Civ4 AI Survivor Season 3: Wildcard Game Alternate Histories

(Note : "A" column tracks the number of war declarations initiated by the AI, "D" the number of times the AI is declared upon, "F" the points for finish ranking, and "K" the number of kills.)

Amicalola: It became clear almost immediately that the Season 3 Wildcard Game was a battle between two leaders, with everyone else playing a supporting role at best, and cannon fodder at worst. Suryavarman and Kublai Khan absolutely crushed these games - they won 80% of the time, hogged over half of the combined first/second places, and were never first to die. They killed an average of over 3 leaders per game between them, died only twice each, and were both in contention to win for basically every game. Indeed, each of them scored more than the bottom-8 leaders combined! These high scores were especially remarkable in such a large game, and emblematic of the fact that no one else really stood a chance. So what were the reasons for this dominance?

To begin with, it is important to note the general weakness of the high peaceweight leaders on this map. Victoria was by far the strongest, but that only amounted to a single win and second place finish. The issue for the peaceniks was clear: they were outnumbered, and the Montezumas and Suryavarmans of the world were more than happy to take advantage. Each of these leaders were dogpiled in basically every game, following the rough pattern of Joao -> Lincoln -> Victoria/Washington. This becomes evidently clear when looking at their death dates: all four leaders died most of the time, in that rough order. Joao in particular was extremely screwed, stuck in a jungle-heavy start and surrounded by three leaders (Kublai, Suryavarman, and Suleiman) who attacked him in virtually every game. Lincoln was generally very weak, and would fall to whichever random warmongers attacked him (and this truly could be any of them), while Washington was locked in a death struggle with Napoleon that usually ended with the other leaders putting him out of his misery late in the game. This left Victoria, who was individually quite strong, but could never fight against the hordes of enemies that would come for her- in my notes, there are about ten variations of "Victoria gets to redcoats extremely early, and dies to a macemen dogpile."

The next thing to observe is who profits the most from the peaceniks' demise. Joao was the clear weakest leader here, which left Suryavarman, Kublai Khan, Suleiman, and Montezuma as the immediate biggest beneficiaries. However, Suleiman struggled greatly with production issues in this game, while Montezuma was generally highly ineffective; thus, Kublai Khan and Suryavarman were able to split the Portuguese land most of the time. Lincoln and Victoria's deaths also seem to benefit Monte the most, however as mentioned, he was shockingly weak, so instead it was Brennus, Suryavarman, and Kublai who tended to absorb the land of these two, while Washington was killed by Kublai most of the time. This left Suryavarman, Kublai Khan, and (less often) Brennus or Pacal with strangely disjointed empires in the midgame, while the late-game would typically revolve around these leaders trimming the fat left behind: Suleiman and Montezuma.

Eventually, most games came down to a competition between some combination of these four leaders, and each had different strengths and weaknesses that explained their performances. Beginning with the big dogs, Suryavarman was usually the 'biggest' leader by this point, with the most territory and population under his control. Sury was usually behind in tech until this point, but could often outproduce the other leaders, and win the game through more fighting. In contrast, Kublai Khan often tried to convert his (usual) tech lead into a spaceship win before Suryavarman's bigger-sized empire could catch up with him. When the two fought, which happened a decent amount, it could go either way - sometimes Sury's size would prove the difference, while other times Kublai would run him over with mechs and modern armour against puny humans. Either way, it sure was fun to watch! As for the other two, Brennus' inability to declare war at pleased proved a huge benefit in Games 6 and 9, where other top leaders fought inconclusively and allowed Brennus to establish a tech lead. This didn't happen often, but it did occur enough to feel like a genuine plausibility, rather than a total outlier. Meanwhile, Pacal tended to fold like a house of cards when attacked, but he was not attacked often, and in a few games this allowed him to become a dominant power (particularly Game 14, where he won the group's sole cultural victory). In sum, there were three key ways this game could play out: either Kublai would leverage his obscene starting position into dominance, Suryavarman would fight tooth-and-claw against the world to do the same, or another leader would get a series of lucky breaks to overtake them.

