Civ4 AI Survivor Season 2: Playoff Game Three Summary

The final playoff game before the championship pitted a mixed field of leaders against one another. Two of the game's best economic leaders returned to action, with Gandhi and Mansa Musa hoping to repeat their dominant victories from the opening round of competition. If either of them were left to develop unmolested, it would be difficult to stop their performance. At the other side of the spectrum, this game also featured some of Civ4's most aggressive leaders, particularly in the form of Ragnar. Only Montezuma and Shaka are more likely to declare war than the Viking chieftain, and his presence in one of these games always makes for a wildcard element. Ragnar was joined by several other low peace weight leaders, Kublai Khan and Mao Zedong, lending a Chinese flair to the precedings. Both had proven to be capable competitors in the first round, if not so dominant as some of the other leaders. Finally there was Joao, a mostly unremarkable leader in Civ4 who snuck into this playoff round by virtue of emerging on top in the Wildgame game. I can never pronounce his name correctly, and I was unsure what to expect from the Portuguese.

The map for this contest ended up with four civs in the west and a bit of a gap until reaching the two leaders off in the east. On first glance, this appeared to be a map that was favorable to Gandhi, since none of the insanely aggressive leaders would be located on his doorstep. Mansa Musa would have the more difficult task of bordering both Ragnar and Kublai (with a yellow Mongolian civ for this game to distinguish his borders from Mali's brown). Mao and Joao appeared to have the tougest starting locations, both crammed into the southern tundra reaches of the map, although Joao had much more room for expansion. With a series of desert patches and long rivers of floodplains tiles, this looked like a fun arena for competition.

The luck of a goody hut immediately swung this game in an unusual direction. Kublai popped Mysticism from a hut on the first turn of the game, and immediately began racing Gandhi for the first religion at Polytheism. Gandhi did win this race by virtue of turn order in an otherwise tied outcome. Mansa Musa founded the Meditation religion and opted for Islam. Generally speaking, this game would be divided religiously between a Hindu east and an Islamic west. As far as the location of the initial starting settlers, all of the western leaders went further west, while Gandhi and Joao hugged the eastern coast. This left an even larger patch of empty territory in the middle of the continent, which soon began spawning barbarians left and right, including several barbarian cities. It would take a long time before this region was fully pacified by the Deity AIs, even with their huge anti-barbarian combat bonuses.

Ragnar's capital was especially strong in this game: double (dry) corn resources, double ivory, a plains copper, silks, a nice mixture of grassland and plains hill tiles, and fifteen forests available for chopping. Then his second city had another corn resource and another copper, located again in a fertile river valley. Very good stuff indeed. Meanwhile, Gandhi's Holy City culture was squeezing Joao already, and the Portuguese leader further managed to lose his first non-starting settler to a barbarian archer. WTF Joao. This would have the effect of crippling Joao's growth curve, leaving him stunted in comparison to his neighbors. This was compounded by several barbarian cities sprouting up that blocked Joao's expansion to the west. Once again, Gandhi appeared to have found his way into an excellent situation by the luck of the map draw.

Then just to prove that Joao was no fluke, Ragnar managed to lose his first non-starting settler to a barbarian warrior too, despite having an archer and an axeman nearby. Comment from the Twitch chat: "Can I change my victory prediction to Barbarians?" There were four barb cities already on the map by Turn 35, more than any of the AI-controlled civs. As for Ragnar, yet another dream starting position appeared to be in the process of getting wasted by the AI.

While Ragnar continued to sit on his initial two cities, Mao Zedong was quietly expanding out to five settlements, filling up the southwest corner of the map. He was claiming a lot of the land that should have belonged to the Vikings by all rights. Mansa Musa constructed Stonehenge, but Gandhi managed to build the Great Wall, which was probably the more useful wonder for this particular map. Protection against the barbarians was highly valuable in his semi-isolated position. Then Ragnar launched the first war of the game at an extremely early date, invading Mansa on Turn 49. Since Mali had skirmishers on defense and access to copper for axes and spears, along with Stonehenge to pop borders for defense, this attack seemed unlikely to succeed. Even worse, the AI in Civ4 normally will not build settlers while at war, which meant that Ragnar would remain stuck on three cities until signing a peace treaty. Unfortunately, that quirk of AI logic applies both ways, which meant that Mansa Musa would stop expanding as well to churn out units. In other words, Ragnar might be throwing away his own game, but there was a good chance that he'd be torpedoing Mansa's game at the same time.

