Civ4 AI Survivor Season 3: Game Seven Writeup


Game Seven featured the return of Huayna Capac, the Season Two champion of Civ4 AI Survivor. The Incan leader had appeared in a total of five games coming into this match, and managed to win four of those five games outright. He has also collected eight kills across those five games, which was far and away the highest amount of anyone. Between this stellar past history, excellent leader traits in Financial/Industrious, and the awesome Incan civilization, Huayna Capac was far and away the favorite to win this match. The other five leaders were a collection of has-beens and also-rans that had done little to distinguish themselves in past seasons. Bismarck and Charlemagne had both been eliminated in their opening round match in the first two seasons, falling as prey to stronger leaders from other civs. Mehmed had been a popular pick to do well in past years, only to fall on his face repeatedly and achieve little. With his strong traits (Expansive/Organized) and the above average Ottoman civilization, there's no good reason why he should have performed so poorly. Tokugawa has one of the worst AI personalities in Civ4, and his isolationism almost invariably dooms him to a weak showing. Somehow Tokugawa had managed a second place finish in Season Two's opening round, which still amazes me to this day. And then there was Wang Kon, the Korean leader. He had also failed to make it out of the opening round in the first two seasons of Civ4 AI Survivor, and made a name for himself mostly for trolling the rest of the competition in various bizarre ways. Given this dubious collection of characters, Huayna Capac seemed likely to find his way to a victory despite the weak terrain at his capital.

The map for this game had the six leaders located in three pairs, with Charlemagne and Wang Kon in the north, Tokugawa and Bismarck in the center, and Huayna Capac and Mehmed in the deep south. This was another game divided sharply along the lines of peace weight, with the two northern leaders and Bismarck falling above average on the scale, while the two southern leaders and Tokugawa graded out with low peace weight scores as "evil" leaders. Some kind of north vs south setup in this game seemed inevitable, with the two leaders in the center playing a crucial role. There was also the question of religion to consider, as Charlemagne, Huayna Capac, and Wang Kon all tend to place a fairly high priority on that element of the gameplay. Korea and the Incans both start with Mysticism tech, and they would end up splitting the first two religions, with Wang Kon establishing Confucianism and Huayna Capac randomly choosing to found Taoism. Much later, Charlemagne would found Christianity on Turn 59, and those would be the three religions that played any sort of role in this game.

For the past few games of AI Survivor, the AI leaders had been doing a good job of sending their initial starting settlers into logical parts of the map. That streak came to an end in this game, as Charlemagne immediately sent his extra settler up north into the tundra to found his second city. Then to prove this wasn't an isolated incident, Mehmed did the same thing and beelined directly into the deep southern tundra with his own settler. Charlemagne had at least picked a good city location with two food resources, but Mehmed was apparently keen to lock down a city with a tundra deer and nothing else of interest. Truly baffling. Then to make matters worse, as all of the leaders were finishing up their first naturally-built settler and getting ready to plant their third city, Mehmed somehow managed to lose his new settler to a barbarian archer:

Note the 6 XP on that barbarian unit as a result of killing the defender and whacking the Ottoman settler. Mehmed now had an iceball for his second city and no third city as a result of this barbarian activity. He was off to a disastrous start just like his Ottoman partner Suleiman in Game Five. Seriously, what was up with the leaders of that civ?! Anyway, Bismarck and Tokugawa had both headed north with their initial settlers, and that seemed to be setting up the map perfectly for Huayna Capac. He had planted his own second city in a nice spot to the northeast, and when Taoism popped up in there it began pumping out Incan culture in all directions. With Mehmed seemingly crippled and the central leaders expanding in the opposite direction, the Incans were off to an excellent start.

As the early turns continued to roll by, we began to see sharper distinctions opening up between the six leaders in this game. Bismarck had a truly insane capital in this game, with a corn resource, double grassland gems, double spices, a banana resource, four hills for production, a river for extra commerce, and about a dozen forests for chopping. However, I had seen this drill before in the previous season where Darius of Persia had a similar capital and managed to throw it away with some terrible management. I suspected that Bismarck would find a way to screw this up, and that's exactly what happened, as the German leader stopped expanding way too early and began building wonder after wonder in his capital - without stone or marble to discount the cost. This left ample open space for Korea to expand down from the south to claim, and helped the Ottomans recover from their horrendous beginning. Mehmed managed to complete Stonehenge on Turn 45, and he would be able to capture a series of barbarian cities later on that helped to preserve some of the territory to his north from being claimed by other civs.

