Wednesday, February 17th Recap
DAMWON Kia vs KT Rolster (Net: no action)
This game one went full-on Ringling Brothers in a hurry. Canyon brought out the Kayn, a pick I think that we’ll be seeing from time-to-time on this patch and the next got off to a great 3-1 start at the expense of KT’s bottom lane but other than an AP support Jarvan DAWMON just lacked front line which just allowed KT to win any 5v5 situation and gave very little room for error to DAMWON. This game would have ended way sooner but DAMWON managed to steal the first and third barons (yes…) and turn this into one of those 40+ minute, nobody knows what’s going on, games where you should just queue up the Benny Hill theme song. Eventually KT took it down with an elder drake but the fact that this was even close after the start that they had perhaps speaks to problems closing. DAMWON’s comp was not a hard one to just group up and siege on and they never really did that.
Game two the script was flipped. DAMWON jumped out to one of their classic huge gold leads in the first twenty minutes with an Ornn centric team composition but struggled to deal with Doran’s Camille creating havoc on the map. KT clawed it all the way back to a gold lead in their favor before punting it away setting up for baron #3.
Game three was a clinical DAMWON chokehold win off of a ridiculous performance from Canyon. The dude is good, who knew.
I don’t know if it was just too much celebrating over the holiday or what but just like the next series, these were really weird games today. It was way out of character for both of these teams, DAMWON especially but sometimes when you get two comps that are running just support tanks you get situations like game one. Game three easily could have devolved in a similar manner. This might just be a reflection of the LCK scrim metagame right now being a little loose and less “traditional” in team composition construction or maybe this was just an odd day.
For what it’s worth, the model suggested KT +1.5 maps and the OVER time totals and kill totals which would have hit in all three games. I figured this would be a much more dominant showing from the champs so I just passed.
Liiv Sandbox vs Fredit Brion (Net: -5.17 units)
Game one pitted Sandbox’s 1-3-1 Camille + Twisted Fate split composition. Summit did his usual and built himself into monster status and Brion had no way to deal with it but a very VERY greedy gold card attempt by FATE when siegeing mid inhibitor tower turned into a prolonged chain of events that all cascaded off of one another. Sandbox really shouldn’t have been grouped as five at any point and then made a critical error once they finally did. That overextension triggered an engage by Brion off of a Zoe bubble from Lava tagged Croco and as Sandbox attempted to retreat a really disjointed and bizarre fight resulted in an ace. An ace, that turned into a baron force by Brion and an extremely close fight that Brion managed to ace them at yet again. Back-to-back aces then turned into a third cloud drake for Brion. At this point Brion’s team fight composition was online and with barons and soul point dragons constantly forcing Sandbox to group their time had run out.
(moments before disaster)
Game two was the opposite with Sandbox having a superior scaling composition but they couldn’t say no to any fights and couldn’t get on the same page in those fights and just spoon-fed Brion a lead big enough that they were far enough ahead that scaling didn’t matter.
I’m not going to lie this was a frustrating series to watch because Brion were severely outplayed in game one and it literally looked like Sandbox were just tilted in game two. They were playing with no patience, no understanding of what was important in the big picture, and it looked to me in my VOD review of the game that there were split calls in just about every circumstance. Sandbox are reminding me an awful lot of Vitality who have a similar profile.
Sandbox are quickly becoming the LA Chargers of League of Legends; an immensely talented roster but always finding new and creative ways to throw games away or just straight up getting unlucky. To me, that’s a coaching issue and this staff should be held accountable. Prince was brought in from FunPlus and will be eligible to play in their next match but unless he’s going to be an in-game leader or provide a positive contribution in that realm, I’m not sure how much of a difference it will make. We’ll see I guess.
This was an absolute slam dunk play by my model which lined this a lot more aggressively than I would have at -530 / +380 (too much) we had a double digit edge (18+%) on the book price here. Even if you significantly tone that down this was still a great spot with Sandbox getting a huge win over Hanwha Life before the break, getting the week to reset their mental game. Brion had basically nothing to show except the DAMWON series and they still don’t.
This was a massive underperformance but at some point we need to ask ourselves if Sandbox are an outlier. Are they the Chargers? A team that pops on the stat sheet but can’t put it all together? It’s sure seeming like it but as I look across the league and see a lot of quality teams just getting really sloppy in closing games out, I start to question whether or not this is all a more big picture issue. It’s still February, we’re halfway into the Spring split but that’s still rather early in the year, typically we start to see more polished looks around now but maybe that isn’t the case this year. I’ll be monitoring this closely.
LCK Net Total: -5.17 units
Daily Net Total: -5.17 units
Two very VERY sloppy series to re-open the Spring split which is very uncharacteristic of the LCK. Perhaps a little too much partying over the break who knows.
