LoL Worlds 2021 - Group A Midpoint

The first round robin of the League of Legends World Championship Group Stage is done, which means we're halfway to the final bracket! Tomorrow's matches are all about Group A, so let's check in on their standings and see what we can predict going forward.


Current standings

Here is each team's current record, after the first six games in the group:

Team Name Wins Losses
DWG KIA 3 0
FunPlus Phoenix 2 1
Rogue 1 2
Cloud9 0 3

Cloud9 have struggled to find any real footing, though they did almost win their game against FunPlus Phoenix in what would have been a spectacular upset. DWK KIA are sitting at a pretty 3-0 right now, after taking a game against the other obvious frontrunner. While it's looking like DK and FPX are the likely two teams to move on, FPX and Rogue are only a single game apart. RGE are the weaker team, according to the market, general sentiment, and my model, but the opportunity for a huge upset is there.


Simulation

I had previously simulated all twelve games of Group A. With half of the games played, I can simulate the second half and combine it with each team's current record, to see how things are likely to play out tomorrow.

Expected record

I had originally predicted the following expected records:

Team Name Wins Losses
FunPlus Phoenix 3.6 2.4
DWG KIA 3.24734 2.75266
Cloud9 2.62558 3.37442
Rogue 2.52708 3.47292

With half the games locked in, things will shift around. DK, for example, are likely to end up with significantly more than 3.2 wins given that they already have three under their belt. Here's the new expected record for each team:

Team Name Wins Losses
DWG KIA 4.71742 1.28258
FunPlus Phoenix 3.78002 2.21998
Rogue 2.19142 3.80858
Cloud9 1.31114 4.68886

C9 are still expected to pick up 1.3 games, on average. They range from even (against RGE) to disfavored in each matchup, so there's a decent chance of another 0-3 round, but they're likely to grab at least one game, and there's a small chance of a double comeback (less than 30% on top of their one likely win).

DK are looking at an incremental 1.7 games, on average. None of their matchups are guaranteed, but they have a realistic chance at a 6-0 finish as well.

FPX are looking at a similar incremental number of wins, as my model finds them to be at just about the same caliber as DK. Starting a game up means that DK are very likely to place higher than FPX, but FPX would have to make multiple catastrophic errors to not at least take second.

RGE, like C9, are likely to pick up one more game, at most. There's always the chance of a second victory, but it's not looking great.

We can also map out the probability of each possible record that each team can achieve:

Most likely outcomes?

  • DWG KIA: 5-1
  • FunPlus Phoenix: 4-2
  • Rogue: 2-4
  • Cloud9: 1-5

Expected placement

Since tiebreakers are looking less likely with a perfect 0-1-2-3 spread of wins so far, records are fairly synonymous with final placement. Still, there's the possibility that FPX and DK tie at 4-2 or 5-1, and some edge chance that Rogue go 2-1 while FPX go 1-2 and they both tie at the edge of elimination. Taking tiebreakers into account, where is each team likely to place?

Team Name % 1st Place Median Place
DWG KIA 71.45% 1st
FunPlus Phoenix 25.18% 2nd
Rogue 3.23% 3rd
Cloud9 0.14% 4th

DK are almost guaranteed first in this group. The literal only way for C9 to get first is for them to win all three of their games, and every other team to win and lose such that the final records are 3-3 for every team, and then C9 win some crazy tiebreakers. Apparently that happens close to one in a thousand times, but I wouldn't really count on it.

DK may be a lock for first place, but the top two teams move on. Intuitively, those two teams are almost certainly DK and FPX, but let's see the numbers:

Team Name % Top 2
DWG KIA 96.02%
FunPlus Phoenix 80.25%
Rogue 19.20%
Cloud9 4.53%

There's a one-in-five chance that RGE pull off something crazy and supplant FPX, but again, this group is looking pretty much guaranteed.

For good measure, here is the distribution of each team's possible placements:

The expected spikes are there, with each team being most likely to finish in a different place.


Wrapping up

We had previously claimed that this group had two obvious tiers of power: DWG KIA and FunPlus Phoenix in the A-tier, and Rogue and Cloud9 in the B-tier. That prediction was probabilistic, so the first half of this double round robin could have played out differently, but it went pretty much according to expectations.

With only six games played by each team, a single extra win or loss in the first three can drastically change a team's final outcomes. The possibility of DK being up a game against FPX in the first three was embedded in the original prediction with some probability, but now that that's reality, they are almost certain to end up ahead of FPX.

There's still some chance of an upset coming from C9 or RGE. It likely won't change the fact that those two teams are getting eliminated, but might affect the relative placement of DK and FPX in first and second. Upsets can make for good wagers, so keep up with our suggested bets to see how our predictions stack up against the market, and where we would suggest staking your money.


For more in-depth match analysis, discussion, and predictions, as well as quantitative betting strategy tips, subscribe to stay in the loop! You can also follow us on Twitter or Facebook for quick prediction snapshots of upcoming matches. Alacrity is your fastest path from simply gambling to beating the market. Stay ahead of the game!