Board 8 > The 128 Greatest GameFAQS Contest Matches of All Time - The Top 64

(+) Yeah! (+)

23. Mega Man vs. Ms. Pac-Man (2002) R1

https://gamefaqs.gamespot.com/a/forum/e/e45953d3.jpg

Mega Man 92.3% 63039
Ms. Pac-Man 7.7% 5257
TOTAL VOTES 78743
https://board8.fandom.com/wiki/(6)Mega_Man_vs_(11)Ms._Pac-Man_2002
https://gamefaqs.gamespot.com/poll/968-west-division-round-1-mega-man-vs-ms-pac-man

Can the honor of the greatest ever contest debut really be held by an entrant in a contest where *everyone* was making their debut? After all, when everything is unexpected, and the entire concept of a character contest is unknown territory it can be a little harder to make a dramatic first impression. Whats good? Whats bad? These are open debates which the contest is, in theory, partly about settling. It takes a lot to stand above the rest in these circumstances. Despite all that however, if I had to give an answer as to what the greatest ever contest debut was, I would be hard pressed to come up with a better answer than Mega Man. Most of the Noble Nine would need some time to burnish their legacies; Mega Mans would be established on Day One.

Despite the overall weaknesses of the bracket in 2002, it took a while for the blowouts to really get started. Due to some questionable seeding decisions, the contest elites were thrown halfway decent entrants in the first round instead of contest fodder, and so it took two weeks for Knuckles, of all characters, to become the first entrant to ever break 80%, barely doing so in a SFF fest with Akira Yuki. But instead of ushering in a string of blowouts, the contest continued on much the same as it had before, with numbers in the mid to low 80s seeming to be the upper limit of what a character could do. Link did break 84% in an SFF match of his own, but even Sonic struggled to do much better than low 80s again Pitfall Harry, hardly anyones idea of a beloved favorite.

Then came Mega Man Vs. Ms. Pac-Man. Well, first came the picture for Mega Man Vs. Ms. Pac-Man, which after a public outcry would become the first match picture to ever be changed. I honestly have no memory of this incident, only after the fact comments, so I dont know what the initial picture was. Im guessing Mega Man had MM1 box art as a joke? Not sure what else would cause that reaction. In any case, even after the pic was changed, we still werent prepared for what was coming. If we had thought that 85% was the upper limit for what a character could pull off in a match, Mega Man would prove us completely wrong, blowing past that marker entirely en route to a 92% victory over Pac-Mans female counterpart. What we were seeing was something that just 24 hours prior we had thought to be near impossible. A character was actually breaking 90% in a contest match, and aside from the entertainment of the currently ongoing match, that result had implications for the entire contest. Mega Man, once a step down from the Marios and Sonics of the world in the boards estimation, was now right there on their level, one of many potential winners in a frequently shifting contest landscape.

And then, in Round 2, Mega Man would go out and do it again, breaking 90% for the second match in a row, this time against Serious Sam, setting a Round 2 record that has never been broken. Mega Man was no longer just one of the stars of the contest. After the one two punch of his first two rounds, he was *the* star of the contest, the one whose exploits every other character was trying to replicate. It wasn't that people entirely trusted that Mega Man could pull off the contest victory, but the results spoke for themselves. He had quickly turned into the one to beat, and expectations were sky high for his upcoming match with Sephiroth, one of the only characters who had so far come close to matching his apparent strength. Ive written previously about how that match turned out, but despite the Blue Bombers relatively early exit, his legacy remained intact. Mega Man would go on to make the Final 4 for the next three years straight, but despite reaching that height of contest achievement, his most impressive year would remain his first one, at least if the board's reaction to him is any indication, and those later Final Fours were, in a way, justice for what could have been back in 2002.

Going into 2002, it was taken for granted that their iconic status would put characters like Link and Mario and Snake and Sonic above most of the rest, and even a more shocking elite like Crono could be explained by the sites demographics. But Mega Man? Sure hes an icon, but not on par with, say, Mario, and Mega Mans games, while decently popular on the site dont seem enough to make up the difference, either individually or in the aggregate. But despite all this, his popularity in 2002 *was* on par with Mario, or at least thats what it seemed. I dont think anyone would have seen that coming in June of 2002, but Mega Mans strength would end up being one of those shocks from 2002 that we now take for granted. That, along with an entrants ability to top out at over 90%.

22. Link vs. Crono vs. Solid Snake vs. Cloud Strife (2008) R6

https://gamefaqs.gamespot.com/a/forum/1/16b4189b.jpg

Link 33.68% 49345
Crono 12.23% 17916
Solid Snake 27.76% 40663
Cloud Strife 26.33% 38581
TOTAL VOTES 146505
https://board8.fandom.com/wiki/Link_vs_Crono_vs_Solid_Snake_vs_Cloud_Strife_2008
https://gamefaqs.gamespot.com/poll/3306-contest-final-link-crono-snake-cloud

When it comes to our Contests, Finals matches have a bad reputation. And not just because one series tends to always emerge victorious in them. For one, Finals are known for being boring. The chances of the two strongest entrants (or at least the strongest entrants in their respective halves of the bracket) in the contest just happening to be close enough in popularity to produce a legendary, nailbiter match are not particularly high. And its true that weve only had one or two finals that were decided by margin of 2% or less. Almost more of a drag than being boring though, is being predictable. And once again, I dont just mean because of the LAW. Not only are most contest winners obvious, the runner up is usually what is expected as well, the whole contest being a buildup to see these two entrants duke it out in the end. Maybe theres a healthy debate between two or three options for 2nd Place, but very rarely does something come out of left field to upset that calculation, especially in the absence of an offsite rally. Say what you will about the four way contests, but they upset that consensus. 2007s Final will be coming up later, but first up is the Final Match of 2008.