Here was where we stood 50 turns into the game:

Germany's painful lack of expansion was already apparent at the time of this picture, and it would only get worse with time. Charlemagne was the leader at the moment with five cities, as expected for an Imperialistic leader, and despite placing Prague up there in the far north, he looked as though he would wind up with a solid result from the expansion phase. There was another small peninsula in the far northwest where he could place two cities and reach seven or eight total cities via peaceful means. Wang Kon probably would have been in trouble since he wasn't expanding very quickly, but the failures of the Germans were helping out Korea in a big way, as there was still plenty of space available. Wang Kon would later capture the barbarian city of Tartar and even get another city to the SOUTH of that on the coast. Like I said, not a very good showing from Bismarck here. Tokugawa was making good use of a nearby copper resource to build some early military, and he was doing about average in expansion. Going for the Pyramids with no Industrious trait and no stone was a more dubious choice that did not portend well for the future. And Huayna Capac looked OK for the moment, although there were some troubling signs from him as well. The Incan leader had sent his first two naturally-built settlers down into the southern tundra. This left more territory open to be claimed by Tokugawa in the north and Mehmed in the east. In particular, Huayna Capac was suffering from a lack of happiness resources, just barely missing out on a gold resource and an ivory resource that would have let his cities grow larger in size. When the Japanese were able to beat the Incans to a resource cluster southwest of Tokyo, it confirmed that Huayna Capac was going to be playing from behind in the early portions of this game. Going for those cities in the deep south had cost the Incans a lot of land that could have been theirs.

One thing that did favor Huayna Capac was the spread of his Taoist religion. A missionary brought the religion north to Japan first, and Tokugawa converted to Taoism on Turn 62. A dozen turns later, Mehmed also converted after Taoism popped up in one of his cities. This had the makings of an alliance between the southern leaders and Tokugawa in this game, all of them sharing the same religion and predisposed to like one another by virtue of having similar peace weights. Mehmed was also on the rise, somehow, as he captured the two barbarian cities to his north, first Alemanni and then Libyan. Tokugawa managed to finish his Pyramids build, eventually, helped along by connecting a stone resource that had been in the disputed area between his capital and Huayna Capac. Then Tokugawa was on the move again, kicking off this game's first war a little later with an invasion of Germany:

There was no alliance between the Axis powers in this game. Bismarck had managed to reach a grand total of five cities at this point, and two of them were recently placed locations buried in the jungle. All of the other leaders had seven or eight cities, and therefore despite the outstanding terrain at the German capital, Bismarck looked to be the runt of this particular game. Tokugawa had more military strength and production ability, and yet he had chosen to attack before having Construction tech for catapults, and that turned this initial invasion into a bloody stalemate. Bismarck apparently didn't feel very threatened and decided to go right on ahead with building more wonders, completing the Mausoleum in his capital while his western border cities were under siege. For the next two dozen turns, Japan and Germany slugged it out with one another, this conflict serving to weaken both of them without achieving much of anything else. The true winners in this war would be the other nations surrounding them.

Shortly after Turn 100, the game heated up in a big way with a double war declaration:

The big surprise here was Huayna Capac's decision to attack Wang Kon. This was one of those choices that makes sense from an AI coding perspective while failing the common sense test that a human would apply. Wang Kon had the highest peace weight of any of the leaders in this game, and along with founding his own rival religion, that made him the "worst enemy" of Huayna Capac. Unfortunately for the Incan leader, this was a very poor target for an invasion. The Korean cities were so far away that there was no chance of ever capturing and holding any territory, not with the Incan production centers a thousand miles away. Making matters worse, Wang Kon was ahead of Huayna Capac in both economic research and military strength. Huayna Capac would have been much better served by declaring on Bismarck and then working to vulture away some of the scraps of that war. He also might have done better by attacking Japan, which was ruled out diplomatically because of their shared Taoist faith. Perhaps in that respect it had been a mistake sending a missionary to go convert Tokugawa. In any case, this war served no purpose and would backfire badly for the Incan leader.

