Sunday, February 14th Recap
100 Thieves vs Team Liquid (Net: -1.0 units)
Immortals vs TSM (Net: no action)
Evil Geniuses vs Dignitas (Net: +1.5 units)
Cloud 9 vs FlyQuest (Net: -2.64 units)
Golden Guardians vs CLG (Net: -2.99 units)
LCS Net Total: -5.13 units
Daily Net Total: -5.13 units
Brutal Sunday. Really a brutal weekend. LCS remains a very high volatility league at least for the time being. It’s still rather early but with quick turnarounds for a few of the teams that were looking terrible caught me off guard. Few of the positions I had were straight up mistakes. I haven’t made a lot of those this year but I was way too heavy on EG and Liquid last weekend for this early in the season. I always preach not laying money with heavy favorites early on even when it looks like a decent value, especially in the best-of-one leagues. Didn’t follow my own advice. Lesson learned, back to the LCK.
LOL Champions Korea
Season Long Trends and Macro Discussion
40 total matches, 98 total games
Average kill spread is +/- 5.525 kills
Favorites have an average team kill total of 13.975 kills
Underdogs have an average team kill total of 9.1 kills
Favorites are 25-15 straight up, 15-25 against the map spread.
Favorites are 43-55 against the kill spread and have covered their team total 41 out of 98 games (41.8%).
Underdogs have covered their team total 59 out of 98 games (60.2%) and have swept a series 2-0 on 7 occasions (17.5%)
Kill Totals have gone UNDER in 52 out of 98 games (53.06%). There have been an average of 24.45 kills per game.
Average Margin of Victory is 9.86 kills with the most frequent margin of victory being 10 kills.
Games have an average time of 32:50 with 54.7% of games eclipsing the 32:00 mark, 47.6% over 33:00, and 31.7% over 34:00.
On the surface level there isn’t really anything surprising here except maybe underdogs covering their team total so often. In matches where the underdog has covered a team total in two or three of the games, the average has been 9.25 kills, higher than the overall average. For the most part, these matches have been extremely large favorites or matches between two closely matched teams (by the odds). A lot of the more moderate favorites have been covering their kill spreads but most favorites in the LCK have struggled to cover these large spreads.
A little more granular on the outlier scenarios.
When there is a favorite of -300 (1.33) or greater:
Favorites have only covered the map spread in 11 out of 20 matches (55%)
Favorites have covered the kill spread in 20 out of 46 games (43.5%)(avg spread: 7.6)
Favorites have covered their team total in just 20 out of 46 games (43.5%)(avg total: 14.9)
Underdogs have covered their team total in 29 out of 46 games (63%)(avg total: 7.75)
Kill Totals have gone UNDER in 28 out of 46 games (60.9%)(avg total: 23.2)
Time Totals have gone OVER in 26 out of 46 games (avg total: 31.55)
In short, favorites aren’t dominating the way you’d think which makes a lot of sense given the lower scoring matchups we’re seeing. Generally speaking, favorites are still getting the job done but big favorites are struggling to dominate their opponents. This could mean spreads need to be adjusted downward, which has already happened, or just that there is a lot more parity in the league than anticipated. Just like in football, it’s much harder to cover big spreads when there is a low projected total and that follows suit here.
TL:DR – Big favorites struggled early which is no surprise but it is something to keep an eye on moving forward to see if this is a longer standing trend.
Teams that stand out:
T1:
As favorites, T1 are 2-5 against the map spread, 11 out of 18 games have gone under and they’ve covered their kill spread less than half the time. This makes a lot of sense given the line up shuffling and the unders we’ve been hitting with Faker in the lineup. Over when Clozer is in, under when Faker is in. The rest of the lineup doesn’t seem to have much correlation but more on that later this week.
DAMWON:
Have only covered the kill spread in 38.9% of games but have mostly delivered in terms of justifying their price tag.
