by Zealously

The GSL Group Selections have time and again been the source of almost unending confusion, rage, and grief. There's no telling what the players will do, from committing GSL suicide by picking the strongest available players, to creating 100% mirror-match groups, or even choosing to play their own friends and teammates, we've seen it all happen in the GomTV studio. Tonight, the first Group Selections of 2013 will go down, and there will most certainly be much more confusion coming our way.

In this post, I will attempt to analyse the underlying reasons for players making the picks they do, identifying the most significant factors in these picks, as well as try to predict player picks. Please note that suicidal tendencies cannot be accounted for, nor the possibility that they will be tricked, taunted, persuaded, threatened, or flattered into making spur of the moment decisions.

Code S Ro16 Group Selection Format

The following was taken from the previous season's group selection thread by Shellshock1122. Reading through the post will help give you a better idea of how group selections work.

The four players with the highest GSL 2012 point rankings are placed into four groups - 1st place to Group A, 2nd to group D, 3rd to group C and 4th to Group B. This season, DongRaeGu goes to group A, Life to group D, MC to group C, and Squirtle to group B.


There are three rounds of picks.

• Round 1: The #1 ranked player picks from the remaining players, then the players ranked 2nd to 4th make their picks (A - D - C - B order).
  
• Round 2: The player picked into Group D during round 1 then begins the second round of picks. The players picked in groups C, B, and A follow (D - C - B - A order ).
  
• Round 2: The player picked into Group A during round 2 then begins the final round of picks. The players picked in groups B, C and D follow (A - B - C - D order). The final two players effectively have no real 'choice.'

The Five Key Factors

Obviously, the first thing a progamer will ask himself when making a pick is "Which opponent do I have the highest chance to beat?" If that were the only criteria they had in mind, then group ceremonies would end up being a fairly dull affair, with 90% of the selections being extremely obvious and predictable. However, maximizing one's win rate is not the only thing progamers have in mind at the group selection, and there are a number of key factors outside of the game that influence their decisions. Here are the five most important factors, in no particular order:

1) The ”Friend” Factor
If there is one trend that has remained almost unbroken throughout all GSL seasons, it is the rule of thumb that you do not, if it is even remotely possible to avoid it, pick friends or teammates. This one doesn't need much explaining: picking your friends and teammates might mean that you end up eliminating the other, ultimately preventing them (or yourself) from being successful and possibly creating quite a bit of hard feelings.

Another thing to consider about picking team mates are the mind games – Life will know Bomber well enough to counter what Bomber should be doing before he does it, and Bomber will play to counter Life's counter before any of them have actually even done anything yet. With this in mind, picking a teammate is not only a recipe for hard feelings, it becomes a dangerous thing to do even if you're generally the more skilled player.

The most well-known exception to this rule came in GSL November, where two-time champion MMA made the controversial decision that his fellow team mate, YugiOh, was the weakest player available to him. Without mercy, he picked him and sent him right out of Code S. In general, however, players will not pick their teammates or close friends.


2) The Revenge Factor
The desire for revenge is a natural human emotion, and it is a desire one must consider when looking at the reasons behind player picks in the Group Selections. A recurring theme throughout every tournament in GSL history has been the desire get back at player X for beating you that one time a few months back. However, it doesn't always play out the way the vengeful party would hope.

On the contrary, it is a fairly common occurrence for the player seeking revenge to get demolished yet again. It seems obvious that picking a player who already beat you isn't the smartest strategy unless you are really confident that you just had that one bad day, but it doesn't stop progamers from continuing to try. On occasion, things do end up working in the vengeance seeker's favor. Some players become overconfident after a particularly crushing victory over an opponent, and fail to prepare adequately for the rematch.


3) The Overconfidence Factor
You are DongRaeGu. You have just advanced from the Round of 32 in dominating fashion, beating Bomber and Polt on your way to the Round of 16. In the Group Selections, Polt – the player you dominated just freaking yesterday , is up for grabs. Surely, the most logical decision is to pick that player, right? You owned him once, why shouldn't you be able to own him again? Oh, Polt beat you and got you kicked out of Code S. How on earth did that happen?