As for the other war announcement, Mehmed's entry into the conflict against Germany sealed the fate of Bismarck. The German leader was already losing his duel with Tokugawa, and with another leader joining the fray, the disparity in cities was now too large to overcome. Mehmed was researching Construction and once the Ottomans had catapults there would be no place to hide from their miltiary forces. Mehmed actually started off the war by attacking and capturing the city of Berlin on Turn 109, and if the Germans couldn't protect their own capital, that meant they couldn't protect much of anything. Reduced to four scattered cities and at war with two stronger opponents, the writing was on the wall for Bismarck.

The only leader who still remained above the fray at this point was Charlemagne. The leader popularly known as "The Burger King" due to his resemblence to the mascot of an American fast food chain, he had been slowly building up his power in the northwest corner of the continent. Charlemagne had done quite well for himself in the landgrab phase, picking up almost all of the disputed land between his start and the Japanese to the south. This had been helped by Tokugawa's terrible research and pitiful culture; the Japanese leader had refused to research Mysticism tech for the first 80 turns of the game, and Holy Roman culture was dominant along the entire border. Charlemagne finally decided to declare war on Turn 121, invading the Japanese with a huge force of horse archers, swords, and catapults. The target of this attack was the Japanese capital of Kyoto, and we knew that Tokugawa was in serious danger when it fell almost immediately to the attackers from the north:

Tokugawa had fallen afoul of one of the oldest adages in Civilization: don't spend 30 turns laboriously building the Pyramids, instead build a bunch of units and go take the Pyramids away from the guy who did build them. Exhausted by the war against Germany, now it was Japan that was beginning to collapse in the wake of a new threat. Over the following turns we watched as Mehmed systematically dismantled Bismarck while Charlemagne did the same to Tokugawa. In both cases, the central leaders who had started the initial war were the big losers. The other conflict between the Incans and the Koreans was a bit of a sideshow, with the Incans slowly but surely getting pushed back out of Korean territory as the turns passed. Since Germany had so few cities to begin with, it wasn't too long before Mehmed had completed his conquest with our first elimination of the game:

So much for Bismarck and that godly starting position. That's three seasons of Civ4 AI Survivor and three early exits for Bismarck; he does not appear to have a very strong AI personality. Mehmed claimed all five German cities, with the Japanese gaining exactly nothing for their long years of fighting. Amazingly, Mehmed was now the score leader (!) Yes, despite settling in the tundra and losing his settler to a barbarian archer, he was now the one topping the scoreboard. Given his aggressive personality, it seemed likely that Mehmed would soon join one of the other ongoing wars. We wondered if he intervene in time to save Japan, or to turn around the sagging fortunes of Huayna Capac.

In the meantime, the two northern civs continued to advance in their respective wars. One of the great surprises in this game was how well the two of them were doing. Neither Charlemagne nor Wang Kon had done very much in their previous outings of AI Survivor, and yet both of them were leading powerful empires on this map and contesting for the pole position on the scoreboard. They also remained on good terms with one another despite their differing religions due to similar peace weight scores on the high end of the spectrum. Charlemagne had conquered his way through half of Japan by Turn 150, leaving the Japanese with three scattered cities remaining and little hope of survival. As for Wang Kon, he managed to break through and take his first Incan city on Turn 165:

Losing Tiwanaku was a devastating blow for Huayna Capac. It had been his second city and was the Holy City for Taoism, a heavily cultured location that controlled a disproportionate amount of this region. Huayna Capac had been building a lot of wonders, and he had an outside chance of winning by culture even if he wasn't at the top of the score rankings. Tiwanaku would have been one of his three Legendary cities in such an attempt, however, and the loss of this spot made it significantly more difficult to pursue that victory condition. It was apparent by now that this war had been an unmitigated disaster for the Incans, and a peace treaty of any kind to rebuild would have been in Huayna Capac's best interests. Instead, the war continued for the moment unabated, with the Koreans slowly grinding onwards.