The big question that loomed over the Finals that year was L-Block. Would it repeat its journey to the finals from 2007, and if so, would it able to triumph yet again? This question ended up being answered a round or two earlier than we all thought, and after its shock boot at the paws of Pikachu, any serious debate around the Finals seemed to end. We would almost certainly be getting the Link > Cloud result that we had predicted in 2007 as well, but which had gotten sidetracked by L-Blocks run. There would be some minor upsets along the way, but for the most part the endgame for 2008 appeared to be falling into place right on schedule. Snake had managed to slip by Cloud into First Place in the final Semifinal, but he had barely done so, and only because Sephiroth was in the poll as well. Surely that was the last excitement wed get for the contest however, and once Sephy was removed, things between Snake and Cloud would revert to normal, and we would get the contest ending that we had all come to expect. Right as we began to settle our brains for a long winters nap however, the unexpected would happen, just one more time.

Buoyed by his upset of Cloud the previous round, Snake started off strong, going toe to toe with Link in the polls initial minutes. No real surprise there, and equally unsurprising, Cloud started off far behind in a distant third. As the minutes ticked by though two things became clear. First of all, Clouds numbers against Snake were bad. Like really bad. It was one thing for Cloud to fall behind in the early going, and it was another to fall this far behind. One hour in, and Cloud was still closer to Crono than he was to Snake. And as time went on, and Cloud only gradually recovered, one thing became clear: he wasnt going to be able to come back. The hole he was currently in was just too big, and based on the trends from the previous day, if Cloud did eventually start coming back he would never be able to do so by enough to actually overcome Snake. This was stunning. Cloud finishing ahead of Snake in this match was considered so obvious that people had already started to congratulate the prospective contest winners as the previous match drew to a close. What would transpire the following day was considered irrelevant. Those congratulations would prove misplaced though as, for the second year in a row, Cloud would end the Final Match in Third Place, a position which yielded his supporters no points whatsoever.

But that was only part of what Snake was pulling off in the early going of this match. As I said earlier, he was going 50/50 with Link in the first minutes, but as an underdog in the Finals, against Link himself, that was expected. But then, the poll updated, and the match stayed 50/50. And it updated again. And once again, the poll stayed 50/50. This was not some flukey Board Vote that Link would spend the rest of the Night coming back from; this was real, honest to goodness (albeit momentum fueled) strength. And for the first, non-rally fueled, time in five years, Link had a match on his hands. As Cloud surged back overnight, Snakes percentage would start to drop, but Link was right there with him, dropping in percentage too, the two of them even falling off in unison. While Link had managed to take the lead for an update or two in the opening hours, as the poll moved solidly into the Night Vote, so did Snakes hold on the lead. It remained very tenuous, Snake never ahead by more than 100-150 votes, but a lead is a lead, and Snake was determined to drag his out as long as possible. The character who several contests ago had struggled to put away Frog was now going toe to toe with Link. This was the strongest Snake would ever look, one of the only times a non-Clinkeroth character ever looked to be on that level, and for six hours the contest order was on the verge of being upset.

And then the Morning Vote came in, Link ran away with it and ended up winning by well over 8,000 votes, with Snake having to settle for Second Place. Cloud would put up an extremely spirited comeback attempt for someone in such a hopeless position, but as previously stated, he would have to settle too, in his case for Third Place. For the second year in a row, we would be denied the Link > Cloud final we had all come to expect, and while Link would still win, as expected, the journey to get to that point over the last 24 hours had been anything but expected. Aside from rallies, there has only been one moment of weakness from Link over 20+ years, a match which is coming up later, but for a quarter of the Finals in 2008, we thought we might have a second. Snake might not have slayed the giant, but, for possibly the last time in history, he had shown that he was mortal.

It's a shame Draven robbed us of a chance to see Snake beat Link in 2013

Leonhart4 posted...

It's a shame Draven robbed us of a chance to see Snake beat Link in 2013

Story of GameFAQs contests since like 08. Rallies take away all the genuinely interesting internal results

I say this as someone who has over the years come to appreciate L-Block despite hating it at the time, but all rallies/cheating since have just existed to no fun zone the cool other things going on in the bracket.

Yeah, I'm in the same boat as you. I hated L-Block at the time, but I appreciate its run now in hindsight.

All other rallies have sucked.

...I was one of those few Ms Pacman voters. Voted more based on how much I liked the games than anything else. I was shocked at the size of the blowout then and even now it seems like it shouldn't have been that high.

We know Pac-Man folds like a house of cards against seemingly ANY recognizable icon, so it's hardly a surprise for that to be the case with Ms. Pac-Man.

Damn every time i remember to check this thread again it reminds me how insanely hype these things were until the outside rallies like Drave became a thing.

I don't think rallies killed the contests per se, because the votal decline was inevitable either way, but they did sully 2013 and 15. Leaves a bitter taste because those were the last two before that decline really started to kill.

That said, I did still find a lot of enjoyment in 2018. But it wasn't really the same.

2018 was a fun contest

21. Crono vs. Missingno (Winter 2010) R1

https://gamefaqs.gamespot.com/a/forum/1/1adaaee2.jpg

Crono 49.71% 45277
Missingno 50.29% 45804
TOTAL VOTES 91081
https://board8.fandom.com/wiki/(4)Crono_vs_(13)Missingno_2010
https://gamefaqs.gamespot.com/poll/3789-jenova-division-round-1-crono-vs-missingno

Should a loss in a 12 hour match count as breaking the Noble Nine? Just as we were quick to answer a similar question when 4 way contests were announced (No, it doesnt), we were equally quick to answer it when 12 hour matches first came on the scene (Yes, it does). Whether we were quick because our previous confidence in 4 way screwiness hadnt been misplaced, or because we were desperate to have the Noble Nine break officially, we might have been a little too hasty in our analysis, done as it was before a single 12 hour match had even taken place. Sure, 12 hour matches arent as abnormal as a 4 way match is, but over several contests, I think we learned that they dont really give a complete result either. Both AM and PM matches retained far from 100% of the votes that a 24 hour match did (people speculated that PM matches had ~70% of the votes from a 24 hour match, and that AM matches had just under 60% or so, I think), and those votes werent distributed proportionally to what a 24 hour vote would have been either. AM matches particularly would favor characters with strong Night Votes, and PM matches, while not as extreme as AM ones, would disfavor said characters. But at the start of 2010, we didnt know any of this, and a decision was made, a decision which would come into play much sooner than anyone had anticipated.