With that said, here is a look at the individual leaders:

Leader Summaries

Suryavarman of the Khmer
Wars Declared: 64
Wars Declared Upon: 23
Survival Percentage: 90%
Finishes: 9 Firsts, 4 Seconds (53 points)
Kills: 33
Overall Score: 86 points

Suryavarman was the deserving winner of this set of Alternate Histories, racking up the most wins, kills, and points of the group. He advanced the most often, and had several more near misses from third place. Sury was also the most flexible leader, winning thrice by Domination, Spaceship, and Diplomacy each. This was all especially impressive considering his starting position: whereas Kublai had a fertile, shielded, and isolated-from-barbarians start (as did Victoria), Sury's position was central, surrounded by enemies, and ravaged by barbarians in every single game. Suryavarman basically never exited the landgrab in pole position, and his barb-infested civ was sometimes in the bottom three on the scoreboard by Turn 100. He would often miss out on the premium central land, and have to content himself with settling the tundra to keep up in cities. Despite this, Suryavarman grimly soldiered on, fighting his way to the top in nearly every game. Suryavarman's most common opponent early on was Joao, but he also fought Suleiman, Montezuma, and a surprising Brennus in multiple games. Suryavarman was able to individually overpower any of these opponents with sheer brute force, and both of Brennus' first-to-die games saw Suryavarman turn back a Celtic invasion with extreme prejudice. Once he'd absorbed one of his neighbours, Suryavarman was a force unleashed, and he would generally go on to paint the map red; it was only a question of whether someone else (usually Kublai) could win first. Sury's only deaths came from two extreme outliers: in Game 9 he attacked Victoria first, while in Game 16 Washington killed him en route to a near Domination win. His only major issue was technology, where Sury would often find himself behind the other major powers (though he was usually able to pull off conquests anyway). Suryavarman was otherwise an extremely consistent leader, contending for victory in nearly every game, and proving that his seeded status is most certainly not a fluke.

Kublai Khan of the Mongols
Wars Declared: 70
Wars Declared Upon: 16
Survival Percentage: 90%
Finishes: 7 Firsts, 4 Seconds (43 points)
Kills: 28
Overall Score: 71 points

Kublai was the other dominant leader on this map, advancing just over half the time and with a survival rate equal to Suryavarman's. This was less impressive in context though - where Sury had a tough central position, Kublai Khan's start was probably the best on the map, with the rare combination of spacious territory and barbarian-insulation. He was also remarkably isolated, facing less than one defensive war per game, meaning Kublai could pretty much always pick his battles. With that in mind, Kublai had a few paths to success. Sometimes, he solo-killed Pacal (in all of Games 1, 6, 7, 12, and 16). Alternatively, he was very well-placed to profit off of Joao's inevitable demise, and he also devoured Lincoln surprisingly often in the early game. Any of these regular occurrences set Kublai up for success, typically by continuing the warpath against Victoria, Washington, and Suleiman. Though he was not usually able to beat Suryavarman in a fight, he was able to out-tech the Khmer, and thus spaceship wins were Kublai's bread and butter. Unlike Suryavarman, who was almost always a contender, Kublai also had a few also-ran games where he never became dominant (such as Games 3 and 7). These occurred when his early fights stalemated, which was more common than Suryavarman's for whatever reason; perhaps Kublai built less units, or was less likely to join dogpiles. This disparity was likely the difference between their performances, and Kublai was definitely slightly weaker than Suryavarman overall. None of this is to say that Kublai wasn't a strong or deserving victor - it's just a little disappointing that the Real Wildcard Game went to the leader handed the victory on a silver platter, rather than the one who fought the world for it.

Brennus of the Celts
Wars Declared: 46
Wars Declared Upon: 11
Survival Percentage: 75%
Finishes: 2 Firsts, 4 Seconds (18 points)
Kills: 17
Overall Score: 35 points

After Suryavarman and Kublai Khan, there was a very big drop off to the remaining 9 leaders. Brennus was clearly the best of the rest though, with a pair of victories and a smattering of second-place finishes. Brennus was a bit of an odd bird in these games, because sometimes he was very strong, and sometimes he was a complete trainwreck. In both of his first-to-die performances, he attacked Suryavarman only to find that he'd Chosen Unwisely and get reverse-conquered. Other times, Brennus would expand terribly, perhaps as the result of barbarians. Alternatively though, Brennus could come out of the early turns as a strong contender, and pick smart wars against high peaceweight leaders to gain more territory. His games typically revolved around religion, which was both a strength (when cultural pressure pushed barbarians back early) and a curse (when it caused him to attack leaders much stronger than him, or kept him peace-locked while Suryavarman and Kublai continued expanding). The surprising part of Brennus' games were how little he attacked Washington; instead Brennus near-invariably went West, into Suryavarman or further into Lincoln/Victoria, which was probably a mistake; this was also religion oriented, with the two almost always sharing one. This leads into Brennus' other key strength: everybody loved him! Between peaceweight and religion, Brennus faced only 1 defensive war for every 2 games, and he was even safer than Kublai from attack, despite his central position. Overall, though there was some variance, not declaring at pleased seemed like a strength here - most notably in both of Brennus' wins, which saw him stay out of inconclusive fighting to establish clear tech leads. This was a pretty schizophrenic performance, and I'm still not sure if Brennus is actually a good leader or not.