The one thing that could save Huayna Capac was the entry of Mehmed into the war against Korea. About half of the Livestream chat was praying for this result, and the other half was hoping for the exact opposite out of a desire to see the pregame favorite get taken down. Eventually half of the crowd saw the outcome that they desired, with Mehmed declaring war on Wang Kon on Turn 174. Before their conflict could properly get started, however, Charlemagne completed his elimination of Tokugawa:

Tokugawa once again managed to self-destruct, ending up with too many opponents and not enough economy. His research rate had been the worst in the game, and there was no answer for the knights that Holy Rome started packing halfway through this conflict. Charlemagne was now running neck-and-neck with Mehmed for the most territory in the game, with both of them having absorbed one of the civilizations that started in the center of the map. It looked like we might be heading for a 2 vs 2 scenario, with the two northern civs against the two southern civs. The northern civs would be favored in that matchup if it were to occur, largely due to the weakness of Huayna Capac. However, Charlemagne now returned back to peace again in the wake of his victory, while the Ottomans and the Incans were both engaged against Korea. If Mehmed could pull off the feat of eliminating or crippling Korea while Charlemagne sat on the sidelines digesting his military conquests, then the southern alliance might be able to come out on top.

It was immediately apparent that Wang Kon was no match for Mehmed. The Ottoman leader had been able to rest up his army and replenish its numbers, and now used them to hit Wang Kon's forces like a truck. First Namp'o fell to the Ottomans followed by Pyongyang, with the latter city being the Confucian Holy City. Meanwhile, most of Wang Kon's military was still down in Incan territory, where they wrestled away another border city from Huayna Capac. That was a strange decisions to say the least. I mean, the city capture was nice and all, but the Korean core was beginning to fall apart back at home. It looked like Wang Kon's days of trolling were about to end. Given another two or three dozen turns, he would be eliminated and Mehmed would emerge as the unstoppable runaway leader.

That was the moment when Charlemagne decided to strike:

Holy Rome was now in a position to flip the tables on Mehmed. Charlemagne had used the past 20 turns to integrate his Japanese conquests and reload his military, which had now grown to an enormous size. I managed to catch the main stack in that screenshot with roughly 60 units joined together of assorted different types. There were also lots and lots of Holy Roman knights running around everywhere, and that spelled real trouble for Mehmed. Some of the Holy Roman units had managed to kill most of the Ottoman defenders at Munich, allowing Wang Kon to slip in and capture the city for himself (with Wang Kon moving after Charlemagne in turn order). This was one city that likely should have been taken by Charlemagne and ended up in Korean hands instead. As for the other war declaration in that screenshot, the long running battle between Huayna Capac and Wang Kon had finally ended on the previous turn... and then Huayna Capac declared war again one turn later. Was that an Apostolic Palace forced re-entry into the same conflict? Mehmed had conquered the wonder away from Wang Kon, and that might have caused this weird situation. In any case, it meant no rest for the weary as the Incans were right back to war with Korea again.

For the first dozen turns of the war, there was no noticeable change in Mehmed's fortunes. He continued pressing Wang Kon hard, and in fact managed to capture the city of Seoul, which had been under siege before Charlemagne joined the conflict. Korea's capital shifted all the way over to Tartar, a former barbarian city on the eastern coast:

At this point, nearly the entire Korean core was in enemy hands of some kind. Wang Kon only held on to his extreme northern holdings and his cities along the eastern coast. However, the cities taken by Mehmed were swamped on all sides by Korean culture and were essentially impossible to defend against the new attack from Holy Rome. Charlemagne could take advantage of Open Borders with Korea to move through Wang Kon's territory and recapture these cities with ease. After taking them, he eventually liberated the cities back to Wang Kon again, with Korea somewhat undeservedly regaining access to its core cities. Over time, the front lines of the war slowly shifted southwards, moving out of Korea and down into the former German lands owned by Mehmed. Just as Charlemagne had hit an exhausted Japan at the perfect time, it was increasingly clear that he had achieved a similar feat against the Ottomans.

