob and good tech rate so he can roll if he gets big.

The possible variance on this map comes from JC catching a break (dogpiling one of the peaceful guys on the heels of toku) and getting huge and/or justinian + SB being heathens to each other as distraction.

The dark horses are liz (peace vassal ---> culture) and toku (if he vultures cities off JC in the event that JC is the distraction instead).

The other AIs have a reasonably low chance in this game. Hammy's peace weight will drag him to war but it's hard to picture him beating toku/JC enough to overtake the other continent. Ram is a toasty sandwich. Sitting bull drags tech and generally pursues all victories poorly (never seen him win anything but culture), and is a top "DoW me" candidate for 3 different AIs on the map so not likely to go runaway.

So in conclusion, Justinian is the least likely DoW target and a solid techer. Most of the religions will be on that continent IMO so they'll have a big advantage in tech.

I think Justin whacks his continent, Julius his. You get a Cold-War-style Space Race with Julius ultimately winning...

TheMeInTeam said:

Now now, play nice:

http://forums.civfanatics.com/showthread.php?t=352494

What could possibly take 1.5 whole hours ?

Click to expand...

Curses! Foiled again!

madviking said:

I think Justin whacks his continent, Julius his. You get a Cold-War-style Space Race with Julius ultimately winning...

Click to expand...

Very interesting prediction. I might post the pics up later tonight, maybe in 7-8 hours.

One idea to improve fairness would be to play each map several times with the same start locations but the AIs shuffled round. A bit like duplicate bridge. To get the complete set on an 8 player map like this would mean 8! or 40320 runs. I'll volounteer to do a couple of them

The random draw of the group stage, i.e. which are in each group of 8, probably has a huge effect though so it would be pointless trying to be too even-handed. I would think just 8 rotations, giving each AI a chance in each spot, would be reasonably fair.

My pick for a joker in the pack in this game is Hammurabi. His natural direction of expansion will put him right up against Ram early on, so that might spark tensions of one sort or another. Looking forward to the result

Bookies favourites for the eventual podium out of this group - Justinian and Liz, maybe Julius a long shot.

My pick for a joker in the pack in this game is Hammurabi. His natural direction of expansion will put him right up against Ram early on, so that might spark tensions of one sort or another. Looking forward to the result

Click to expand...

Ham and Ram both have almost no chance of winning outside of "vassal very early, whore tech handouts from their master, and go culture). Sitting Bull and Liz are in similar boats...IMO culture is their only hope and they need not to get beat down (although liz will probably get some religions).

There is a small but reasonable chance that every single religion gets founded by Justinian + liz. If that happens, nobody on JC's continent has any hope at all since at least 1 religion is a big part of the AIs success, giving it culture, , and important civics.

This otta get interesting. I'm hoping for Liz to pull up an upset here, or maybe a Stalin conquest could be nice to. Let's see how this game, and the rest of the tourny will go.

Kid R said:

One idea to improve fairness would be to play each map several times with the same start locations but the AIs shuffled round. A bit like duplicate bridge. To get the complete set on an 8 player map like this would mean 8! or 40320 runs. I'll volounteer to do a couple of them

The random draw of the group stage, i.e. which are in each group of 8, probably has a huge effect though so it would be pointless trying to be too even-handed. I would think just 8 rotations, giving each AI a chance in each spot, would be reasonably fair.

My pick for a joker in the pack in this game is Hammurabi. His natural direction of expansion will put him right up against Ram early on, so that might spark tensions of one sort or another. Looking forward to the result

Bookies favourites for the eventual podium out of this group - Justinian and Liz, maybe Julius a long shot.

Click to expand...

During the summer I'll do most of the 8! trials!

1. Justinian (space)
2. Justinian (culture)
3. Justinian (diplomatic)
4. Julius Caesar (space)
5. Julius Caesar (diplomatic)
6. AP cheese (anybody)

Click to expand...

Very good predictions, TMIT.

I was thinking that, in all fairness, I should send the top 2 leaders to the second round, but have "alternates" for leaders who I think performed best under their circumstances. For this game I just viewed, I opted to have the top 2 AI's advance, but then I had a third and fourth AI also advance too.

For the second game I'll also attempt to introduce more "fairness" in starts, such as replacing tundra with grassland.



Speaking of which ...

Round 1, Game 1

Part 1

By the time it was turn 10, I saw some interesting things.



Sucks to be Sitting Bull. Justinian founded his second city right by the borders, then founded Buddhism there!

By turn 50



It seems like TMIT was right about Justinian. Justinian basically blocked off the northeast part of the continent, and boxed Stalin in by capturing the barbarian city of Vandal. Meanwhile, Tokugawa was smart enough to settle towards Ramesses II! You can also see that Liz's chances of winning are slimmer and slimmer because of how she founded Nottingham to her south rather than north.

And one more thing ... the continent with Rome, Japan, Egypt, and Babylon has NO RELIGION yet! Liz went and founded both Judaism and Hinduism - in the same city!

It wasn't until 650 BC that the religion-less continent got a religion, courtesy of Julius Caesar. You can also see that Justinian, Sitting Bull, and Julius Caesar are currently the board leaders. Don't be fooled by the WHEOOHRN sign - Justinian went out of WHEOOHRN soon.



Part 1 concludes at the 100 turn mark.



My prediction about Elizabeth was right. 5 cities, boxed in .. and no chance to win. Justinian, meanwhile, successfully got Stalin to convert to Buddhism so they are now friendly with each other. Sitting Bull is doing a good job holding his own with his tundra. In fact, he has 11 cities - more than Justinian! Could we possibly see an upset??? Meanwhile on the other continent, Egypt is boxed in with 4 cities while Japan is growing large.

EDIT: Hammurabi = Great Library, Ramesses II = Mausoleum of Maussolos

And on turn 101 ... the first war of the game occurred!