DRX:
Kill total unders have cashed in 15 out of DRX’s 22 games this season (average total: 23.09 kills). Obviously they’ve been the surprising underdogs with a 4-2 record against the map spread as dogs but they’ve failed to cover the two matches they were favored in. Over 11.5 towers has cashed in 16 out of 22.
Afreeca:
Afreeca have been a bit two-faced on a per series basis but overall have settled into a rough favorite but tremendous underdog. Afreeca are 7-2 against the kill spread and in covering their team total as underdogs. They’re just 2-6 against the spread as favorites.
Liiv Sandbox:
Sandbox have had a rough bout with variance this season. It felt like they simply couldn’t buy a win despite some solid performances. With a match win before the break to end on a good note I’d expect a lot of regression but more on that later.
Sandbox have been great underdogs despite only taking down two of their seven chances in the role outright. They’re 4-3 against the map spread, 11-5 against the kill spread, and have covered their team total in 11 out of 16 tries (average total: 8.93). They’re also forcing a lot of cross-map trades with the tower total over 11.5 hitting in 10 out of 16 matches as dogs.
Don’t be fooled. Sandbox are much better than their record.
Some notes about Post-Break:
The mid-Spring break is occurring a week earlier in the season calendar than it did last year (it was in March but the LCK started later). Obviously last season we had the COVID outbreak throw things for a loop and force a condensed schedule but this was when the majority of the “upsets” happened. There were six underdog 2-0’s in the thirty matches played in the two weeks following the break and another to open the third week after. This makes some sense if you think about it. Generally the stronger overall teams and coaching staffs gain an advantage as the season goes on in LOL. Most weaker and middle-of-the-table teams aren’t as strategically deep and once some film is gained the amount of improvement those teams make is typically outclassed by the ability of the stronger teams to plan against them. In other words, strong teams improve as the season goes on more than weaker teams improve with the occasional exception.
While there were a ton of upsets in those weeks following the break, MOST of the matches were chalky 2-0 sweeps by the favorites. It’s a significantly more muted advantage for the underdogs especially with most of the strong teams getting their footing beneath them. I’m sure we’ll see a lot more line up shuffling from some of these teams but I’m largely going to be looking for the favorites to control things more than they have been in the LCK but I do have my eyes on a few teams that I think are due for some positive regression.
Lastly, here are my current power numbers for the LCK as we head into the second half of the Spring split.
LOL Champions Korea
Week 5 – Day 1
KT Rolster +599 (+1.5 maps @ +163, -1.5 @ +900)
vs
DAMWON Kia Gaming -909 (-1.5 maps @ -213)
Total Maps Played: 2.5 (over +200 / under -263)
Kill Total: 22.5 (over -105 / under -123)
Kill Spread: +8.5 @ -125 / -8.5 @ -103
Team Kill Totals: 6.5 / 14.5
Time Total: 31:00 (over -108 / under -122)
(Individual player model positional ranking in paranthesis, “living” number)
(Dove presents a marked upgrade statistically speaking but I’m guessing Ucal will get the start again after some time off)
KT Rolster have been what I like to call a “three-quarter faced.” They’re inconsistent with three-quarters of their performances being underperformances and one-quarter overperformances. I did have reasonably high expectations for this roster but other than Doran being the best top in the LCK so far there hasn’t been a lot of great things to point to. Ucal has been very up-and-down to the point where he was benched for Dove and the bottom lane has been underwhelming. Blank and Bonnie have been fine.
Part of me expects KT to improve in the second half. They’ve had some flashes but they’ve also dropped series to some of the weaker teams in the league (1-2 to Brion, 0-2 to Nongshim). Overall they present as about a league average team statistically but it’s more a case of averaging the extremes of great and terrible than a consistently “average” performance. So which are we going to get? Frankly, I don’t know. More on this in a bit.
DAMWON haven’t been quite as dominant as they were in Summer but I think people forget that that was perhaps the single best season of professional LOL ever played by a team. They’re also doing it in a different looking way in some aspects than they were last year. Their gold differential at 15 minutes isn’t quite as ridiculous as it was last season but something I wanted to draw attention to is how they’re accruing that advantage compared to last season.
