EPT![[image loading]](http://www.teamliquid.net/staff/pachi/banner/SPL_080930.png)
Shinhan Proleague 2008/2009 Playoff Coverage by Riptide and Heyoka
IF you can keep your head when all about you
Are losing theirs and blaming it on you,
If you can trust yourself when all men doubt you,
But make allowance for their doubting too;
If you can wait and not be tired by waiting,
[...]
If you can make one heap of all your winnings
And risk it on one turn of pitch-and-toss,
And lose, and start again at your beginnings
And never breathe a word about your loss;
If you can force your heart and nerve and sinew
To serve your turn long after they are gone,
And so hold on when there is nothing in you
Except the Will which says to them: 'Hold on!'
[...]
Yours is the Earth and everything that's in it,
And - which is more - you'll be a Man, my son!
- Rudyard Kipling, If
![[image loading]](http://www.teamliquid.net/staff/riptide/proleague200809/pl_playoffs_jangbilit.jpg)
It was the best of times, it was the worst of times. It was a weekend of four bo7s, two ace matches, and a some of the best Starcraft we have seen this year. It was the end of the line for some, and a new beginning for others, who rose from the bottom to do the impossible. Titans clashed, battles were fought, blood was shed, and in the end, it came down to a single match, one deciding showdown. Two fell, defeated, and two rose, to save their team and live to fight another day.
CJ Entus, heroes to many, enemies to some, sat in silence, a moment from elimination, their fate resting on the capable shoulders of a determined young Zerg. Today, Kim Jung Woo played what was perhaps his most important game to date. This wasn't the OSL final. This was bigger than that, bigger than him, bigger than any or all of his hopes and dreams combined. This was about those he had sat and trained with, for hours, days, weeks, and months. This was about two years of Proleague, and the fate of an entire team. It wasn't about medals or money, or fame. No, as EffOrt took the steps to his booth, and settled in, he felt with every nervous beat of his heart, that this next game was about pride.
EffOrt has been tearing up a storm recently. He has been dominating the proscene with an impressive 67.01% of wins over 97 games (as of this writing). Effort is indisputably one of the best Zerg players that has ever existed, both in the land of statistics and the land of pure badassery. Like a tiger hidden deep within the brush, he watches his prey, and waits for the perfect moment... to STRIKE.
For those of you who doubt: Effort recently defeated Jaedong in a convincing 2-0 match in the GOMTV Classic Season 3. What more proof is needed to crown a Zerg king? It is his best matchup, clocking in at a spectacular 69.7% winrate over 33 games! Faced with the super ace match against the unlikely YellOw[arnc], there was little doubt that Effort would once again reign victorious with his stellar ZvZ ability. Or was there? Prior to the Ace Match, Effort's track record versus the infamous Yarnc, Luxury's evil twin, was a drastically uninspiring 3:1 in Yarnc's favor. Not good.
The last time these two played? One day. The battlefield? The all too familiar Heartbreak Ridge. The result? Yarnc 1: Effort 0. Perhaps this could be the day things would change. Perhaps this could bring about the dawn of a new era. Perhaps the shining hope of CJ, the ferocious, determined, and fearsome Zerg Effort would rise to the stage. And perhaps, he would conquer. The stage is set. The setting? Heartbreak Ridge. Actors, take your places.
The doors close. The only sound that can be heard in Effort's booth is the clang of his beating heart, smacking rudely against the confines of his rib cage. This isn't fear. This is rage. The sound of imminent death.
"I will not lose again. Not to the likes of you. Prepare to be humbled."
A few feet away, the Hite SPARKYZ player can only hear the clattering of his own teeth. He runs through his bo for the last time as a shiver runs down his spine. He closes his eyes and sees a large beast pouncing on him, rendering him helpless. The countdown begins:
5
"I will not lose again."
4
"Prepare yourself."
3
"I will destroy you."
2
"No mercy."
1
"gg."
This wasn't just any old pro team. This was the team that had produced the Maestro himself. For years, Kim Jung Woo had looked up to Ma Jae Yoon as the Zerg overmind, the perfect player, the destroyer of anything and everything. Now, here he was, about to fill those shoes. Closing his eyes, he clears his head and listens to his cold, calm, inner voice, honed by months of training with the best team in the country. An Entusman always keeps his head. An Entusman never shows his hand. When called upon, an Entusman delivers death, and does so swiftly.