This had not been a noteworthy game in terms of technological research, with most of the leaders moving at roughly the same pace through the tech tree. Mehmed did manage to reach Rifling tech before Charlemagne, but the Holy Romans had gone for Military Science instead and had grenadiers aplenty to counter the Ottoman rifles. This seemed to neutralize any such edge there, and Charlemagne's larger territories slowly wore down the Ottoman military. By the time that Charlemagne reached Rifling tech for himself, the Korean core had been reconstituted and the nothern Ottoman territories were beginning to fall:

Mehmed's score had now fallen below Wang Kon's score, a result that would have been all but unthinkable earlier. While this was going on, Huayna Capac and Wang Kon had signed peace and then started another brief war and then signed peace again. The Incans were still building a lot of wonders, including the Statue of Liberty (?!) a little bit later, and it looked like Huayna Capac was thinking about a cultural victory attempt again. He would need to survive to have any chance of pulling that off, however, and the diplomatic situation in this game was not looking favorable for the Incans. The Ottomans had fought valiantly for a long time, but they continued to be pushed further and further back with each passing turn. As Charlemagne gained more territory and Mehmed lost it, the fortunes of war began to swing ever faster against the Ottomans. By the time that Mehmed's capital fell on Turn 267, the Ottomans were in a state of total collapse, and the final blow came sixteen turns later:

It was a massive collapse for Mehmed from the top of the scoreboard to out of the game entirely. He had come back from a terrible opening to eliminate the Germans, then found the situation reversed and himself slowly rolled back and eventually crushed. Charlemagne had instead emerged as the runaway AI, and he looked to be on the cusp of winning a Domination victory. I thought that conquering all of the Ottoman lands might be enough for a victory, but it turned out that he was still about ten percent short of the necessary land area. All of those cities liberated back to Wang Kon up in Korea had limited the amount of territory that Holy Rome acquired in the war against Mehmed. Charlemagne would need to conquer the Incans as well if he were to finish this game via Domination. With that said though, Huayna Capac was extremely unpopular with both of the northern civs. He had declared war on Wang Kon twice, and a combination of peace weight difference and "you declared war on our friend!" penalties made him the worst enemy of Charlemagne as well. It seemed like only a matter of time before the war horns blared again and the yellow horde of Holy Rome rolled over the remaining Incan domains.

As it turned out, when the next attack came it was Wang Kon returning to war with Huayna Capac, not Charlemagne. This renewal of their age-old conflict was almost comically one-sided, with the Koreans fielding infantry against the longbow defenders of the Incan cities. Perhaps this game wasn't going to end in Domination after all, not if the Koreans were able to claim all of the Incan cities. But then Charlemagne entered the war as well:

Now it was a race to see which of the two northern leaders could claim more of the spoils of war from the fall of the Incans, and whether Charlemagne could claim enough land to get over the Domination hump. Wang Kon had taken an important step before Charlemagne even entered the war, capturing the Incan capital of Cuzco as pictured above. This was the repository of several crucial wonders, including both the Statue of Liberty and the Sistine Chapel. The AI loves to run specialists in Civ4, and both of these wonders buff up their performance significantly. Holding the Statue of Liberty was likely worth two or three additional cities given that it affected every settlement on the map.

While Wang Kon was slightly ahead in technology, Charlemagne had a vastly larger army and most of the Incan cities were captured by Holy Rome. Out of the five remaining Incan cities at the time of the above screenshot, four of them would end up getting captured by Charlemagne. However, the fifth and final city of Machu Picchu was captured by Wang Kon, and he claimed credit for eliminating Huayna Capac:

Thus we bid farewell to the Season Two champion of Civ4 AI Survivor. Huayna Capac had never really gotten off the ground in this game, with everything unravelling once he launched that silly war against Wang Kon. When the other leaders with a low peace weight score were picked off one at a time, it left Huayna Capac alone to face the raging storm of the two northern civs. Maybe if he had focused more on settling the central parts of the map instead of the southern wastelands, maybe if he had picked a different opponent for his first war, maybe then it would have turned out differently. Instead, Huayna Capac joined a long list of other big pregame favorites that had failed to make it out of the opening round this season. There was some irony here with Wang Kon claiming the kill, as these two leaders had been matched up against one another in the last season as well. This year it was the Troll King who had emerged on top.