Scandinavia and Mali continues to trade units against one another, displaying an impressive level of tactical incompetence in the process. Joao made repeated attacks against a nearby barbarian city, suffering some unlucky dice rolls in sword vs. archer combat and promoting the barbarian defenders. Kublai Khan kept settling the mostly desert region to the north of his starting position, while Gandhi teched away and slowly pushed further west. Despite the endless war with Ragnar, Mansa Musa still was leading the game in GNP and sitting near the top of the scoreboard. I'm always amazed at how strong the Mansa AI performs with anything economic in Civ4. He likely could have held off Ragnar indefinitely and possibly even conquered Scandinavia in time, except that there were other AIs on the board with equally low peace weight, and it was only a matter of time until they intervened. Kublai took the plunge on Turn 96:

Kublai's impressive landgrab had him on top of the leaderboard at this point in time, and in an amazing development, Kublai and Mao appeared to be leading the field in research on Turn 100. I was as shocked as anyone at that, here in a game with both Mansa and Gandhi. Almost as soon as Kublai's war began, Ragnar decided to cash out and sign peace with Mansa. However, this wasn't enough to stop Kublai from pushing through and taking the first city of the game, the Malinese city of Walata. This left Mansa with only five cities, not enough to be competitive in the long run. Gandhi had eleven cities already, and with that capture Kublai had ten cities.

Would Ragnar use this interval of peace to rest and rebuild his civ? Of course not. Half a dozen turns later, Ragnar initiated another conflict by declaring on Gandhi, grabbing the newest Indian city in the middle of the map. Perhaps there was a chance for Ragnar to get back into the game after all. He did have ivory at the capital, and Construction tech was allowing the Vikings to run around with elephants and catapults. On the other hand, Gandhi would now finally start building some military units, and there was a possibility that the sleeping giant had finally awoken. Both the Kublai/Mansa and the Ragnar/Gandhi wars would end up stalling out, with no further cities changing hands after the initial strikes. With longbows and castles starting to appear on the field, further offensive gains seemed unlikely for the moment. Eventually Rangar signed peace with Gandhi, giving back the one city that the Vikings had captured. What a waste of time.

The peace treaty between Ragnar and Gandhi did have one major side effect, however: it allowed Ragnar to redeclare war on Mansa Musa almost immediately thereafter. With Ragnar plowing into the soft underbelly of Mali, and Kublai launching a renewed assault from the west, suddenly Mansa Musa was in dire straits. Djenne and Kumbi Saleh both fell, one city to each of the attackers, and those losses sliced Mansa's territory by half. The attacking forces then converged on Mansa's capital city, which was taken on Turn 148. The final blow came shortly thereafter, with Kublai claiming Mansa's scalp on Turn 159:

Even as Mansa collapsed in the central part of the map, Mao had launched his own war against Gandhi in the east. Mao and Kublai were trading top position on the scoreboard back and forth, with Mao's surprising performance fueled by a number of useful wonders and lots of happy/health resources. Gandhi continued to build more wonders and more temples even as Mao battered his western cities, and he had been slowed by a weird lack of health resources. Even with all that territory, Gandhi only controlled three health resources, limiting the size of all his settlements. Mao inexplicably signed peace at a point when he had Gandhi on the run, and the world returned to peace for the first time in ages on Turn 160. It soon became clear that this peace had been forced on Mao via the Apostolic Palace, which was under India's control. Despite this reprieve, Gandhi's position remained tenuous. Kubali, Mao, and Ragnar all hated him, and together they controlled roughly 70% of the continent. Gandhi's longterm prospects for survival looked grim.

Sure enough, Kublai declared war on Gandhi a short time later. Ragnar was sitting at a diplomatic "Cautious" stance with Gandhi, despite +10 relations from shared religion and shared favorite civics. That was the perfect example of different peace weight in action. The Mongolian invasion sent 59 units streaming across the Indian border... only to be stopped by Gandhi's control of the Apostolic Palace, which once again forced an end to the conflict. It was the same mechanic that had forced peace with Mao earlier in the game. How long could Gandhi hide behind the forced peace from the Apostolic Palace?

Thwarted in the attempt to invade Gandhi, Kublai then turned around and attacked Portugal. This was the rematch of the Kublai/Joao conflict from the opening round, although this time Mongolia seemed to have a much stronger position. Kublai grabbed the Portuguese city on his borders, only to be forced into peace yet again by the Apostolic Palace. That wonder was becoming a real pain from a spectator perspective. We had the worst possible confluence of events here: a Hindu, ultra-pacifist Gandhi doing everything possible to stop war declarations coming in from the aggressive Islamic civs to the west. Kublai invaded Gandhi again on Turn 189, then was forced back to peace again on Turn 191. Argh! This was really getting to be a pain in the rear.

But there was one other conflict that wasn't related to the Hindu civs on the map. Ragnar inexplicably declared war on Mao's China, which was much larger and more technologically advanced. Although Ragnar was attacking with 77 units, and the two civs were even in military strength, the greater size and technology of China looked like it would prove decisive in the long run if the war protracted. For what it's worth, Ragnar did capture the first city of the war. However, Mao bought Kublai into the war on his side, and that spelled instant doom for Ragnar. Mongolia had a massive army of cuirassiers while the Vikings had nothing better than maces and longbows. Scandinavia began crumbling from all sides, with Kublai capturing nearly all of the spoils. The war looked more or less like this:

Midway through the war, Kublai finished research into Rifling tech and began upgrading everything to rifles and cavalry. This snowballed an already uneven war that much further, sending the remaining Viking units scurrying for the exits. Despite some funky AI pathfinding in the final stages of the war, it was only a matter of time until Kublai finished things off. Ragnar was defeated on Turn 224, with Mongolia claiming nearly all of his conquered territory. Kublai's yellow borders were now a massive blob in the center of the map, and a Domination finish to this game was looking more and more likely.