One thing I want to make perfectly clear is that the vast majority of these upsets or surprises so far in the LCK have a lot more to do with poor play by the teams that should win than overperformance or an “underrating” of certain teams. DRX? Brion? These teams aren’t very good they’ve just been the beneficiaries of some downright awful play by their opponents. They aren’t forcing mistakes with good defense or building leads through proactive play, they’re simply there on the other side of the rift to recover the fumble. As frustrating as it is when you get into ruts like this you’ve just got to stay the course, keep grinding, and continue to hone and improve your process. Maybe something has changed? Perhaps there’s a reason all these good teams are playing terribly? Perhaps there is not and we’re just catching some brutal variance.
A Brief Moment of Reflection:
I have an on-going joke with The Gold Card Podcast crew about my “Inverse Confidence Model.” The joke is that when I’m very confidence or bullish on a play it seemingly always finds a way to miraculously lose while my normal, day-to-day grind has been incredibly consistent over the years. While it’s an inside joke that has been the case so far this season (and why I typically don’t fire heavily early in the season like I have in the past). There is a lesson to be learned from all of this. I could sit here all day and just blame bad luck but let’s take a look at my resume so far this season and critique my positions.
Here are all of my 3+ unit selections from 2021:
On 3+ unit plays this season I’m a dreadful -6.32 units so far (-12.6% ROI). That’s terrible so what’s going on here? Let’s break down each of these positions starting from the bottom (beginning of season).
- T1 moneyline -303 vs Hanwha 3.03 units (win): The interpretation from me was that this was an underrated/overrated situation. I used T1’s priors, graded Hanwha as a league average team (used Afreeca from Summer), and compressed it closer together to account for some early season rust and variance. I sitll couldn’t come up with a number this cheap for T1 so I fired heavily on day one, something I don’t usually do. Worked out. This was a good wager in my opinion.
- Afreeca moneyline -139 vs DRX 4.17 units (loss): Again, even in optimistic scenarios for DRX I could not get anywhere remotely close to this. Afreeca had some change but carried over the prior top trio and even if I diluted Summer Afreeca I would made this line more than double what this was against a collection of players I graded as slightly below average but gave an early season bump to until they get figured out. Afreeca played horribly. This was a good wager in my opinion.
- EDG -1.5 maps vs LGD @ -175 3.5 units (win): LGD are awful, kept the same weak coaching staff, and had massive turnover. EDG have a stacked roster and were on the up and up in Summer, expected them to be a league favorite and sure enough they’re atop the table right now. This was a good wager in my opinion.
- 100 Thieves moneyline -383 vs Immortals 3.83 units (win): This was for the quarterfinals of the Lock-In tournament and 100 Thieves had met my expectations and then some while Immortals were playing their academy team. It was also a best-of-five which tends to cut down on variance for moneylines. This was a good wager in my opinion.
- Liquid -385 vs FlyQuest 3.85 units (win): Also from the Lock-In quarterfinals. Best-of-five good vs bad. This was a good wager in my opinion.
- BiliBili moneyline -400 vs LGD 3.0 units (win): BiliBili had looked shaky here and there but had mostly handled their weaker competition and were competitive against their stronger opponents early. LGD were better than I thought in their first series but immediately looked worse in their next few. This was a debateably good wager but maybe a little heavy handed in my weight on it.
- KT Rolster moneyline -133 vs Nongshim 3.99 units (loss): This was a weird situation where Nonghsim opened as -141 favorites and it was bet all the way to KT -133 (-145 by close). I had KT in the ballpark of a -210 or more favorite so it was still a tremendous value even at the closing price. I personally got this at +115 but for writeup and record keeping still backed it at -133. In hindsight, this was a great play that deserved the extra weight but I maybe should have capped it at 2x instead of 3x. It also looks a lot better now given how these two teams have looked in the rest of the season. My model would make this -205 for KT right now. This was a good wager in my opinion just too heavy.
- FPX moneyline -385 vs LNG 3.85 units (win): FPX were actually underrated in this spot according to the model. They had the second best kill agnostic gold per minute of any team on earth at the time of this match (only to DAMWON). LNG, meanwhile, were a decent team but most definitely being overrated by the market as they were coming off of strong performances against decent teams until that point and had been spoon-fed a few victories by their opponents. FPX should have been twice this price in this spot. This was a good wager in my opinion.
- G moneyline -400 vs Nongshim 4 units (win): Again, an elite caliber team against a team that was not only overrated by the market in pre-season and continued to be, but one that had been massively underperforming in Nongshim. Also mid difference. This was a good wager in my opinion.
- T1 moneyline -625 vs Brion 3.125 units (win): This wager was made with the caveat that Faker starts. Once it was announced near game time I fired. Again, it looks weird because of the big number but this is a cheap price for T1 I made it closer to -800 at the time of wager. This was a good wager in my opinion and more appropriately weighted.
- DAMWON moneyline -625 vs DRX 6.25 units (win): Similar reasoning, even bigger line (made it closer to -1300). This was a good wager in my opinion.
- Team Liquid moneyline -357 vs TSM 3.57 units (loss): This was arguably my dumbest wager of the entire season. For as good as Liquid have looked overall and as poor as TSM had until that point, Liquid weren’t unflappable and TSM were a team I expected to improve. This was also for a best-of-one in February….I’m a moron! This was an absolutely awful wager.