With all of Japan, all of the Ottoman domains, and half of Incan territory under his control, Charlemagne seemed certain to end this game with a Domination victory. We were thinking that his borders in his recent conquests were about to pop and that would be it. So where was he sitting on land area? Holy Rome was very close to a victory after those recent conquests popped their borders:

Charlemagne was at 63.30% out of a needed 64% land area. That was near enough that we started looking into the math of what he needed. There were 801 total land tiles on this map, and a Domination requirement of 64% meant that a leader needed to have 513 tiles within their borders to win. Charlemagne was sitting at 507 tiles at the moment, which indeed equated to 63.3% of the total land area as indicated in the screenshot. So he needed only six more tiles in order to win by Domination, and surely he would be able to manage that, right? Only a funny thing happened as the turns passed: Charlemagne's land percentage began to go DOWN, not up. First he fell to 62%, the 61%, then below 60%. What was happening? Wang Kon's culture was proving to be dominant over Charlemagne's culture, that's what was happening. The combination of the free Statue of Liberty specialist plus Sistine Chapel and Eiffel Tower for bonus culture everywhere was causing the Holy Roman borders to get pushed back all over the map, and especially in the former Incan lands where neither of these leaders had much in the way of culture established as yet.

It soon became apparent that Charlemagne wasn't going to have enough territory to cross over the finish line, despite how incredibly close he had been. This is where Charlemagne's earlier generosity was coming back to bite him in the rear end. He had liberated three different cities back to Korea that had been conquered by Mehmed, and without those cities it was doubtful that Wang Kon would have been able to stave off Domination. There was also the case of Munich, a city where Charlemagne had killed most of the defending Ottoman units only for Wang Kon to swipe in and capture the city at the last second. And then there were the Incan cities themselves; both leaders had surrounded the final Huayna Capac city with large armies, and it had been essentially chance as to which one of them had been able to capture it. If Charlemagne had been the one to capture Machu Picchu instead of Wang Kon, the game also would have been over. For that matter, if the OBSERVER CIV hadn't been occupying nine tiles off in the corner of the map, then Charlemagne also would have been the victor. It was that close. But instead Charlemagne was going to come up a half dozen tiles short of the Domination victory. I was not happy about this - I had Domination as my victory condition and Turn 310 as the finish date in the picking contest, and both would have scored points from a Charlemagne victory here. Wang Kon was trolling once again.

OK, well if Charlemagne was just short of winning with his current territory, what about attacking Wang Kon, who was vastly inferior in military might? The diplomatic screen looked as follows:

Wang Kon was "Friendly" with Charlemagne and would never declare war on the Holy Roman leader. The opposite relationship only had a score of +2, and that was enough enough for "Pleased" relations. However, Charlemagne is one of the AI leaders who will never attack at "Pleased" relations, and therefore he would never declare war on Wang Kon so long as their relations remained at that level. I think that Charlemagne wanted to declare war and go for the Domination victory, as he kept building lots of military units and running them around in circles. His programming stopped it from happening though. We kept checking the diplomatic screen to see if there were any changes, and nope, these leaders remained locked at the same +2 level for the rest of the game. No such luck there. The Diplomatic victory condition was invalid with only two leaders remaining (each of them would vote for themselves), and neither leader was even remotely close to a Cultural victory, with no cities of either leader above 10k culture.

Therefore it was going to be another space race, and neither leader was anywhere close to finishing the tech tree either. It took dozens and dozens more turns for these leaders to slowly research all of the remaining Modern era techs. Wang Kon was slightly ahead in tech at the time that the Incans were eliminated, but Charlemagne had a vastly larger empire and quickly closed that gap. With all of those cities, he was guaranteed to win the space race, right? We checked their respective beakers/turn and it was closer than expected, with Charlemagne getting about 3800 beakers/turn compared to roughly 3000 beakers/turn for Wang Kon. The enormous military expenses being paid by Charlemagne had to be hurting him in that respect, and I suspect that the Statue of Liberty was also helping out the Koreans noticeably. When it came to researching the necessary techs, Charlemagne took one of the dumbest paths possible through the tech tree. He researched both Advanced Flight and Stealth, neither of which is necessary to launch the spaceship. This looked like it might be enough to get Wang Kon back into the space race, only for the Korean leader to research the optional Artillery tech and waste two turns himself. Charlemagne finished the final missing tech (Fusion), and then needed to build the last remaining spaceship parts, the two Engine components. As soon as he did this, he wasted a turn of Anarchy revolting to State Property civic. What the heck Charlemagne?! That narrowed the gap even further, and Wang Kon had built the Space Elevator earlier to make his spaceship production that much faster.