Missingno was much like L-Block, not just because they were both joke characters that threatened to derail an entire contest, but because they both kind of came out of nowhere to do so. L-Block wasnt some oft talked about joke entrant that wed been dying to get into a contest for years, and neither was Missingno. Even in the heady days after Pokemon R/B/Ys strong showing in the 2009 Contest, when every Pokemon under the sun was discussed as a new entrant, I dont recall Missingno ever coming up. I cant remember anyone expressing any interest in seeing it in a contest until it showed up on Day 1 of the Vote-Ins for 2010. Missingno would win its Vote-In with ease, even despite the inclusion of two other Pokemon in the poll. This should have been a sign, but as Vote-Ins were a brand new concept for us, its performance here went mostly unremarked upon. And once the bracket was released, Missingnos potential, to the extent that we thought the glitch had any, was noticeably wasted, being forced to go up against Crono in Round 1. Crono himself would have an unremarkable path laid out in 2010, set to win two matches before going out to Sephiroth in the Third Round. Yoshi might put a scare into him in Round 2, but for a character long thought to be the one to break the Noble Nine, his path was considered a wasted opportunity.

While many groundbreaking matches in contest history tend to have a bit of foreshadowing leading up them, aside from the Vote-In mentioned above, there really wasnt anything to cling onto here or anything that implied anything more than the 60/40 Crono victory we were all expecting. Pokemon was having a strong contest and Chrono Trigger a somewhat weaker one, but nothing out of the ordinary or extreme enough to set off any warning bells. Yoshi beat Jak very solidly in the AM match that day, and then, at 12:00 PM, on February 14th, Valentines Day, the breakup of the Noble Nine got underway. While Missingno would do very well in the first seconds of the poll, Crono would be the one in the lead at the freeze. Shockingly, he was leading by a single vote, and while it was clear that Missingno would be stronger than we all thought, there was no reason yet to see this as anything more than the typical Joke Fueled Early Vote. That single vote lead, it turned out however, would be the only one Crono would get for the rest of the match.

When the first update hit, 10 minutes in, it became clear that the Early Vote had not been favoring Missingno, it had been favoring Crono. Missingno wasnt just leading, it was leading by 95 votes! And while its percentage would stabilize at just over 51%, its lead would keep going up and up over the first several hours of the poll. 100 votes, 200, votes, two hours in and Missingnos lead had crested 600 votes. It didnt take long to discover that our increasingly common nemesis 4chan was responsible for hundreds of those votes, if not the lead entirely, but for Crono/Missingno to be close enough for such a rally to even matter was truly shocking. As a 12 hour match, rallies had a greater impact obviously, but this would be the most popular 12 hour match of all time, pulling in 91,000 votes when all was said and done. That should have been enough to stop something like this from happening. Was this how the Noble Nine would end? Not with one of the expected breakers like Vincent or Ganondorf, but with the help of 4chan, and with a literal glitch?

Not if Crono had anything to say about it. Finally, two and a half hours in, Crono would start fighting back, picking at the lead here and there, but as is often the case whenever someone tries to fight back against an ongoing rally, any gains that Crono made, any momentum he could build up, was countered by a newfound intensity from the forces offsite. By 7 PM, with only five hours left in the match, Crono was right back where he had started, the lead still far over 500 votes. This time though, Crono was able to start a comeback with some juice, something that might actually go somewhere. Over the next two hours, Crono would take 400 votes off the lead. All he would need was one more hour or so at the same pace, and he would be able to take back the lead. One hour was all he needed, and Crono had three hours left to work with. Unfortunately, that run from 7-9 PM would prove to be Cronos last, as he stalled out for the next hour or two, and then saw any remaining momentum overrun by renewed rallies from offsite. Cronos initial lead would remain his only one, and Missingno would win the day, or well, the afternoon, to be more precise, stunning us all.

The unsatisfying nature of the result led many to grumble about whether or not the Noble Nine really did break with this match, and while, as I mentioned earlier on, there was a point to be made about the vagaries of 12 hour matches, it was hard not to see the complaining as a bit of sour grapes. Not only had the Noble Nine lost, it had lost to an outsider, not someone that anyone could get the bragging rights about predicting. It was both a climax and an anticlimax at the same time, not the way that things were supposed to be. However, more than anything, it was history, history that had been made in the most bizarre way possible. After 8 years, and seven and a half Character Battles, the Noble Nine had finally fallen.

To a rallied meme character in a 12 hour match? nah. Still don't consider this as a "real" loss.

MAJOR Crono fan. I loathed this match.

This was a harbinger of things to come for Pokmon, and it all started with a literal jumble of pixels.

Lame result and essentially a fluke based on what Missingno did in 2013

Ive honestly hated all of these non-characters post-2007. L-Block was kinda funny as a novelty but after that it just kinda became sad.

How many non-characters are we talking about?

UltimaterializerX posted...

I cant look it up on the guru site, but I think I picked missingno to win? Maybe?

Your Oracle prediction was Crono with 54.11%

Only BeTheMan and GameBopAdv had Missingno in the Oracle

I think Palmer was one of the only people to predict Missingno. Or maybe I'm mixing that up with Charizard.

