Sunday, August 15th Recap


Team WE vs BiliBili Gaming (Net: +0.41 units)


T1 vs Hanwha Life (Net: -2.95 units)

Nongshim vs Gen.G (Net: +2.7 units)


Fnatic vs Team Vitality (Net: +1.0 units)


TSM vs Team Liquid (Net: +4.53 units)



LPL Net Total: +0.41 units

LCK Net Total: -0.25 units

LEC Net: Total: +1.0 units

LCS Net Total: +4.53 units


Daily Net Total: +5.684 units



LCK Summer 2021

Playoffs – Round One – Day One


Below are the economy/objective model power rankings for the LCK after the Summer regular season.


LCK After Summer Reg Season
Team Power # Rank
DWG 1.274 1
T1 0.566 2
AF 0.542 3
GEG 0.424 4
LSB 0.341 5
NS 0.108 6
KT -0.374 7
BRO -0.622 8
HLE -0.712 9
DRX -1.547 10



#3 Nongshim RedForce -169 (-1.5 maps @ +119, -2.5 @ +348, +1.5 @ -357, +2.5 @ -1429)


#6 Afreeca Freecs +140 (+1.5 maps @ -152, +2.5 @ -500, -1.5 @ +257, -2.5 @ +681)


Map Totals: 3.5 maps (over -263 / under +200), 4.5 maps (over +186 / under -244)

Map Moneyline: NS -145 / AF +113

Kill Total: 23.5 (over -109 / under -120)

Kill Spread: -3.5 @ -115 / +3.5 @ -114

Team Kill Totals: 12.5 / 10.5

Time Total: 33:00 (over -151 / under +113)

Eco/Obj Model Projected Line: +122 / -157 (map), +168 / -204 (series)

Model Suggested Play:  Afreeca all ways (very strong)

NS as Favorites Win/Over Loss/Under Averages (Odds, Totals, Spreads) AF as Underdogs Win/Over Loss/Under Averages (Odds, Totals, Spreads)
Matches as Favorites 8 5 -227 Matches as Underdogs 8 4 +171
Against Map Spread 3 10 +143 Against Map Spread 9 3 -200
Against Kill Spread 14 19 4.0 Against Kill Spread 20 10 +4
Kill Totals 15 18 23.42 Kill Totals 12 18 23.33
Team Kill Totals 15 18 13.04 Team Kill Totals 17 13 9.83
Game Time Totals 20 13 32.8 Game Time Totals 21 9 32.58
Dragons over 4.5 21 12 Dragons over 4.5 20 10
Towers over 11.5 14 18 Towers over 11.5 17 13



League Rank NS Tale of the Tape AF League Rank
8.0 -94.7 Gold Diff @ 10 min 785.0 1.0
7.0 -162.6 Gold Diff @ 15 min 165.7 4.0
5.0 -22.0 Gold Diff @ 20 min 103.3 3.0
-27.4 GPM first 20 min vs Avg 4.6
5.0 5.1 Gold Diff / min first 20 15.9 3.0
3.0 118.2 Gold Diff / min Rest of Game -10.4 6.0
5.0 1775.9 Gold /min (GPM) 1760.9 7.0
6.9 Gold / min vs Avg -8.1
4.0 41.0 Gold Diff / min 9.9 5.0
4.0 0.5 Gold Percent Rating (GPR) 0.2 5.0
6.0 1608.3 Kill Agnostic GPM 1615.9 5.0
4.0 13.7 Kill Agnostic Gold Diff/min 0.0 6.0
8.0 1893.2 GPM in wins 1874.7 10.0
7.0 262.1 Gold Diff per min in wins 253.3 9.0
8.0 1609.2 GPM in losses 1625.5 5.0
3.0 -273.3 Gold Diff per min in losses -279.8 4.0
41.3 Adjusted Gold Diff / min 10.2
8.0 -21.4 Win-Adjusted GPM -39.9 10.0
7.0 -29.2 Win-Adjusted Gold Dif/min -38.1 9.0
5.0 52.2 Dragon Control % 52.2 5.0
6.0 52.2 Herald Control % 54.9 4.0
2.0 54.7 Baron Control % 44.4 7.0
9.0 Quality Wins? 11.0
45.0 % of wins as Quality 44.0

(These numbers use a composite blend of trending and season long performance)

Economy/Objective Model Series Outcome Projection (favorite):

Series Outcomes
Wins Losses Probability
1 3 24.639%
2 3 20.657%
0 3 19.592%
3 2 14.910%
3 1 12.835%
3 0 7.367%
(Series Win): 35.112%



The Summer head-to-head was split 1-1 with Nongshim winning the first 2-0, Afreeca the second 2-0.


Quantitative Analysis:

From a quantitative perspective this is a slam dunk Afreeca position but this comes with a caveat. Nongshim have been one of the worst early game teams in the LCK. You could argue whether or not this is by design or not, I’d argue sometimes that it isn’t, but one thing that’s constant is that they remain cool under pressure and seemingly always find some pivotal play in a high leverage moment. More on that below. Afreeca have been better than previous iterations in regards to punting leads but we have seen it a few times here.

The thing to keep in mind here is that the entire LCK with the exception of recent DAMWON and DRX for the whole season, have fallen within a pretty set range in terms of metrics so it’s important to see, relatively speaking, where they lie on the spectrum and the model attempts to do that by comparing just how far off of average the teams are through standard deviations, weightings, and more.

The long and short of this quantitative handicap is that Nongshim are, and continue to be, incredibly overrated. Their expected wins is significantly lower than their actual wins. Even when you look at the adjusted and more advanced metrics these two teams are very evenly matched.