Finally we reached this point on Turn 359:

Both leaders had all of the spaceship techs researched. Charlemagne was due to finish both of his Engine components on the next turn, while Wang Kon was three turns away from finishing the Stasis Chamber. Despite all of the shenanigans and the stupidity from Charlemagne in his tech path, Holy Rome was still going to win the space race. Charlemagne launched the spaceship at the end of Turn 359, and Wang Kon launched two turns later at the end of Turn 361. With a lead of two turns, the game was finally over, and we only needed to wait for the Holy Rome spaceship to arrive at its destination in Alpha Centauri to see the victory message appear. I was pleased to see this outcome, as I had picked Wang Kon to finish in second place (behind Huayna Capac) and would score 5 points in the picking contest when Charlemagne won the game. Losing out on the Domination victory and the early finish date had been annoying, but this was still going to be salvageable from a picking contest perspective.

The victory message didn't appear on Turn 370, which was a little bit odd. I guessed that I must have miscounted the ten turns somehow; it's a little bit wonky sometimes in how the ten turn delay gets counted in terms of the Spaceship victory condition. But there it was popping up on the next turn, letting us all know that the winner of the game was... WANG KON?!?!

The Livestream viewers were treated to a show as they watched me sputter in disbelief at the victory message stating that Korea had won the game. How was that possible? Charlemagne launched his spaceship first, two turns faster than the spaceship from Wang Kon! We had to go back and reload to the previous turn to get some answers. Here was the telling piece of information:

Charlemagne launched with only one Engine part, and that slowed the progress of his spaceship. Some of the Livestream viewers thought that Wang Kon sabotaged the other Engine part with a spy, but when I went back and watched the Livestream again, that did not appear to be the case. Instead, Charlemagne was thrown off due to the vagaries of the production queue. He had two spaceship parts completing on the same turn, one in his capital and one in a nearby city. It's possible to launch the spaceship in Civ4 with only one of the two Engine components, and for some reason Charlemagne AI decided to launch his spaceship AFTER the Engine component finished in his capital but BEFORE the production queue cycled forward to the other Engine component in his other city. Thus he launched with only one Engine instead of two Engines, and that delayed his spaceship by two turns. Both of the spaceships therefore ended up reaching Alpha Centauri on the same turn (1951 AD = Turn 371) and then somehow Wang Kon won the tiebreaker, likely by virtue of having a spaceship with all parts finished instead of a spaceship missing a part. I don't know how the tiebreaker for simultaneous spaceship arrival is determined, but that seems the most likely explanation.

Unbelievable.

Wang Kon therefore ended up winning this game in the most unlikely way possible. I don't think we've ever had an AI leader outright win a game from such a seemingly weak position. We've had leaders survive on near-death status before, or get carried into the playoffs in second place on the coattails of a much stronger first place leader. But to win the game in such a ridiculous fashion? The only comparable situation was Huayna Capac's miracle Cultural victory in the Season Two Championship, and he was never as weak in that game as Wang Kong had been in this game. Korea lost almost every single core city at some point in time, and only by having those liberated back to Wang Kon's control, plus sniping several key city captures away from Huayna Capac, did this game even survive long enough to reach a space race in this first place. Charlemagne was six tiles away from winning by Domination. He wins the game if he captured even one more enemy city, he wins the game if he doesn't waste time researching Stealth and wasting a turn in Anarchy, he wins the game if he didn't somehow launch his spaceship in between finishing the last two components on the same turn. A completely ridiculous number of lucky things had to break Wang Kon's way to emerge with this most unlikely of victories.

One of the Livestream viewers summed up this game the best: Wang Kon had taken his trolling to a whole new level. He had moved beyond trolling the other AI leaders: he was now trolling the observer civ in the real world. Wang Kon's antics ended up costing me 15 points in the scoring contest for this game. Hmmpf. It appears that the Troll King will ride again in the playoffs and have another opportunity to screw with us again.

I am not looking forward to that game.