Pacal of the Mayans
Wars Declared: 27
Wars Declared Upon: 29
Survival Percentage: 50%
Finishes: 1 First, 3 Seconds (11 points)
Kills: 5
Overall Score: 16 points

Pacal was a truly feast-or-famine leader here, but in the Gandhi mold rather than the Julius Caesar one. Frequently, Pacal would be attacked early on, and this would basically end his game. In particular, Kublai Khan was able to completely solo-kill Pacal in a full quarter of the games, with Victoria and Montezuma also doing considerable damage whenever they attacked. Pacal's expansion was responsible for this, as he consistently settled few cities and struggled to claim his fair share of the map. However, when he was left alone, Pacal was able to become a technological powerhouse, scoring a hodgepodge of kills and second places, as well as the only Cultural victory of the Alternate Histories. These games were outliers though, and Pacal's low performance (relative to the Khmer and Mongolian giants) is particularly embarrassing considering the diplomatic free pass he often received, fighting the fewest wars in the Alternate Histories. In particular, Financial/Alive was not as strong a trait combination as we've come to expect, with Pacal surviving in many games where he was still unable to place. Pacal was basically Gandhi-lite here, as he displayed all the peacenik weaknesses while generally unable to capitalize on the diplomatic good fortune his peaceweight afforded, barring a few exceptions.

Victoria of England
Wars Declared: 27
Wars Declared Upon: 63
Survival Percentage: 20%
Finishes: 1 First, 1 Second (7 points)
Kills: 5
Overall Score: 12 points

Victoria was the strongest of the high peaceweight leaders on this map, although that isn't saying much. This was the result of a beautiful starting position that was spacious, fertile, and isolated from enemies. She was generally one of the leaders coming out of the landgrab, alongside Kublai and Brennus, and would typically be strong until around Turn 200, when Joao and Lincoln had been finished off and the warmongers would come for her next. She was able to pull out a single win in Game Four, an outlier game where she was largely left alone, and she was able to solo-kill Pacal. Otherwise though, the above pattern played out in virtually every game, and Victoria was totally unable to find the success she experienced in the Real Wildcard Game. This was very disappointing, considering her starting position that basically screamed 'winner!' The other thing of note about Victoria's performances was that she fought with Lincoln far more than expected, in something like a third of the games. This was usually a lose-lose scenario where both leaders were slowed down and less ready to face the incoming Aztec/Khmer/Mongolian/Mayan/French/Celtic/Ottoman hordes. However, it did suggest that perhaps a high peaceweight bond isn't as much of a shield as we tend to think, which has some interesting implications for the future.

Suleiman of the Ottomans
Wars Declared: 41
Wars Declared Upon: 21
Survival Percentage: 55%
Finishes: 0 Firsts, 0 Seconds (0 points)
Kills: 11
Overall Score: 11 points

Suleiman was the leader I found myself rooting for the most across these games - although perhaps this cursed him. Sulei was handed an extremely rough start, without much land and thoroughly buried in jungle tiles. He also neighboured the two strongest leaders on the map, Kublai and Suryavarman. Despite this, in the Real Wildcard Game, he only barely missed out on second place thanks to an unfortunate late war; this pattern would be repeated throughout the Alternate History, with Suleiman handing in a series of strong performances against all odds, only to find himself with zero placements across twenty games. Suleiman's best game was probably Game Two, where he navigated a near solo-conquest of Joao into becoming the games dominant leader, only for second- and third-place Kublai and Suryavarman to tear his empire to the ground. He had several near-misses though, such as Games 8 and 16. Suleiman's biggest problem, other than the aforementioned start, was production - once he got the massive jungles cut down, his cities were in beautiful grassland valleys that lacked badly for hills. This meant that he would lose 1v1 wars even at tech parity, which happened repeatedly against Suryavarman, Kublai, and Brennus, and was responsible for almost all of his deaths. Still, Suleiman tried his best to escape these issues, fighting many wars and scoring many (deserved) kills; he was almost certainly unlucky here, and this was a tragic series of games for the Ottoman sultan.

Washington of America
Wars Declared: 27
Wars Declared Upon: 51
Survival Percentage: 25%
Finishes: 0 Firsts, 1 Second (2 points)
Kills: 8
Overall Score: 10 points

Napoleon of France
Wars Declared: 37
Wars Declared Upon: 26
Survival Percentage: 60%
Finishes: 0 Firsts, 2 Seconds (4 points)
Kills: 6
Overall Score: 10 points

I'm presenting the Eastern duo together because their performances were intrinsically linked, even ending up with the same scores. Washington and Napoleon fought in basically every game, and this was nearly always a death struggle. Brennus intervened surprisingly little in this squabble, and as their only other neighbour, these two were generally undisturbed by the rest of the map for hundreds of turns, in a fascinating situation. In this mortal duel, there could be three outcomes. The first (and most common) was a stalemate, for which the outcome was clear: better leaders who'd done bigger things would clean up Washington in the late-game, with Napoleon remaining an also-ran. Less commonly but more interestingly, one of these leaders would solo-kill the other and go on to be a major power. Napoleon was able to kill Washington slightly more often (for example Games 3, 8, and 10), and this combined with the first outcome was responsible for his much higher survival rate. However, Washington's duel-wins were more impressive, especially Game 16 where he went on to conquer the entire Eastern half of the map, including Suryavarman, before finally losing to Kublai in the very late-game. Overall, these leaders were very unlikely to achieve anything of consequence - they could only achieve things by killing each other, and were too evenly matched to do so reliably.