Kill agnostic economy is a metric I’ve developed to attempt to remove the volatility of kills impact on a team’s gold generation. Kills (and first blood) tend to be less repeatable than solid fundamental macro play. Kill agnostic gold differentials are a purer measure of how a gold lead is being generated that encompasses everything from creep and jungle gold, tower platings, and tower kills. It’s not always about scoring the kill but how you can leverage advantages from poking an opponent out of lane or being quicker to a rotation, and more. Teams with higher kill agnostic economy levels tend to have repeated success because those results can be duplicated consistently and it indicates a strong fundamental team.
DAMWON have the highest kill agnostic gold per minute of any team in the world at 1838. Let that sink in for a second. If they scored no kills, they’re generating more gold per minute than JDG’s entire economy is producing and they’re doing that all while keeping their opponents to a total gold per minute (including kills) of just over 1700. As a matter of fact there are only a handful of teams in the major leagues worldwide that can even eclipse that number in total gold per minute. All of this is to say that DAMWON are still a ridiculously good team they’re just doing it in a less flashy manner than they were for most of the Summer. Don’t let that cloud your judgment, although maybe you should let it impact your fantasy decisions.
Even if KT are playing their best League of Legends that we’ve seen this season I still have a hard time giving them a chance in this series. Coming off the break week should hopefully give them time to figure things out but it also gives DAMWON time off, something they haven’t had a lot of in the past couple of years. My model says there’s a slight value on the KT +1.5 maps here but I just can’t bring myself to do it regardless of the lineup KT announces. This is a pass on the sides for me.
Other Markets:
Kill Total Model Projections:
Odds-Weighted: 25.695
Time-Implied: 26.264
Underdog Win: 23.994
“Gelati” Total Projection: 25.923 kills
Combined % of Games OVER Kill Total: 63.125% (DWG 56.25% / KT 70%)
Volatility Rating: DWG 0.34353 / KT 0.24276 (League AVG: 0.3184)
The projections point to a high total here but both of these teams have been highly volatile in their kill total outcomes with a lot of results at the extreme highs and lows and not too much in the middle. Model obviously suggests an over. Lean over but pass.
Time Total Projection:
Combined AVG Game Time: 33.703 minutes
Combined % of Games OVER Time Total: 76.875% (DWG 68.75% / KT 85%)
Volatility Rating: DWG 0.18933 / KT 0.13935 (League AVG: 0.14566)
Model likes an over here as well but DAMWON have been winning in an average of 31.74 minutes and they have a fair amount of wins under that number. This team is capable of speedrunning anybody so I’ll stay away from this tricky price.
In terms of other derivatives I’m not seeing a lot here. The model suggests over 4.5 dragons but again, given the potential DAMWON boat race here I’m not sure I want to toy with that at all.
My Picks:
no wagers
Fredit BRION +154 (+1.5 maps @ -213, -1.5 @ +479)
vs
Liiv Sandbox -192 (-1.5 maps @ +164, +1.5 @ -769)
Total Maps Played: 2.5 (over -102 / under -125)
Kill Total: 22.5 (over -114 / under -114)
Kill Spread: +4.5 @ -128 / -4.5 @ -101
Team Kill Totals: 9.5 / 12.5
Time Total: 34:00 (over +164 / under -217)
(Individual player model positional ranking in paranthesis, “living” number)
(Route and Leo are very similar profiles and rank right next to each other)
I somewhat spoiled this one in the overall breakdowns earlier in the post but Sandbox look like a prime candidate for positive regression in the second half. Their statistical profile is not just better than their record but significantly so. My economy/objective model grades them as a league average team (#6) right around the level of Afreeca and about a half tier behind Hanwha Life from a metrics perspective. They’re also quite a bit ahead of the bottom four teams. In many ways this aligns with what I’ve been watching on film for them. They’ve had a few really bizarre games. Admittedly, some of this is their own doing but at least part of it can be chalked up to variance.