Having won yesterday's game vs Hite, CJ walked into their second game with an advantage, an advantage that Effort quickly capitalised on by beating type-b, avenging Savior's loss in the process. An unlikely duo, however, once again put their playoff spot in question. While Yarnc taking Kwanro was not too much of a surprise given his ZvZ win on Day 1, what shocked fans everywhere was go.go beating Iris, and even more so, Justin beating Movie. Yes, Justin. Movie is certainly not the brightest bulb in the chandelier, but this is Justin aka Crocus aka Bifrost aka Casy[Name] aka TheMarine[Name] we're talking about. The man is 8 - 15 in his TvP career, and Saturday's unlikely hero Movie chooses to two base arbiter him. Talk about going from hero to zero.
Two days straight of playing, and finally, everything decided in a ZvZ. EffOrt steps out, and faces Yarnc, the same Yarnc who had trounced him quite convincingly a little over 24 hours before. S-Class is a word that is thrown around a lot in the community. S-Class gamers are frequently crowned as such for their skill and performance. S-Class gamers have flawless macro, and can make their units dance, dance, dance all day. Starsense, you say? Well, yes. That too. We'd like to propose the idea, however, that the term is best defined in situations like this, when with everything on the line, a player can keep his or her head, and do what needs to be done. Clutch.
You can have the most fantastic muta micro, and be able to control each ling individually. When the game is on the line however, when the clock is ticking, when the fans are covering their eyes because they cant bare to look, this is when the S-Class gamer steps up with a cold, unwavering gaze and delivers a flawless game.
With the heat of a thousand eyes upon him, Kim Jung Woo kept his wits about him and stayed in the moment, playing tactfully against a very aggressive Yarnc. Holding on despite being at a disadvantage in terms of muta count, EffOrt defended his two hatches and kept his few remaining drones alive, slowly harassing when he could and riding the econ advantage he had got from taking out some of Yarnc's workers. Pundits will argue that it wasn't the most impressive of games. They will say that EffOrt's mutalisk micro was nowhere near Lee Jaedong's. The fact of the matter though, is that all that talk is irrelevant. Under immense pressure, and with the knowledge that he had lost to that same opponent just a day earlier, EffOrt, the Zerg heir in waiting, managed to play a near perfect game.
Talking of perfect games, however, would not be complete without mentioning the exploits of a chubby young Protoss by the name of Heo Yeong Moo. When we spoke of January's boys in last week's PL update, we didn't give them much of a chance against a STX SouL that had streaked through Round 5. Then again, that is the beauty of Starcraft, that statistics and predictions and analyses aside, it's all about performing on the day. Indeed, the day was Saturday, and Samsung was performing brilliantly. Firstly, they took the game to an ace match, which was extraordinary in itself, and overdid themselves by taking that as well to go 1 - 0 in the series.
Some gamers hide behind the face. You know what I'm talking about. The CJ face. The fuck you face. The I'm too cool for this face. (-_-). Not my hero. My hero subscribes to the ChoJJa religion. He plays with the sparkle in his eye and a tug on the corner of his mouth. He brings joy to his team mates and causes fan girls to swoon.
Switch.
In the booth my hero is a monster. Not content to merely win, he smashes his opponents down a well, Spartan style. When he games his opponents are left shaking their heads. How can this innocent child bring so much joy to his friends and so much pain to his enemies? +1 speedlots may hold the answer, but nobody had time to ask, let alone blink, today.
Maybe this is what STX is talking about in the van on their way home. Maybe this is the question Calm will be mouthing to himself as he is unable to sleep tonight. Maybe this Jekyll and Hyde routine is the source of Jangbi's strength.
CJ, beware. Jangbi is coming at you. Coming with a smile.
Though Stork avenged Saturday's loss by coming back from the brink of defeat to take Modesty out the following day, FailBatZero totally dropped the ball by losing today's game as well. After yesterday's embarrassing build order loss to Calm, the erratic Terran was completely schooled in TvT by Hwasin, and with Juni, yoOn and Jangbi losing a game apiece, STX went 4 - 1 and took them to a super ace.
Of course, as all heroes of Broodwar do, Jangbi stepped up when asked of him, and despite losing his first PvZ of the day, went all out crazy on Calm. With a forge spinning, and two stargates going up, Jangbi's build order was discussed in detail on IRC, with some even suggesting that he had gotten it from combatex. A couple of seconds later, and there were ton of zeals in the Zerg main, happily having a go at anything and everything. For once, you didn't need to understand what the commentators were saying. Calm's face said it all.
Powerhouse is probably not the first thing that comes to mind when you hear Jangbi. This weekend, however, that's exactly what he was. Coming in from losing a game vs one Zerg, he took the opposition's better Zerg to the cleaner's with build order jugglery that requires, in the very least, nerves of steel. If his build had been scouted, and countered, he would have been quite far behind, a lead that a player like Calm would have had no problems exploiting. Indeed, though much admired for his crazy storms and elegant PvT, Jangbi's true strength, and the reason he is so invaluable to KHAN, was seen today. Coming from a loss, to deliver such a convincing win in the same matchup just a few hours later is not something just any player can do. It's not something even the best of players can do easily. Today, however, Jangbi did it, and did it in style.