While Ragnar had been dying horribly in the center of the map, Gandhi had swapped to Free Religion civic. This was his attempt to pursue the Cultural victory condition, but by leaving Hinduism, Gandhi lost the chance to keep trolling the rest of the world with forced peace via the Apostolic Palace. In a bizarre twist, Gandhi decided to start a war himself, attacking Mao shortly after Ragnar's defeat. That was... not the best decision. Mao was ahead of Gandhi in tech though, and since Gandhi didn't have Open Borders with Mongolia, he couldn't even reach the bulk of the Chinese cities off in the far west. While Gandhi and Mao traded units with one another to no purpose, Kublai was finishing Assembly Line tech and powering up with factories all across his territory. He also had Mining Inc. founded and close to 20 copies of the relevant resources to power it, combining with the Deity cost discounts for an insane amount of production. Kublai was totally out of control, and everyone was simply waiting for him to make his next move.

The inevitable Mongol invasion of India took place on Turn 236. Well over 100 units came screaming across the border, and Gandhi immediately called on the Apostolic Palace for a forced peace... with Mao. Whoops. That meant that Kublai had ten turns to work with before Gandhi could call on the Apostolic Palace again, and he intended to make good use of that time. Waves of cavs and infantry began rolling up the western Indian colonies. Many of the Mongolian stacks lacked siege, which was a good thing as it meant the AI wouldn't waste time trying to bombard down the defenses. Twenty turns into the war, the front looked like this:

That right there is a picture of a runaway Deity AI laying down the hammer. The Indian defenses were collapsing left and right, even as Gandhi, uh, dialed up the cultural slider? He was trying to go for that Cultural victory, but none of the Indian cities were over 15k culture, and Kublai was refusing to allow any more time for cultural pursuits. The odds that Gandhi could hold out for 167 more turns to win by culture were not good, not good at all. Meanwhile, Mao and Joao had gone to war in the south, a conflict that was going almost as poorly for Portugal as the northern war was going for India. Although Mao lacked the overwhelming power of Kublai, he was still strong enough to slowly take Portuguese territory.

Gandhi spent his final turns building caravels in his last cities for some inexplicable reason. The final two Indian cities fell at the same time on Turn 263, eliminating Gandhi from the competition. This had been a dream game for Kublai, as he claimed his third kill of the match. His Domination percentage was hovering close to 60%, and he needed only a tiny bit of additional land to cross the victory margin. Howveer, after the Mongolian borders popped in ex-Indian lands, it looked as though Kublai was still short of what he needed for victory. With only three civs remaining now, the question was whether Kublai would enter into the China/Portugal war in search of that small remaining spot of territory. In the meantime, Mao slowly pushed further east, grinding down Joao with numbers, not any kind of technological edge. Portugal was reduced to a single city remaining, with Joao hanging out for dear life in his capital.

Then just as it seemed as though Portugal was about to be eliminated from the game, a popup message appeared stating that Kublai had won a Domination victory. What the heck?

After investigating, it turned out in retrospect that Kublai had indeed met the 64% land required for Domination. What had been holding him back was the POPULATION requirement for Domination, where he had been short of the same 64% mark. With his seemingly-endless series of wars over, Kubali had shifted the cities in his empire over to peaceful builds and started growing his population upwards. A half dozen turns of that had been enough to trigger the victory message with no further conflict. It was an odd ending to a game that had been so heavy on warfare, and came at the last minute possible to save Joao's life. Two or three more turns, and it would have been just Kublai and Mao left standing.

This was another game that went in an unexpected direction from what the field had been predicting. Out of 85 entries, not a single person picked Kublai to win - not one! You would think that someone would have lucked into that by accident if nothing else. However, Kublai had seen almost everything break his way throughout the game. Kublai and Mao never fought a single war throughout the game, allying over a shared Islamic faith and serving to guard one another's flanks. Off to the east, Ragnar managed to self-destruct both his game and Mansa Musa's game simultaneously, leaving Kublai free to landgrab most of the northwest uncontested. Even if a lot of that land was a desert wasteland, the Deity AI gets so many bonuses that land of any kind is worth having. Kublai was able to swallow up Mansa's territory with help from Ragnar, and then turn and devour Ragnar in turn when the Vikings were caught on the wrong side of the religious dividing lines. Gandhi was too peaceful to leverage his sizable empire into a military threat, and Joao was simply too far behind from the early stages of the game. The barbarians had played a massive role in this game too, especially by capturing those early settlers from Ragnar and Joao. Neither leader ever truly recovered from that setback, leaving Kublai and Gandhi to dominate their respective corners of the map.

That left Kublai and Mao to advance on to the Championship game. How would they fare against the other winners of the playoff round? There was one more game still to come, and plenty of fireworks left in the final round of the competition.