EPT
I would rather that my spark should burn out
in a brilliant Blaze than it should be stifled by dry-rot.
I would rather be a superb meteor, every atom
of me in magnificent glow, than a sleepy and permanent planet.
The function of man is to Live, not to exist.
I shall not waste my days trying to prolong them.
I shall use my time.
-Jack London
Save the Last Dance
As fans of StarCraft, we’re fascinated by the old-timers. We Love underdogs, but we Love our personalities of yesteryear even more. It’s a curious phenomenon that, being a die-hard Zerg (and by extension,
Jaedong) fan, that I found myself cheering for 
Sunny in the ACE v Oz match. We cheered when ACE v FOX gave us NaDa v 
Anytime, more than we cheered when KTF v Oz gave us Flash v Jaedong. I remember getting excited every time KTF and ACE faced off, in hopes that the coaches were as nostalgic as I hoped they were. Don’t get me wrong; I haven’t been following progaming since the times of
BoxeR and YellOw, but who could resist the prospect of another peek at the greatest Rivalry in StarCraft?And we also feel despair as the old-timers lose their grip on the game they once dominated.
More often than not, the old-timers lose and we sigh and wait for ACE to send another of our washed-up heroes. However, every once in a while, after we’ve given up hope that one of our favorites has finally reached the end of the Line, that they’re not the player they once were, we get one Last look, and that magical occurrence: the comeback.
Sometimes it isn’t an old-timer. Sometimes it’s a former bonjwa, sometimes it’s a player who never made an impact giving it another shot. But each time, we’re thrilled. We can’t wait for the next game, and the next S-class victim of our first idols. Nobody cheered harder for YellOw, except for maybe Last Romantic, than me in the Classic season 1. He showed us he still had the game, and reminded us of what made his generation of players great: their heads. YellOw’s Classic run was, we all knew, his Last, but he gave it all he had the final time around.
There have been other comebacks like Jin Ho’s, some longer, some shorter, some more glorious or more frequent. Following this paragraph, I profile several comebacks of three distinct variants. I have classified the different types of comeback and provided examples, but first, there is the pivotal question: what’s in a career comeback?
From what I’ve investigated, and I admit I have limited data, particularly in comparison to my Last two articles, I can say these things for sure: a career comeback is a prolonged return to approximate original form after a prolonged removal from that form. I have defined this temporally, which may seem counter-intuitive, but I explain it this way: slumping players play fewer games. A return to form is correlated to two things: confidence and practice. This is consistent with intuition, I think. StarCraft is a very mental game, but there is doubtless a physical component. A temporal assessment rather than an assessment dependant on a constant number of games has the advantage of being perfectly accurate on, brace for it, time. An assessment based on the number of games up until a present range depending on an arbitrary player ‘form’ or baseline win rate would dramatically and deleteriously extend and skew periods of slumps until well past a recovery.
Now, from my examination, I am prepared to say that there are three types of comeback. Please forgive the cliché names.
The Roman Candle
This type, I think, is most common. It is fairly simple; a player comes on the scene, starts performing well for some time, consistently, and then a traumatic event, or perhaps a strong stylistic counter, or even general adaptation of the game, happens, and a slump ensues. After several months, or longer, of significant underperformance as juxtaposed against the consistent pre-event performance, a resurgence occurs.
This resurgence may be brief, maybe only a month or two, or it could be much longer. The necessary provision, and this is not arbitrary, is that the resurgent period does not exceed one cycle of the individual leagues. This is precisely why I like to call it the Roman candle, because it is in and out in a flash, the mystical one Last run. Invariably, these fail somewhere along the Line. Of course, this definition is not limited to individual leagues/3-4 month cycles. The example I took for this was XellOs’ 4 month late 2006 resurgence.
After the end of his dominant period, which, according to my fact-finding expedition, found its informal end on the 25th of October, 2005,
XellOs slumped hard. His dominant-era win rate was approximately 62.28%, falling to 51.1% in the nine month period between 10-25-05 and his resurgent period between 8-4-06 to 12-14-06. During that time, XellOs was again rocking at 61.1%. But XellOs woke up on December 15th, 2006, and realized that the magic was gone. Since that day, XellOs’ overall win rate is a dismal 43.59%.There are a few more examples of Roman candles that I can think of. I’m afraid
Pusan just slumped out of a Roman candle period. Between 4-5-07 and 12-16-08, Pusan was at a 60.34% win rate. Between 5-17-2005 and 1-27-06, Pusan was at a solid 60%, then he slumped to a nasty 48.15% (26-28), Only to swing up, and now back down. I’d say it looks like he’s not coming back up. He’s 0-5 in 2009. DaezanG is another pretty clear example. There’s another name I should put here right now, but I can’t make myself do it. See if you can guess who it is before I profile him. It isn’t that hard.The Phoenix
I’m gonna go straight up and say it. I don’t like the name because it’s pompous, corny, and cliché all at the same time, but it is apt for the description of this kind of comeback. I think this is well described with an example, so I’m going straight to the profile after a very brief overview. This comeback could be described simply as coming out of a protracted slump, but it’s far more than that. This is the renewal of a career in a dramatic way. To illustrate how it’s not just removal from a violent slump, consider Jaedong after being derailed by Flash. Jaedong slumped, but slowly returned to somewhere close to form. That is the rational and expected fluctuation in a career. The Phoenix is the sudden and violent return of a player, and then that player maintaining heightened performance with consistency for a very long time.