As KHAN celebrate this victory and begin preparing to take on a CJ led by EffOrt, Jangbi no doubt knows that a lot depends on how he performs. Samsung fans, however, should not be overly worried. If today's game was any indication of his prowess under pressure, the Protoss should have no problem facing bigger and badder obstacles as his team progresses through these playoffs.
![[image loading]](http://www.teamliquid.net/staff/riptide/proleague200809/pl_playoffs_jangbicookies.jpg)
"Coach, could I have those donuts you promised me now?"
"A shop, Jangbi. You can have a donut shop."
As much as the weekend was good for some, for others it was terrible. Lethal Leta, the Hite pocket rocket, the wrangler of wraiths and destroyer of Zerg - yes, that Leta - failed. The Ace Match Bonjwa not only dropped his ace encounter yesterday to Movie, he lost to SkyHigh twice, and got bad mannered in the process.
Bad manner, you ask? Where? Here, sports fans, here, in the form of a unique new Terran build.
![[image loading]](http://www.teamliquid.net/staff/riptide/proleague200809/pl_playoffs_skyhighcc.jpg)
The offensive CC, the new face of bm.
As much as we expected some bad manner from FireBatHero during these playoffs, the attitude actually came from CJ's own Skyhigh. First used by the young Terran in Saturday's game vs Leta, the offensive CC is the new and cool way to rub in a defeat. Truly taking bad manner to the next level, this technique was soon picked up by go.go, who built a CC in Iris' nat, and finally perfected by Justin, who used it to add salt to Movie's already festering wound. In the end, what goes around certainly seems to come around, with CJ getting the offensive CC twice today, a fact that Skyhigh apparently hadn't yet grasped as he gave the camera a thumb's down in a manner that was truly uncharacteristic of a CJ player. EffOrt, despite being the one who actually bagged the ace match, was content with a few high fives before quickly slipping on his CJ face once again.
At the end of the day, however, one simply cannot help but empathise with both Leta and Movie, two players who really got the short end of the stick. Leta, having led his team for so long, crumbled under pressure to a beautiful pincer manoeuvre with storms thrown in, to go from the brink of victory to certain defeat. Moreover, he lost two consecutive TvTs, and in doing so, went 0 - 3 in the playoffs. Movie, on the other hand fell even further, going from CJ's saving grace to that guy who lost to Justin in a little over twenty four hours. For these two players, the weekend that passed is one that they'll want to forget. For many others, however, and I'm sure for many of us as well, this certainly was a weekend to remember.
As these playoffs progress, we have no doubt that there will be more great games, and that many more will be called upon to give their all for their teams. When this happens, when player faces player at the end of the line, and when time stands still and a countdown begins, we'll be here, watching, waiting for a hero to rise, just like like they did this weekend. Yes, in those five seconds of stillness, in that calm before the storm, we will hold our breath together, and remember why we come together to watch this great game. So, thank you for sharing this weekend with us, and here's to many more!
Statisfaction
Team depth is something that is talked about constantly in standings and live report threads. People make matter of fact statements about how certain teams (STX, CJ) are deep and others (Oz, KT) are one man shows. When seeing discussions about these kinds of things, it always makes me wonder what parts of it are merely perception and what parts we can back with some kind of data. Sometimes charts and numbers fly around in those threads, certainly we can find a more satisfactory way of rating these things than has been done so far.
This project began with a totally different approach. I had a theory that teams send out their worse squads when they are playing weaker opponents, assuming it was a necessary approach to developing players. This turns out to be true on some level, as the chance of a new player getting his shot does increase when facing eSTRO/ACE caliber teams. Overall, though, a teams active/good roster is actually only about five players, so the lineup SKT fields against WeMade is more or less the lineup they field against Oz or CJ.
This means that any given team is predictable to a large degree, which then means it may be possible to quantify and rate how solid each team is relative to each other. With playoffs being a best of seven series, this is a great time to examine how deep each team is. We can see which teams can expect to benefit from the longer set, and which teams will struggle to survive in a longer series.