Conclusion: According to model these odds should be flipped. By the numbers I happen to agree but read the caveat mentioned above.


Qualitative Analysis:

Both of these teams are enigmatic.

Nongshim continue to get away with murder in seemingly half their games and have yet to be punished for it. On one hand you could criticize other teams for not punishing or call this extremely favorable luck (perhaps a blend of the two). At some point you have to give them credit for consistently accomplishing this but just how much you give is where this becomes complicated. Teams like Nongshim are difficult to handicap in a traditional sense (looking at the data, matching it to what you see, etc) but they’re a lot like Spring EDG in that you have to give them the benefit of the doubt that they’re a “better team than their numbers.” I think this is obvious but how much better? I think that’s where I’m going to disagree with a lot of people. I don’t give as much credit to “outplays” as others do because most of the time it means you did something wrong to get there. I’m always going to be more bearish than my peers on teams like this. That doesn’t necessarily mean I think Nongshim are bad or anything they’re just overrated.

Afreeca aren’t without their own … perplexities … is that a word? This team seems to elevate to their level of competition but that on its own doesn’t tell as anything. I don’t work in “seems” and “feels” here if I can help it. In the vast majority of their games, Afreeca are in favorable positions and that ultimately leads to more wins than losses no matter how you slice it. Afreeca don’t even index heavily into early game to do so which tells me this is just a strong fundamental team that’s being undervalued by the market based on the results of this season.

If you look at this match on an lane-by-lane basis I think you could make reasonable cases for both sides in more or less all of them despite the public and/or broadcast narrative. The only real clear advantage would be Deokdam. I’m more or less looking at this as teams not individuals. Nongshim have clearly had more individuals step up in big moments this season but that’s not always as repeatable as people assume. Sort of ties in to what I mentioned above. Nongshim might have the biggest edge but I personally think Kiin is the best overall player of the ten (+) in this match for what that’s worth.

If it wasn’t obvious I’m going to be on Afreeca here. To me this is a stylistic mismatch. I much prefer teams to be in winning positions than not and Afreeca have done that much more consistently than Nongshim have this season overall. We’ve also seen them handle this team in their most recent matchup. Obviously I’m not as bearish on Nongshim as the model is but I do think this series is roughly a coin flip and that’s a sizeable adjustment both from the book price AND from the model price. If I split the difference we’re roughly even money and I’m getting +140 on the dogs here.

Conclusion: Not as bearish on Nongshim as the model is but make this roughly a coinflip, take the dogs.


Derivatives and other markets:


NS League Average AF
Combined Kills / game 23.541 22.74 22.221
Combined Kills / min 0.685 0.68 0.659
Kills per win 16.087 15.93 15.304
Kills per loss 9.709 7.39 8.144
Deaths per win 7.44 7.09 5.84
Deaths per loss 16.42 16.34 16.29
Average Margin of Victory (AMOV) 8.59 9.10 9.08
Average Margin of Defeat (AMOD) 8.32 9.20 8.81
Combined Avg Kills / game 22.881
Time-Implied Total 23.666
Book Odds Weighted 24.577
Rating Weighted 24.685
Underdog Win Projection 25.013
“Gelati” Total Projection 23.744
Volatility Rating 0.30774 0.2890 0.36119


Average Game Time 34.93 35.50
Avg Game Time (in wins) 33.12 35.94
Avg Game Time (in losses) 33.693 34.977
Combined Avg Game Time 35.215
Book Odds Weighted 34.39
Rating Odds Weighted 34.49
Volatility Rating 0.22664 0.17062 0.16606
% of Games over Time Total 60.86957 60.86957 60.86957


A lot of the derivatives for this series are either off the board or priced appropriately. I’d look over the course of the next few hours and closer to game time for more markets to open and try to snipe an Afreeca first herald at anything better than -140. Nongshim’s early games are lacking and they’re only picking up first herald in 45.65% of games compared to Afreeca’s 67.39%.  Afreeca first tower is a similar comparison and percentages so if you get a better number there I’d look to that. Kills and time totals are appropriately priced and I’ll just be sticking to sides here.

Reasonable case for the tower total over if you think these games end up with one team jumping out and the other stabilizing which has somewhat been the calling card for Nongshim but the price is close enough that I’m just passing. Intriguing that they float the 11.5 towers instead of 12.5. Number is only a short edge to the under surprisingly with a combined 51.1% of games going over the 11.5 towers with a price of -164 (ends up being 45.5% vs that 48.9%). I personally think these games will be more hard fought and tense so I’d lean to the over but there’s a chance Afreeca actually just steamroll the early games to wins.

My Picks:

Moneyline: Afreeca +140 (2 units)

Map Spread: Afreeca -1.5 maps @ +261 (1 unit)

Map Spread: Afreeca -2.5 maps @ +687 (0.25 units)

Prop: Map 1 Afreeca first herald @ -115 (1.15 units)

Prop: Map 2 Afreeca first herald @ -115 (1.15 units)

Prop: Map 3 Afreeca first tower @ -104 (1.04 units)

Prop: Map 4 Afreeca first tower @ -104 (1.04 units)

Prop: Map 5 Afreeca first tower @ -104 (1.04 units)


I believe in accountability. For years I’ve tracked all of my picks publicly. 2021’s selections will be via this spreadsheet but it isn’t updated until AFTER the games have started. The Esports Department subscribers get the first look.

Check out The Gold Card Podcast and you can find me on Twitter @GelatiLOL

(all lines from Nitrogen unless noted otherwise)

Leave a Reply