Every time I think of Charizard it frustrates me because I was tied with Palmer for closest to having that mess of a division perfect in matches right, and I almost picked Charizard over Bowser and would have if I had known HG/SS released that day. Palmer missed a round 1 match, predicted Missingo correctly and won the contest (but got no prize lol) so of course it was deserved. But dammit I completely nailed the strength of everything in the toughest division that year.

L-Block, Missing No, Weighted Companion Cube. Probably one or two others.

? Block

Way bigger fan of Seph vs MissingNo. People were so deflated and were thinking a joke character was gonna run again and Seph is like "nope"

20. Aya Brea vs. Donkey Kong (2002) R2

https://gamefaqs.gamespot.com/a/forum/5/576398e5.jpg

Aya Brea 48.94% 30478
Donkey Kong 51.06% 31798
TOTAL VOTES 62276
https://board8.fandom.com/wiki/(5)Aya_Brea_vs_(4)Donkey_Kong_2002
https://gamefaqs.gamespot.com/poll/973-north-division-round-2-aya-brea-vs-donkey-kong

What is the appeal of the Contests? Is it just simply about predicting things, guessing whats popular, etc., and the matches themselves are secondary? Prior to the first Character Battle in 2002, contests were a regular feature on the site, but not the kind of capital-C Contests that we all know and love. Several times a year, every two or three months, Ceej would have a competition for the sites members, and these could be any number of things, ranging from having to predict a game/character based on a partial or manipulated screenshot, predicting the 10 most visited FAQS of the year, or even predicting the results of a demographic questionnaire posed to site members. Im no mind reader, but Im sure that as Ceej was putting together the first Character Battle, he considered it of a piece with these other competitions. A more expansive competition, sure, (he knew enough to create a board to discuss it right away), but something that was at least roughly in line with those other contests. Fun diversions to pass the time and see how much you knew about your fellow users. That was how it had been before, but the first Character Battle would change everything. We would have two other small competitions in that earlier vein later in the year, but aside from the 10th Anniversary Contest in 2005, it was Bracket Battles from here on out.

Once again, I dont know how much of what followed Ceej had foreseen, but it strikes me that the Contest being structured the way it was was no guarantee. There was no reason we *had* to see the polls play out live. We could just find out the result after the fact, or maybe every participant in the contest could fill out their votes along with their prediction, the same way that similar contests had worked. The contest was structured the way it was however, and while it was obvious from the very first match that there was something special with the Character Battle, it would take a while before a match came along that showed what more was possible from these things besides just predicting results. The First Round had its share of excitement, but the individual matches themselves werent anything that required a serious investment of time. You would look at the results, and as surprising as they might be, you would go Oh, thats an upset! or Oh, thats a big blowout! and then briefly discuss the implications of said result with other users for a while before going on with your day.The results themselves were the interesting part, not so much the journey to get there. Despite the desperate pleas for the kiddies to wake up, if there were any kind of drastic vote shifts that occurred in this first set of matches, they went largely unremarked on, and surprisingly, there was not a single match in Round 1 that featured a comeback, or was anything that might be considered a nailbiter. The closest match was Strider/Raziel, which, while a relatively narrow result, never became a 50/50 affair.

Round 2 started off with Mario/Morrigan which played out along these same lines, but then finally, with the second match of the round, we had something new on our hands. Donkey Kong was briefly ahead at the polls beginning, being a massive bracket favorite after all, but after the initial rush of votes, something shocking happened. Aya took the lead. Someone who many thought wouldnt make it out of the First Round was going toe to toe with a gaming icon. And as the Night Vote kicked in, Aya built that lead up, gaining around a 1,000 vote lead on DK by daybreak. We had seen some upsets in Round 1, but this one topped them all. But then, as the Day Vote began, something even more unprecedented happened: Donkey Kong started tocome back? Was such a thing possible? Just you wait til the kiddies wake up had become a joke by this point, and while it might not have been enough to redeem Spyros disaster in Round 1, the overall concept of Vote Trends was turning out to be quite real indeed. DK would tear into Ayas lead as the morning went on, eventually overtaking her and building up a lead of his own.

It was around this time that I, and I imagine a lot of other board regulars, logged on for the day and saw what was going on. Keep in mind that at this point in time, outside of a few obsessives, people werent maniacally tracking each poll from its very start. Not only because in a pre-smartphone, heavily pre-broadband era it wasnt customary to be online 24/7, but also because, prior to Aya/DK, there was no reason to be following along with a match in real time, let alone from the very beginning. Realizing that not only was there a 50/50 match going on but that the balance of power in it was shifting right in front of my eyes,I joined the rest of the site in a brand new practice: mass refreshing the poll as often as possible to see what the results were and where they were trending. Our eyes remained glued to the poll, and for the first time in contest history the way the poll played out generated more attention than the result itself.

Donkey Kong would eventually run away with the Day Vote, putting the match out of reach by the afternoon, but the drama wasnt over yet, as Aya would go on a 1,000 vote comeback of her own in the polls final hours*. Impressive, but ultimately not enough, and Donkey Kong would limp into the Sweet Sixteen. Aside from being one of the definitive Only on GameFAQS results (where else but on this site in 2002 is this a nailbiter), as well as establishing DKs bonafides as a choke artist, this match would be one of the most foundational ones in contest history. Every moment spent sitting in awe of the Day (or Night) Vote, every stomach turning bout of anxiety before refreshing a poll, every back of the envelope calculation to see if a comeback is still viable. They all come back to this match.

*I was able to find contemporaneous commentary from two weeks after this match in the one archived Stats Topic from 2002 where a user states that Donkey Kongs lead got up to 3,000. That would put Ayas comeback at around 1,700 which considering that the match only got 62,000 votes seems quite hard to believe, especially if it came right at the end, when votes should be slowing down. Looking into things, it appears as if Parasite Eve was never released in Europe, which actually would partly explain such an extreme vote shift, but I still find the scale of those numbers questionable.