The first thing I did was sort players in a team by how frequently they were in the posted lineup (only for the past 2 rounds, I drew a line at how current the stat was). This does not include ace games, only regular matches, and does include games that were on the schedule but not played (times when a player was slated to play a game 4 that wasn't needed). It gives a chart that looks something like this:
![[image loading]](http://www.teamliquid.net/staff/heyoka/depth/1.png)
It is a pretty cool way of looking at teams. Here are the ones for the top six proleague teams, notice how some are vastly different:
+ Show Spoiler +
SKT:
Oz:
![[image loading]](http://www.teamliquid.net/staff/heyoka/depth/4.png)
STX:
![[image loading]](http://www.teamliquid.net/staff/heyoka/depth/3.png)
CJ:
![[image loading]](http://www.teamliquid.net/staff/heyoka/depth/5.png)
Hite:
![[image loading]](http://www.teamliquid.net/staff/heyoka/depth/6.png)
KHAN:
![[image loading]](http://www.teamliquid.net/staff/heyoka/depth/2.png)
Note: these numbers omit the last week of proleague, so few teams played their normal lineups I felt it just screwed with the distributions. I also am not showing players with 1 game.
I don't know if this proves anything particularly useful, its probably saying more about who a coach is willing to send rather than anything about where a team's wins are actually coming from. Notice how SKT is essentially a 4 man team, but because their Z time is split between Hyuk and Thezerg (neither of which are likely to perform better than 50% in any season) it looks like a much more dynamic team than it really is. It does show visually how some teams really dry up quickly past their top 4 players, look at the huge dropoff between how often Oz played Lomo and how often they played ggaemo in the past few months.
Using this alone, we can say with some confidence that the teams likely to be worse in a longer format are Oz, Hite, KHAN, and SKT (while they have 6 active players, the last 3 of them have a 40% average win rate). CJ and STX seem to have the most to gain from the longer series. July doesn't even make STX's top 6 when you look at it this way (and accordingly is not being played against KHAN, clearly proof I am right).
We can do better than this. I checked for a couple of different correlations such as ELO and overall win percentage in proleague games (some weighted averages to see if frequency was connected with player strength), and while it looks like some exist, none of them are strong enough to draw any conclusions from.
I stumbled across something that might be useful however. Using the above frequencies as a template, I figured out the "ideal mix" for each team. This is the 4 players that most frequently see play, so for KHAN it is great, FBH, JangBi, and Stork. For each team, this mix is the exact group of players they send out in a match about half the time (with most of the other half having 3 of those players in common). These are the 4 players the coaches beleive in the strongest, and if a match were occur today against an unknown team they are the players that would be fielded. I then did the same thing for the rest of the players who have more than one game appearance (this is to filter for "semi regulars" on the lineup and weed out those who got a game but lost and were sent back to practice). Using these two designated groups, I then tallied up the number of games played by each one, and got the following. Top 4 is the number of games played by the 4 most frequently used players, and "5 Up" is the semi-regulars. The percentage is the portion of total games the "ideal team" played in relation to the entirety of games a team played on the whole.
![[image loading]](http://www.teamliquid.net/staff/heyoka/depth/depth.png)
This does not include winner's league, nor does it include some individuals who only have one game so the total game number will be lower than you find in the R&S charts. Again, its not super good for drawing hard conclusions but it begins to quantify what the above graphs show, in a much more convincing way. Oz relies pretty heavily on the top of their A-team, CJ has a large amount of variety in who is getting played. What we're mostly looking at here is how one-dimensional a team is in terms of who gets played.
To get a little deeper, it is necessary to look at where the wins are coming from. This is the same chart, spearated into two categories, but instead shows the win percentages of each group. This is basically giving up a win distribution chart filtered by who has been actively playing for the last 3 months. T4 W is the number of wins in games the "ideal team" has, T4 L is their losses and then the accompanying percentage is their win ratio as a group. The 5+ columns describe the same things for the lesser played regulars.
![[image loading]](http://www.teamliquid.net/staff/heyoka/depth/dwinrate.png)
This is a little more useful in actually saying something concrete about which teams can expect to benefit from matches played in a longer set. The smaller the gap is between the Top 4 win ratio and the other win ratio, the more depth we can say a lineup has. STX is notable in having the only semi-regular lineup with an above 50% winrate, but the way this groups players, July (who is 13-6 this season) is in that category, so it might not be as significant as I want. Again, this is showing CJ and STX are more versatile teams, and KHAN is going to have serious trouble finding players number 5 and 6 to fill out their matches that are best of 7.
The one thing I want to stress here is that this is an experiment. These numbers should not be considered accurate as far as actually rating (or more importantly, quantifying - yet). I made several assumptions, drew a few arbitrary lines, and am not using as large of a data set as is needed to truly say we proved anything. This is the first step in a long journey. In the future, I think this will be very useful in discovering just how useful any one player is to the team effort (something that is rarely looked at but will be hugely important as free agency becomes relevant). This method is pretty rough for actually describing an aspect of a team, but is a good starting point towards more detailed discussion of ways we can objectively look at teams aside from common notions and subjective perception.
Thanks to Mikeymoo and Manifesto7 for the guest columns and Fomos for the awesome pictures!