This demonstrates the Overconfidence Factor. Being overconfident in the GSL is perhaps the most serious mistake you can make, and it will get you eliminated more or less every time. DongRaeGu knew his ZvT was strong, and was confident in the fact that he had beat Polt just recently. That might have been enough to justify his move, if not for the fact that Heart and Happy were still on the table. Yes, DongRaeGu allowed a measly 2 - 0 victory to make him think that somehow a one of the best Terran players in the world would be a better opponent for him than two much less accomplished Terrans.

This not to say that picking a player you're very confident against is a bad move – Bomber is going to feel good about going up against Protoss, Life against Terran, Soulkey against Zerg, and so forth. But when players start to think less with their brains and more with their egos, they sometimes bite off more than they can chew.


4) The ”Group of Death” Factor
From a long-term win rate point of view, the least reasonable way of picking your opponent is picking an someone you know will be an excellent player. And yet, that kind of thinking is almost contagious in the GSL, with several Code S tournaments ending up with a Ro16 group of death. GSL champions in particular tend to attract a lot of attention at Group Selections, and for good reason – picking and beating them is a gateway to recognition and fame. However, you have to sit back and think, is it worth the risk? If it backfires, it can be a shortcut straight out of Code S.

Take, for example, Season 3's Group of Death containing MC, Symbol, Squirtle and Marineking. Right off the bat there were players available for picking that would have been consider safer, not to mention much easier opponents, than those guys. And yet, they were all very intent on creating a group that no one could safely advance out of. Why? It ties somewhat to the Overconfidence Factor above – they were all convinced that they could beat whoever they picked, and wanted to show off the skills they were so confident in.

The reasoning behind it appears to be that with how hard it is to distinguish yourself from the masses in Korea, creating a group of death and advancing is one way to prove how skilled you are. It is also the most risky way of creating a group, but there always seems to be ”that one guy” who starts it, and everyone else in the group is somehow roped into following his lead.

Something that ties to this is a kind of revenge that often occurs within the Group Selections themselves. Let's say for example, Marineking struggles against Zerg, and says he's inclined to pick a Protoss opponent such as PartinG. PartinG, who does not want to play a strong player like MKP, threatens to use his pick on a strong Zerg if MKP decides to go ahead with his move. MKP decides to call PartinG's bluff, and in response PartinG refuses to back down and brings Symbol into the group. In this example, PartinG knows that Marineking might be favored against him, so he decides that if he's going out, he's going to do his best to have someone eliminate Marineking for him. This also brings us to the fifth and final significant factor:


5) The ”Symbol” Factor AKA. I'm going to be really mean to you now
”Symbol hates LG-IM” is a common thing to read when they both appear in a discussion. No one can blame him, with how he lost to Seed in the most insane reverse-sweep ever in the GSL and proceeded to slump for a while after that.

What he did in the Group Selections once he got back there went beyond simple ”I'll pick the player who beat me and beat him”, and into somewhat controversial territory. By picking Happy, Seed's team mate, he knew that if he was going out, then chances were that one of the LG-IM players would eliminate the other. This struck many as mean, perhaps even evil, but there is a certain stroke of genius to it. Symbol knows that the knowledge that you might have to eliminate your teammate and friend could really throw you off your game and give him the advantage, and he doesn't hesitate to use that.

It also gives Symbol the credibility to make a similar threat in a future group ceremony. Most Koreans are too well-mannered to force a team-kill, but Symbol showed he's willing to protect his own interests at all costs. It's always something to consider – whether a player does it out of desire for revenge or cold logic doesn't really matter, because such a strategy is very potent and it most certainly shows that you mean business.

Mock Draft

Now that we've got all that out of the way, how about a mock draft? There's dozens of different factors that could swing the group selection one way or another, and one influential pick early can change everything. But for the fun of it, here's just one example of how things could go.

First pick: DongRaeGu: Marineking
Though DRG and MKP apparently are good friends, picking the player with the weakest vZ in the Round of 16, and the one that he has also developed rivalry with seems to be by far the most attractive option for DRG. In addition, DongRaeGu has shown in the past that he wants MKP in his group every chance he can get. One tendency that cannot be overlooked completely, however, is DRG's stubborn preference to always pick players he beat in the Round of 32. This hasn't worked out for him so far, as Creator and Polt have both punished him in the Ro16. If he does continue this trend, it seems Innovation, another Terran, could also be on the top of his wishlist.