I miss the days when a niche Square character could hang with a Nintendo icon

UltimaterializerX posted...

Its my go-to example when people claim rallies always ruin contests.

Yeah, you still don't have a leg to stand on by claiming that.

I mean

That Sephiroth was thought at risk of losing at all kinda is evidence rallies do ruin contests.

I think if you had more moments of resistance like that it'd be cool but it's like rallies 25 GameFAQs 2 or something. It's very rarely actually interesting.

19. Donkey Kong vs. Duck Hunt (Sp2004) R1

https://gamefaqs.gamespot.com/a/forum/d/dac50a21.jpg

Donkey Kong 50.07% 48587
Duck Hunt 49.93% 48444
TOTAL VOTES 97031
https://board8.fandom.com/wiki/(6)Donkey_Kong_vs_(11)Duck_Hunt_2004
https://gamefaqs.gamespot.com/poll/1605-division-8-round-1-donkey-kong-vs-duck-hunt

I know the complaints coming about this match already. That there were no stakes. That the contestants were fodder. That this was just a pointless Round 1 match whose winner was always fated to get completely blown out the following round. All those reasons are *exactly* why it is so high on this list. The fact that such an objectively irrelevant match could produce the result it did is what makes this match so special. With the exception of Draven/Undertale matches, there is only one match in contest history to feature a 3,000 vote comeback. Its not a match featuring the Noble Nine or even a pair of Near Elites, its this one. Donkey Kong and Duck Hunt, two fodder games, who, in theory at least, share the same fanbase managed to produce a comeback which to this day has yet to be topped. Theres no greater contest narrative that this match ties into, no implications for anything else. Its just a pure diamond of a match, an objet dart.

While the hype didnt extend much past the match itself, there was still a decent amount of hype for this match going in. Classification-wise, it falls into a group I call Those Matches, 1 point, Round 1 matches that split the board almost evenly, and which actually do end up being relatively close in the end. Everyone knows that the winner will lose in the next round, but these matches inevitably take on outsized importance in board discussion, something about the raw simplicity of a 50/50 split with no past or future complications generating a higher than expected degree of tension. By the time you get to the big name clashes in later rounds, the data points are numerous enough to generate a narrative that can either be accepted or refuted in a couple of sentences, but theres something about the black box nature of a Round 1 tossup that causes people to generate reams of speculative analysis from the smallest of minutia. And there was certainly plenty of analysis spilled on this match. Duck Hunt was a game that everyone had played, but did anyone actually care about it? Donkey Kongs status in gaming history was legendary, but was it too old-fashioned for modern day voters to support? These questions divided the board, and going into the match it was considered anyones game.

The match started with a decently close result, the Board Vote not seeming to favor any one game, much as youd expect in a closely debated match such as this. As the First Night Vote started to kick in however, Donkey Kong began running away with things, and the dream of a nailbiter began to fade further and further away. While it would take 4 hours for Donkey Kongs lead to break 1,000, it would take only two more for it to break 2,000. Donkey Kongs overall percentage broke 56% around this time, and stayed at that number all throughout the morning, its lead frequently increasing by triple digits in the 15 minute updates that were customary at the time. Duck Hunt had looked like it could keep up early on, but those days were long gone at this point. Donkey Kong slowed down a little as the day went on, but even so, its lead would break 3,000 votes just past noon, and would keep rising afterwards. This was a number far removed from what we had seen in any previous comeback, and there was very little doubt in anyones mind that Donkey Kong would not keep continuing on the way it was.

At 1:45 PM EST, Donkey Kongs lead would reach its peak: 3,251 votes. With one exception, it had increased that lead with every single update for the past 11 hours straight. And even after that one cut an hour previously, Donkey Kong had bounced right back and kept increasing as steadily as before. At 2:00 PM however, Duck Hunt would make a 45 vote cut and the match would change instantly. Donkey Kong would manage to barely make a gain the following update, but from then onward, it would be 9 straight hours of Duck Hunt gains. If there is such a thing as a mathematically perfect comeback, this was it. Look at later matches like Halo/Starcraft or SoulCalibur/Kingdom Hearts. There is an hour or two of back and forth when vote shifts happen, or when comebacks begin. In Donkey Kong/Duck Hunt, there was none of that. It was as if a switch had been flipped at the halfway point of the match and all of a sudden the polarity was reversed. One update Donkey Kong was the one gaining and gaining, and by the very next update that role had shifted to Duck Hunt.

While one or even two cuts might be a fluke, it was immediately clear that this was nothing of the sort as the ASV (not that we called it that at the time) began in earnest and Duck Hunt started to tear into the lead in update after update. Initially taken by surprise, once we realized that we had a comeback on our hands, the board got straight to work doing what we do best, crunching numbers. While a 3,000 vote comeback is an extremely daunting figure, the poll trackers made their calculations and determined that if Duck Hunt kept up its comeback at its current pace, it would manage to regain the lead.right as the poll was ending. Time passed, the comeback continued, and Duck Hunts pace was recalculated. Now it was projected to take back the lead.right as the poll was ending. The only thing more perfect than the comebacks immediacy was its consistency. No matter how much the total incoming votes ebbed and flowed, Duck Hunts cuts would match the input perfectly. After every update was posted in the Poll Update Topic, there was a note saying how much of a gain Duck Hunt would need on the next update in order to stay on pace to win. Duck Hunt would hit that mark with a freakish exactitude almost every time. It was on pace to win, but it was going to be a photo finish. As afternoon turned into evening, incoming votes slowed, meaning that if Duck Hunt wanted to keep up its pace it would have to start gaining a larger and larger percentage of the incoming votes. Well, what did it do? It started to gain a larger and larger percentage of the incoming votes.