Montezuma of the Aztecs
Wars Declared: 45
Wars Declared Upon: 44
Survival Percentage: 20%
Finishes: 0 Firsts, 0 Seconds (0 points)
Kills: 6
Overall Score: 6 points

Montezuma was in classic form here, flailing around attacking anybody and everybody. He was a complete disaster in the early turns, getting surprisingly ruined by barbs in most games, and often building wonders instead of expanding. This meant that by the time Monte was able to declare wars, he was already behind most leaders, and he was rarely able to make any progress against Joao, Lincoln, Pacal, or Victoria (his most common targets). This can be seen by the war counter, where Montezuma was surprisingly just as likely to be attacked as to attack someone himself - a bad sign for the most aggressive leader in Civ 4! Most of Monte's kills were outright stolen, including two(!) incidents where Brennus nuked the final city of an opponent, before Montezuma marched some obsolete unit in for the kill (Games 1 and 6). This means he was arguably quite lucky to score even this paltry sum. Monte's only real strong game was Game 14, in which he conquered most of Lincoln and Victoria himself, before dying miserably to a Kublai Khan attack - this was the high watermark, and Monte's game were universally total failures.

Lincoln of America
Wars Declared: 9
Wars Declared Upon: 64
Survival Percentage: 15%
Finishes: 0 Firsts, 1 Second (2 points)
Kills: 1
Overall Score: 3 points

Lincoln had a very rough series of games, even if he wasn't quite as pathetic as in the Real Wildcard. Lincoln's territory was some of the most barb-infested on the map early on, which always left its mark on the poor Western Americans. Even in games where he managed to settle lots of cities, they remained small in size, with few improvements or buildings. As mentioned, Lincoln also fought a lot more with Victoria than expected, which either dragged them both down or led to Lincoln's death, but almost never helped him. Lincoln's one good game came in Game 12, a game that was unusually stalemated early, where he was able to essentially masquerade as a low peaceweight warmonger, eating Victoria and being saved from Kublai Khan by Suryavarman. This was a truly bizarre result though, and Lincoln was probably lucky to place even once...

Joao of Portugal
Wars Declared: 9
Wars Declared Upon: 55
Survival Percentage: 0%
Finishes: 0 Firsts, 0 Seconds (0 points)
Kills: 0
Overall Score: 0 points

...but at least he survived a few times, which was better than poor Joao. The Portuguese were first to die more than half the time, which was a pretty impressive feat in a game with 11 players. This was almost entirely the result of a jungle-heavy central start, surrounded by low peaceweight warmongers who were the top leaders on the map (although it's hard to know if this is because they started next to Joao). Joao was also hurt quite badly by the removal of Deity starting techs. Joao also struggled badly with a failure to research Mysticism, which kept his production capabilities far lower than they should have been in many games. To his credit, Joao did have a few strong games where he remained near the top of the scoreboard early on and settled a lot of cities, but it always fell apart in a vicious dogpile. Joao was frequently attacked by Kublai, Suryavarman, Suleiman, Montezuma, and even Brennus. This was often how Suryavarman and Kublai snowballed their already-strong positions, and it was also where Suleiman scored the vast majority of his kills. In sum, Joao acted as a speedbump for Kublai and Suryavarman en route to their victories, and the Real Wildcard Game was probably the high watermark here.

Conclusions

Unlike many of the other games I've run for Season Three, this game turned out to have mostly followed the status quo. Kublai wasn't the most likely winner, but he definitely proved that his victory wasn't a fluke; the same was true of Pacal for second place, though it would have been more fitting for some combination of the top three to advance. Our real result was certainly better than the contemporary alternatives of Victoria and Suleiman! The biggest outlier seems to have been first-to-die, where Joao was a strong favourite who performed very unusually well in the real game. However, this may be explained by the removal of Deity starting techs, which likely hit landlocked Joao very hard on this map relative to the other leaders. So far though, this has been by far the least-changed game by the techs' removal, which was interesting considering the higher number of leaders. Perhaps the addition of Raging Barbs slowed everyone down so much that the techs didn't matter as much.

Cheers ~ Amicalola