Here I examine the Classic case of this:
July. July came out of nowhere, embarrassed a top player, and snagged a Golden Mouse. After his first few OSLs, July hit a wall, and we all knew he was never coming back. The players were just too good today, his old-school Style wouldn’t cut it against BeSt’s sick macro. We all, to our shame, forgot one thing: July is a StarCraft genius. Still, after trades and a 50.89% winrate post 9-23-05, it was common knowledge that July was done. Fortunately for Zerg fans, he wasn’t, and since 2-28-08 he’s been at a 65.31 win rate. That’s higher than before his slump. July is a beast.The NaDa Syndrome
I’m just going to profile NaDa and that’s going to be the explanation, because I don’t understand the data. At all.
Here goes:
Through 2-26-2003, NaDa was 73-30 (70.87%)
To 9-12-2003, NaDa was at 33-29 (53.23%)
To 3-25-04, NaDa went 45-19 (70.31%)
To 11-14-04, NaDa went 34-28 (54.84%)
To 3-5-2005, NaDa went 43-18 (70.49%)
To the end of 2005 NaDa went BACK to 23-23 (50%)
From jan 1 '06 to sept 1 06 NaDa climbed up to 60%, going 18-12
From there NaDa finds consistency at ~55% winrate, with fluctuations from that point barely if ever exceeding 7% over any significant time frame. For 2 years, NaDa’s win rate fluctuated more than Rosie O’Donnell’s blood pressure. The most salient feature of these data is that these flips are never for a short period of time. The shortest duration is NaDa’s final return to 70%, but in those 4 months, NaDa still played 61 games. So the real question is, is this a comeback? I grappled with that question before listing it. In the end, I figured that it could really be nothing else. Long periods of complete domination interspersed with long periods of mediocrity, finally settling with consistent mediocrity. The other consideration is that it could merely be the simple fluctuation of a career, that players have hot streaks and cold streaks, and that’s how they play. There’s only one problem with that: it isn’t the way that most players play. Sure, a good many players do have hot and cold streaks. That is not being contested. What distinguishes this type of comeback from simple career fluctuation is the sheer duration, and the violence of the swings. Career fluctuation would not swing so violently, but here we see NaDa’s win rate frequently and wildly swinging from mediocre to incredibly dominant. A 70% win rate over a period of months is a rare occurrence. It doesn’t mean it doesn’t happen, but only a few S-class players will Ever see that kind of streak over a couple of months, and at that usually only once or maybe twice. NaDa hit over 70% three times for prolonged periods. That is special. As such, this is the rarest type of ‘comeback’ if you will. The only other one who I have data for is
Light (42.11-64.29-43.48-60-68.75-45.52-59.18) but if there are doubters about this one I suppose I could sift through and find a few more.In Front of Our Very Eyes
So we come, Really, to the elephant in the room if we’re discussing comebacks.
Ma Jae Yoon has been on Fire. He’s broken his streak for consecutive wins, and he’s beaten some of the best players around. Granted, his latest MSL games were shaky, but he made it through with a good strategic decision, which shows that his mind is in shape.
Nearly every MJY fan, and indeed nearly all of TL, has been examining practically every play in every one of MJY’s most recent games, so there’s no need for me to go in depth. It looks like, from the perspective of the data, that MJY has fired off a Roman candle. I have several reservations about that ascription, including his win against Bisu, his strategic builds, and the breaking of his streak, but really the data points to a Roman candle. Here’s why:
From 8-16-03 to 3-2-07, MJY was 116-59 (66.29%)
From 3-3-07 to 9-20-08, he sucked it up at 51-53 (49.04%)
Since then, he’s gone 21-11 (65.63%)
So on first examination, the percentages and time intervals are classic. Dominant era, slump time, return to form, and then a final, precipitous fall, which we obviously have not seen yet. You could look at this, point to the July data and say ‘Could be that one!’ Yes, it could be that one, too, but there are a number of reasons why I think that it isn’t.
So I’ve stated the obvious. Anybody who has read the article can tell which MJY may be and why. Anybody even who hasn’t read this can tell that the future of the Maestro’s career hinges on this Starleague run, and knowing MJY, even despite his post-MSL interview, nothing but first place will leave him walking away satisfied.
His next set is against a Zerg. We don’t know which, but put this in perspective:
YellOw[ArnC] is 3-6 career vs MJY, and his overall ZvZ win rate is below 50%, and ZerO is 0-1 v MJY, and has a 50% ZvZ win rate. Since his return, inaugurated by a ZvZ, MJY is 9-2 in the matchup. None of his opponents were particularly good in the matchup, but as has been established, neither are Yarnc and ZerO.If this postulated categorization of comebacks has true merit, then it must be able to make some accurate predictions at least at greater frequency than TheRock wins games. So I could spit out the easy ones: Anytime, if he hasn’t already sputtered out, will, and Light will fluctuate some more before finding consistent mediocrity. That would take care of both the Roman candle and the NaDa syndrome. The final prediction regards, you guessed it, MJY.
So here is my very gutsy 2-part prediction. Part 1: If MJY wins the Lost Saga MSL, his return will be protracted, and he will again play with the S-class.
I had originally planned this section as a big hype for MJY’s side of the MSL bracket, and despite everybody winning who I thought would win (the outcome of Yarnc v ZerO doesn’t particularly interest or concern me, as I explained above, but if you’re dissatisfied, I called Yarnc), it does seem to fit, so I’ll summarize and conclude in the same section:
No matter how this Lost Saga bracket turns out, somebody will get a chance at glory. For some, it is a coveted and desperate second chance, the fulfillment of a comeback. For others it is the fulfillment of promise. With this in mind, you may be wondering what the second part of my ‘very gutsy’ and rather irrational prediction is.
I’m gonna regret it, I know, but, like with the old-timers, who we cheer for despite our better judgment and who we hope for against our reason, this is one of the best parts about being a true StarCraft fan:
My prediction:
+ Show Spoiler +
Lost Saga MSL 2009 Finals: sAviOr 3-1 Jangbi
Edit: Spelling=fail