As the night wore on, the board crackled with an energy that was normally reserved for the battles of the Noble Nine. But the Noble Nine, for all their strength, had never done anything like this. Theyd never taken a 3,000 vote deficit, and on a dime, started to turn it around. They had never been propelled with this kind of unstoppable momentum for this long without missing a beat. The only entrant to ever show that kind of skill was, well, Duck Hunt. We werent entirely sure what was going on, but it was hard to not accept that Duck Hunt was on the verge of brute forcing its way to victory. And it almost did. Duck Hunt was able to near perfectly keep up its pace until the final minute, but near perfect was not quite good enough, not when youre behind by as much as Duck Hunt was. Donkey Kong would manage to win three updates in the final hours of the poll, which was not much, but just enough to barely hold on. When all was said and done, Duck Hunt had managed to come back by 3,108 votes in a match without major rallies, more than any entrant would ever do before or since. It had not won the match, but it had set a record that would never be broken.

I remember checking the result in the morning, seeing DK had the match in hand, and not thinking anything of it again until the next day when I saw Duck Hunt had almost won.

DK vs. DH was an awesome match. I remember being super nervous as the comeback took place. Definitely deserves a high ranking. I didn't realize it was the largest comeback in history outside of rallies. Maybe it was this match that prompted people to start tracking updates?

Yeah I'm definitely a hater. Did not care about that match. But I also didn't as avidly follow contests until late sp2004 (Starcraft beating Windwaker made me a fan for life)

just a quick headsup, I prob won't be able to archive the topic until Monday due to holiday travel, so be careful with the ration of remaining posts until then >_>

Luckily, there is a site that does that for you (also even if the topic were locked right now it wouldn't purge until around the 31st at this point)

Bump

Should be back home tomorrow evening.

18. Starcraft vs. Super Smash Bros. Melee (Sp2004) R4

https://gamefaqs.gamespot.com/a/forum/0/0fdc3016.jpg

Starcraft 49.73% 48548
Super Smash Bros. Melee 50.27% 49082
TOTAL VOTES 97630
https://board8.fandom.com/wiki/(16)Starcraft_vs_(2)Super_Smash_Bros._Melee_2004
https://gamefaqs.gamespot.com/poll/1660-division-final-starcraft-vs-super-smash-bros-melee

It is not uncommon for a latter round match to have an entrant that no one predicted. What is uncommon though is for both entrants to be unexpected, and thats exactly what we got in the Finals of the 128 Bit Division in 2004. Both entrants had made it here on the back of an underdog run, but one of those runs would be put to an end with this match, and the only thing left was to see which one it was. Melee had gone up against a trio of PS2 icons and beaten them all, most notably FFX, which had been the favorite both to advance in Melee's place, and to win the entire division. There were no sure bets in the 128 Bit Division that year, but if anyone had been willing to place money on anything, it would have been on FFX. As the one to take down FFX, Melee took on the mantle of the one to beat instead, however as previously mentioned, it was not the only game with a Cinderella run going into the match.

Enter Starcraft. Melee's run would be considered the most unpredictable of the contest in any other year, but in terms of upsets, in 2004 it was impossible to beat Starcraft. Its run is considered the most legendary in contest history for a reason, and this can be attested by the fact that three of its four matches that year made my Top 20. Starcraft had hit a perfect storm thus far in the contest, running up against the most anti-voted game on each of the three current systems, but that luck would finally run out with Melee, the only game in the division that could rival the fervor SCs fanbase felt for it. It was one thing to think that Halo or Wind Waker needed to be taken down a notch, but convincing people in 2004 that Melee had its loss coming would be a tougher sell. There was also the persistent question that had been dogging Starcraft all contest, the repeated refrain that questioned how much of Starcrafts success was really due to luck. Its victories tended to come on the heels of a last minute comeback, allegedly aided and abetted by a series of rallies from Battle.net. Aside from the usual drama that has always swirled around tipping point rallies, Starcraft was repeatedly attacked with the speculation that the rallies were just a smokescreen for the real cause of its comebacks: vote stuffing. The fact that these huge swings kept occurring right at the very end of each match was enough to rouse many users suspicions, and while no one could prove it, many people remained on high alert, just waiting for Starcraft to be exposed as the cheat they knew it was.

Melee might have been the slight favorite going into the match, but Starcraft came away from the Board Vote with the lead, narrowly holding onto it for the first several hours of the poll. It would never bring its lead much beyond 200 votes however, and once the Morning Vote kicked in, Melee made quick work of things, wiping out Starcraft's lead with ease and building up a lead of its own. Despite this being a Weekday Match, Melee seemed mostly unaffected by the usual schoolday trends, barely dipping during the Midday Vote, and increasing its margins consistently for 12 straight hours. For most matches, things would have been considered over by this point, but when it came to Starcraft we had learned to expect the unexpected. We were on the verge of the Second Night Vote, right when Starcraft had made its previous moves. Still, this time it was looking to be a bridge too far. Halo had been ahead by just over 1,000 votes at this time, Wind Waker by just over 2,000. Melee was ahead by 3,500. Could the Zerg Rush really overwhelm this kind of a lead? We all stood by, holding our breaths, waiting for another miracle to happen.

And then, shortly after 8:30 PM EST, something even more inexplicable than a miracle took place. Within a minute, Starcraft cut 100 votes from the lead. And then, almost just as quickly, Starcraft did it again. For nearly half an hour, a deluge of Starcraft votes poured into the site, its numbers spiking to astronomical levels as it cut 1,000 votes from the lead in only 20 minutes. We had seen rallies before, but this was factors beyond what we knew a rally to be capable of at this time. Hell, even a latter day megarally like Undertale's was unable to do more than match the pace set here. Matches played out live at this point, and we could watch in real time as hundreds of suspicious votes came in for Starcraft, one after another. There was only one explanation remaining, and that was vote stuffing. The stuffer quickly realized that they had been caught, and within 30 minutes the spike receded, but the damage had been done. The comeback would slow to a trickle but it would keep coming. In the most dramatic way possible, Starcraft's nightly comeback had begun.