Individually Sandbox have been about an average team with a few standout performances particularly from Summit and Croco. The wrinkle here is that they acquired Prince from FunPlus right before the break and Route and Leo haven’t exactly been stellar performers so far this season. I wouldn’t call them detriments despite their individual positional rankings but they haven’t been good either.
I don’t think we’re going to see Prince in this series because the LCK has roster lock windows at a date before the season starts and then it opens at the halfway point again before closing for any trades or moves. If it’s like it’s been in the past and I’ve seen nothing to the contrary, he won’t be eligible until after this match. The teams have only played 8 out of the 9 matches in the first round robin. There could have been a rule change but I haven’t seen it. Don’t misinterpret the announcement. As always, predicting lineups in the LCK is always a wild game so keep an eye on social media before kick off here.
Simply put, Brion have been outclassed. Other than an otherworldly single series effort against DAMWON, they’ve been unimpressive. Statistically speaking they’re the worst team in the league by more than half a standard deviation. The film is a little bit more forgiving than the numbers are. Brion aren’t one of these absolute dumpster fire teams that have no idea what they’re doing at all. If you screw up against them they can punish you but they struggle to engineer their own advantages and when you simply don’t have the horses, so to speak, that the other teams in the LCK have, you’ve got to get creative. They haven’t done that.
Brion did add longtime pro Yaharong and there’s a chance we see Brion come out with a different look out of the break but they’re reminding me a lot of APK/SeolHaeOne from last season, not a terrible team but not one that’s good enough to really compete consistently in the LCK.
I absolutely love Sandbox here. Their prices this week are a reflection of their record and not their game-by-game performance level. This is a team that just swept Hanwha Life who were looking like a candidate to knock on the “elite” tier. While it’s a little rough to lose the momentum off of that, I can’t help but think Sandbox are going to come out of the break flying and I fully expect this team to be make a run at a playoff position even though they’re in a pretty deep hole.
Other Markets:
Kill Total Model Projections:
Odds-Weighted: 23.068
Time-Implied: 23.667
Underdog Win: 23.713
“Gelati” Total Projection: 23.318 kills
Combined % of Games OVER Kill Total: 66.27% (LSB 71.43% / BRO 61.11%)
I project this kill total a modest 0.8 kills over the total here but the important thing to understand here is that these teams are, more often than not, exceeding this total as evidenced above. I’ll be playing the over in this contest especially because I think we’ll some more spunk out of these teams coming out of the break.
Time Total Projection:
Combined AVG Game Time: 32.694 minutes
Combined % of Games OVER Time Total: 51.98% (LSB 42.86% / BRO 61.11%)
Volatility Rating: LSB 0.08219 / BRO 0.15789 (League AVG: 0.14566)
No plays for the time total here although I’d lean toward the under in this case.
Sandbox are definitely a herald team and have secured first herald in 77.8% of their contests win or lose. Brion are very a dragon team. Unless we see a complete 180 degree turn here, which sometimes happens when you see teams this polarized as they plan against each other, this is a pretty massive edge on the market. You could turn and play this the other way as well but I’m sticking to just one because I really do think there is a good chance Sandbox run this matchup over.
My Picks:
Moneyline: Sandbox -192 (3.84 units)
Map Spread: Sandbox -1.5 maps @ +164 (1 unit)
Kill Total: Map 1 OVER 22.5 @ -114 (1.14 units)
Kill Total: Map 2 OVER 22.5 @ -119 (1.19 units)
Kill Total: Map 3 OVER 22.5 @ -114 (1.14 units)
Prop: Map 1 Sandbox first herald @ -128 (0.64 units)
Prop: Map 2 Sandbox first herald @ -128 (0.64 units)
Prop: Map 3 Sandbox first herald @ -128 (0.64 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 can find me on Twitter @GelatiLOL
(all lines from Nitrogen unless noted otherwise)