Second Pick: Life: Soulkey
Life is still upset with how season 5 turned out for him, as he was eliminated from two straight losses against Soulkey in what was considered his best match-up for a long-time. In his Ro32 interview, he stated that he wants to get revenge. Unless the Startale coaching staff talked him out of that pick during this time, it's probable that he'll go through with it.

Third pick: MC: PartinG
MC went on record saying he wants an all Protoss group. Take a moment to take that in. There are three Protoss in the Ro16 and he as many of them in his group because he wants to monopolize the Protoss spotlight. Since he can't take Squirtle, he'll end up having to take the only remaining Protoss in Parting.

Fourth pick: Squirtle: BboongBboong
Squirtle advanced in second place from his group after losing to the vastly improved BBoongBBoong, and I can definitely see him picking BboongBboong to get back on him. Though Curious is also available, Squirtle was his teammate for a long time and as such it is unlikely that he'll be picking the Code S Reader.

Fifth pick: Soulkey: Symbol
Soulkey stated in his Ro32 interview that he wants to create another all-Zerg group this season. RorO is a fellow KeSPA Zerg and on a ZvZ tear lately, and Curious is Life's teammate. So, depending on how well-mannered Soulkey is, we'll either see Symbol or Losira being picked here. But, given Symbol's downright bad vZ record lately, it seems more likely that Soulkey will take him.

Sixth pick: PartinG: Curious
The teammate factor is absolutely something to consider here, but Parting and Curious are no longer on the same team, and the benefits of picking Curious may well outweigh the discomfort of picking a former teammate in Parting's mind. Parting has an insane PvZ record, he should be very confident against Curious with how well he should know Curious' style.

Seventh pick: BboongBboong: Taeja
Although BboongBboong beat a Protoss on his way to the Round of 16, his ZvT is probably still his best match-up. Bomber absolutely dominated his Ro32 group and is an unlikely pick, so this pick becomes a toss-up between Gumiho and TaeJa, with Taeja being the more attractive pick due to his recent injuries.

Eighth pick: Marineking: Gumiho
Marineking wanted nothing more than to get into MC's desired Protoss group, but in this hypothetical scenario, there's no way for him to get a Protoss opponent. What he has left is a choice between his shaky TvZ or his still very strong TvT. Unless Marineking goes insane and picks a strong ZvZer in the hope that DRG will go out alongside him, Marineking has no choice but to pick a Terran. Out of the two available Terrans in Gumiho and Innovation, Gumiho's record in combination with being demolished by Bomber in the Round of 32 should convince MKP that he's the player to pick.

Ninth pick: Gumiho: Innovation
Gumiho advanced to the Round of 16 by beating last season's champion Sniper, proving that his TvZ is nothing to joke about. His TvT is also strong as he has shown many times before, but his crushing defeat at the hands of Bomber may convince him to pick a Zerg and not a Terran to be safe. Still, I cannot help but feel that Innovation's TvT is something he'll consider a safer pick than RorO or Losira, who have both shown powerful ZvT lately.

Tenth pick: Curious: Losira
Curious will both want to avoid picking a teammate in Bomber and RorO's excellent ZvZ. Although Losira stated that he is very confident in his ZvZ skills, Curious is no slouch in the match-up and will likely be more than happy to take the Cat over Bomber and the Samsung Zerg, especially given the fact that he almost certainly knows Losira better than he does RorO.

Eleventh pick: Taeja: Bomber
Bomber was an absolute menace in the Round of 32, stomping both Gumiho and Sniper. Still, I can't help but feel that Taeja's supposedly shaky wrist situation in combination with his already weak vZ makes going for the potential TvT against Bomber a more attractive choice than playing his worst match-up against RorO who is completely confident in the match-up.

Twelfth pick: Symbol: RorO
The only remaining player at this point is RorO, which also means that Life, Soulkey and Symbol all end up in Group Z for the second season in a row.

The groups would then end up being:

Group A: DongRaeGu, MarineKing, GuMiho, Innovation
Group B: Squirtle, BBoongBBoong, TaeJa, Bomber
Group C: MC, PartinG, Curious, LosirA
Group Z: Life, Soulkey, Symbol, RorO.

Believable, or unbelievable? In any case, this is just one way the groups could turn out. The players are always wheedling and dealing during the group selections, so you never know what kind of strange arrangements may come to pass. In any case, feel free to your own mock drafts in the replies!