The spike might have scared us, but Starcraft still had 2,500 votes to go at this point, and if it stuck to previous trends it would wind up just short. Still, trends or no, we had just witnessed the most blatantly obvious example of cheating in contest history, and Ceej was yet to respond. The night and the comeback wore on for several more hours before CJay finally spoke in a series of messages to the board. Yes, there was evidence of hundreds of stuffed votes for Starcraft during the suspicious time frame, and if necessary they would be removed from the final total. However, the scale of the stuffing was so massive that instead of having to sort through the results and remove every last illegitimate vote, Ceej would just wait and see if the stuffed votes would make a difference. If Melee won, the result would stand. If Starcraft won, it would see hundreds, if not 1,000+ votes removed. The match would continue on for several more hours after this, and Starcraft would finish with a spirited last minute comeback attempt like always, but the match was basically over the second Ceej made his post. Even if Starcraft won, it would most likely lose, and in the end it couldnt even do that, Melee pulling off a narrow victory, even with the stuffed votes remaining intact.

Smash Bros. has become such an institution on the site, and such a source of contest strength, that it can be hard to remember a time when its power took us all by surprise. Its not an obvious fit for such an RPG friendly site, and its prediction percentage coming out of this match certainly reflects that. However, against all odds, GameFAQS had declared it the best current-gen game, and its trajectory in the zeitgeist changed for good. Starcrafts reputation would also come away from this match permanently altered. While its contest achievements could not be denied, and its place in history strongly secured, the narrative of the naysayers who had doubted its legitimacy all along had temporarily taken precedence over the narrative of Starcraft as the plucky underdog. It had ended its first contest run mired in scandal, but oh what a run it had been.

Yesmar_ posted...

What is uncommon though is for both entrants to be unexpected, and thats exactly what we got in the Finals of the 128 Bit Division in 2004.

Was Melee really unexpected? I had it winning the division in my bracket.

Underleveled posted...

Was Melee really unexpected? I had it winning the division in my bracket.

I also had Melee, but it had a very low prediction percentage

Pretty sure FFX vs. Wind Waker was the bracket favorite to be there

There is one match left I consider way too high but curious to about it.

Leonhart4 posted...

I also had Melee, but it had a very low prediction percentage

Pretty sure FFX vs. Wind Waker was the bracket favorite to be there

Melee only had < 15% of brackets advancing to this match, and only 22% even had it getting through Vice City. It was very much a wildcard entrant that people thought could be capable of taking the Division Finals, but I think most people thought that because the PS2 was obviously stronger than the GameCube, it would likely fall in the Gauntlet of the PS2's biggest killer apps/series.

That 22% is for the whole site, including all of the non-B8 users that annually believed the GTA entry was going to go deep into every contest. Board 8 had more faith in SSBM.

17. Magus vs. Knuckles the Echidna (2005) R1

https://gamefaqs.gamespot.com/a/forum/a/a1935a62.jpg

Magus 49.14% 50153
Knuckles the Echidna 50.86% 51909
TOTAL VOTES 102062
https://board8.fandom.com/wiki/(3)Magus_vs_(6)Knuckles_the_Echidna_2005
https://gamefaqs.gamespot.com/poll/2088-devil-division-round-1-magus-vs-knuckles-the-echidna

Nobody saw it coming. Thats a phrase that gets thrown around all the time when discussing upsets, but its usually an exaggeration. Theres always some core group of people, no matter how small, who were sniffing around the upset, even if very few of them were willing to take a risk and pick it. Magus/Knuckles was different. OK, sure, Smurf did predict it, but hes a Sonic fanboy, so who knows how much though was put into that prediction. Absolutely no one that was thinking critically about this match had any idea that its result was a possibility. It was a shock out of the blue, made even more stunning by the fact that it wasnt some newcomer or unknown quality that was involved. It was two contest mainstays who took us all by surprise, and forced us to reevaluate how accurate all of our received wisdom was. That helped up the ante even more and helped transform the match from a simple upset, no matter how unexpected, into a match that tied into narratives and character journeys that would span close to 20 years.

Ive written at length so far about Maguss journey through contest history. Im sure its repetitive, but its hard not to pay a bit of extra attention to Magus, the character who has one of the most memorable narratives in contest history. Going into 2005, he was still riding high in the initial stage of his journey though, widely considered to be the strongest character not in the Noble Nine, and one of the only ones with the capacity to break it. And this wasnt based on speculation or dodgy math; it was based off of a direct 1v1 match against Link himself. Even the lol, xstats crew would find that hard to argue against. But yet, there were signs that something was off with that result. While Magus himself would only have two inscrutable matches in 04, the other characters from his 03 fourpack (Ganondorf, Tidus, Sam Fisher) all had underperformances that following year. Even though there was no strong explanation provided as to why, it began to be speculated that Maguss 35% against Link was an overperformance, and, as if to test that theory, a Round 2 Match was set up in 05 between Magus and Squall, seemingly settling once and for all the debate as to who was the strongest Near Elite.

The board by and large sided with Magus, and much has been made over the years of this being the defining example of board groupthink and of ignoring the obvious signs that were staring everyone right in the face. That is certainly true to an extent, and whenever a result that hinted at Maguss weakness was brought up, people would dismiss it with an excuse. What makes this different from other groupthink situations though, is that those excuses werent just a cope; there were a lot of legitimate counters that were brought up, and seemingly contradictory results that dont have a good explanation. Ive tried not to get into the nitty gritty of re-litigating match debates from 20 years ago, but I think for this match its relevant just to demonstrate the sheer volume of coincidences that had to pile up in order to hide Maguss weakness.

First off, one of the big points against Maguss 03 value being legitimate was Tiduss weaker than expected performance against Mega Man in 04. But the thing is, if that match went against his value from 03, his Round 1 match did the exact opposite. Tidus was expected, based on the 03 stats, to have a 50/50 match with Shadow, and thats exactly what he did. It just so happened that Shadows value in 03 was also inflated for unclear reasons, for exactly the same amount. Plus, going back even further, Tiduss 02 X-Stat value was inflated as well, making the 03 value look more legitimate. Didnt Ganondorf do a little worse than expected against Alucard in 04? Sure, but due to Twilight Princess and Villain Contest hype, Ganondorf eventually boosted to his inflated 03 value anyway, once again appearing to legitimize his 03 value. Even matches that didnt directly involve any of the characters in that four pack played a role. Did Sam Fisher look weaker than he should have in 04? Sure, and while Samus overperformed on him, she also legitimately overperformed against Lara the round before (once again for unexplained reasons), making Sams numbers against her easier to dismiss than they should have been. And hadnt Frog almost beaten Snake? Surely that implied something about Maguss strength as well. If all of this made Maguss fourpack hard to decode, things were obfuscated even more by Ceejs decision to make 04 an SFF-fest, reducing the accuracy of its results, and also banning Magus and Shadow from the Villain Contest, which might have exposed either of their real popularities much earlier. As I said earlier, all the signs were there, but it is truly impressive how much was there to mask the signs as well.

In any case, as Round 1 went on, people went back and forth litigating and re-litigating the arguments listed above, not in preparation for any kind of Round 1 match, but in excited anticipation for the Magus/Squall showdown of Round 2. Even Squall supporters werent expecting Magus to lose before getting there. His match with Knuckles was just supposed to be a testing ground to see if he would show any signs of weakness, to see if he would dip below the 60/40 result that most people were expecting. The match started, and from the very start, the Magus doubters were vindicated. Magus was winning, but he was winning by less than 60/40. *Much* less. Still, for the first hour or so, he was building up his lead steadily, even if he seemed to have stabilized around 54%. His match against Squall was in serious jeopardy, but thered be time for that later. And then, around 90 minutes in, as the Power Hour wore off, Maguss time started to run out. His percentage collapsed, and Knuckles was stalling him out consistently, keeping Maguss lead steady at a mere 600 votes. There was little doubt that Knuckles would win the Day Vote, and, with Maguss lead sitting where it was, that would almost certainly be enough to win the entire match as well. The board sat in shock, watching this all play out, as Knuckles did what the trends would predict, coming on strong during the day, and putting the match away for good. Magus would manage to pull off a near 1,000 vote comeback over the last several hours of the match, but it would not be enough. The character that most people had taken to win the entire Devil Division had just been eliminated in the First Round, before the showdown with Squall had even happened.

As stunning as the result was, it was relatively clear what had caused it, even if the scale of Maguss 03 overperformance was hard to grasp. Thanks to all the Magus/Squall debate, we knew that there was a chance Magus had been overrated, but we had no clue that it had been by this much. Even though the Link/Magus result didnt have a very good explanation behind it, we understood that flukes can happen, and that there is a degree of year to year variability when it comes to X-Stats results. What happened in this match however, was far beyond some annual volatility, and called into question how we were predicting matches in the first place. The match wasnt necessarily a serious blow to the infallibility of the X-Stats, since after all the Magus is overrated argument itself had relied on that kind of extrapolation of results, but it certainly called into question how much confidence we should all be placing in our received expertise. The match had been an upset among overall bracketmakers as well, but not a unanimous one. 35% of the site had correctly predicted Knuckles to win. Once the shock of it wore off, the boards reaction shifted more towards embarrassment. How could we not have seen this coming?

The aftermath among users would be nothing of course when compared to the aftermath that hit Magus himself. He had been exposed as a fraud in the most dramatic way possible, his reputation destroyed instantaneously. Magus was not just a bracket favorite, but a personal favorite for the board as well, having been voted as the Boards favorite VG character for the second time in a row only a month before this match took place, and that intensified the reaction all the more. When all was said and done, his strength this year probably wasnt even *that* far off from where we had thought it should be, but Maguss role as *The* Noble Nine Breaker in the boards imaginary, along with the affection the board held for him, was so strong that even this mild drop seemed precipitous. And while Magus might have still looked strong in 05, first round loss aside, the same could not be said for him in subsequent years, as in contest after contest he would lose and be upset in increasingly embarrassing ways. A character who was once thought to be capable of beating Snake, by 2013, had people doubting whether or not he could beat Otacon. Never has a character fallen so far in our estimation, and it all started with this match.

I did all the math back then and there was one scenario that predicted Knuckles > Magus, but I can't remember what it was now. Maybe Luca Blight?

Leonhart4 posted...

I did all the math back then and there was one scenario that predicted Knuckles > Magus, but I can't remember what it was now. Maybe Luca Blight?

I think it was Luca, as we got an additional read on Luca through the Villains Contest that was among the warning signs for Magus going into the Devil Division.

Doing the math again, it's not Luca, although it certainly projected a close match with Knuckles.

It might have been Sam Fisher. Let's see. Assuming Sam Fisher 2003 = Sam Fisher 2004, that projects a 71.81% victory for Samus over Magus. The 2004 X-Stats have Samus over Knuckles with 68.44%. Yep, that's the one!

Now that projects out to a 55% win for Knux, which is obviously not what happened, but the scenario did exist!

Yesmar_ posted...

and also banning Magus and Shadow from the Villain Contest,

wait why

pjbasis posted...

wait why

They weren't explicitly "banned" but essentially antihero rival type characters didn't make the bracket despite having clear support on the board

For instance, FFVIII fans almost certainly nominated Seifer but he didn't get in and Ultimecia didn't either because she wasn't the most popular bad guy from the game