- Sandbox moneyline -192 vs Brion 3.84 units (loss): Instinct is to attack a weak line but Sandbox had a history already of being “Chargers-esque” in their unique ability to find ways to lose in frustrating fashion. Still, this was absolutely a good wager but I perhaps should have done a normal “to win one unit” staking instead of a 2x. Not every massive edge needs to be scaled up for. This is a separate discussion regarding stake weighting for another time but I should have been a bit more subjective on this and just wagered 1.92 to win 1. This was a good wager in my opinion just overweight considering consistency issues.
So really I’ve made one awful wager that I’d like to have back (Liquid), was way too heavy on KT against Nongshim and too heavy on Sandbox vs Brion based on massive edges but with inconsistent or two-faced teams, and the Afreeca loss looks like a bad beat in hindsight. It sure as hell feels terrible to lose plays you’re really confident on but the logic was sound in most of these I just need to clean up the weighting and not fire heavy in February in best-of-ones like a moron (I know better).
Anyway, I like to do this from time-to-time to show everyone that this isn’t easy and perhaps more importantly, to hold yourself accountable and continue to improve. I’m human. I make mistakes and I’m not one to sit here and pretend that I don’t. I own it. I absolutely trust my process, have been beating closing lines, and I’m confident that long term I’ll be successful in this market despite the rocky start so far.
LOL Champions Korea
Week 5 – Day 2
Nongshim RedForce +209 (+1.5 maps @ -182, -1.5 @ +550)
vs
Hanwha Life Esports -263 (-1.5 maps @ +140, +1.5 @ -1000)
Total Maps Played: 2.5 (over +102 / under -130)
Kill Total: 22.5 (over -127 / under -103)
Kill Spread: +4.5 -111 / -4.5 @ -118
Team Kill Totals: 10.5 / 14.5
Time Total: 32:00 (over +106 / under -139)
(Individual player model positional ranking is a “living” number comparing individual performance against other players at their position)
Hanwha have come back down to earth a bit after a blazing hot, blood-filled start to the season. This team is still very good and has plenty of room for growth but currently they sit as the best of the three mid tier teams in the economy/objective model at #4 ahead of Afreeca and Sandbox but significantly further behind T1, Gen.G, and DAMWON. Perhaps the most intriguing aspect of Hanwha is their reliance and dedication to kills. Frequently I use kill agnostic economy metrics to see “if this is real” with teams but you do get certain styles of teams that are just bloodier through and through. Most of the time they’re not as good at League of Legends as they are for your fantasy lineups because it’s a less repeatable, more inconsistent foundation upon which to build your game-by-game economy.
I’m still approaching Hanwha with tempered expectations. There are a lot of red flags and indicators in the underlying data that heavily suggest a regression but they are also improving on film overall and they have Chovy. In short, I think Hanwha aren’t as good as we’ve seen so far YET but they could be… weird right? They feel to me like they’re going to be the gatekeepers and firmly in this tier below the elite teams but I could also see them continuing to improve. I want to see them take a more disciplined approach to the game but there’s a chance we never really see that and this just is what they are. If that’s the case they’ll be a higher variance version of Afreeca.
Nongshim have been a big of a puzzle this season. I wasn’t as high on them as some of my peers but I think it’s safe to say that they’ve underperformed even my modest expectations thus far. That said, there are good things going on here. Nongshim rank 5th in Kill Agnostic gold per minute and 4th in kill agnostic gold differential per minute despite their overall economy metrics being less than exciting. So how do we explain this anomaly especially considering they have BY FAR the worst gold differential at 10 minutes, which is usually around when a lot of the tower-based gold is coming into play? Nongshim are digging themselves into a massive hole early in games and have struggled to claw their way out of it. They’ve only played eight games this season where they were within 1000 gold or leading by more than that in the first fifteen minutes and the vast majority of them have been WELL below their average differential of -1386 (they’ve lost all but one of these).
Typically I don’t like teams that display a lack of proficiency early in the game because that’s a significantly more difficult obstacle to traverse than cleaning up transitional decision making in the mid game. If you have the foundation in place you’ll get more experience and film to learn from. If you’re never getting those leads you never learn how to close. Regardless, it’s not impossible to overcome this deficiency. If Nongshim can clean up their early game and not bury themselves in a multiple thousands of gold deficit in most games they’ll have a much better second half performance.
I was going to go into more detail but Bay has still been the worst performing starting mid laner in both the eye test and my individual player model. He gets to face Chovy. Mid diff, that’s all I’ll say.
Looking my tale of the tape sheet here there isn’t really a huge “early vs late” style edge in this match. Hanwha have had a few big openings but for the most part they’re not jumping out to huge leads on people which would be how to attack Nongshim so far this season.
The economy/objective model likes Hanwha quite a bit at this price point showing a high single digit edge on the market price for the series and a low double digit advantage on the Hanwha spread. I personally think it’s probably closer than that. Both teams are coming off of extra rest and preparation time which typically lends to more competitive matchups (although who knows after what we saw this morning this could be a fiesta). Nongshim should continue to improve, Hanwha it could be regression or progression. Hanwha are also playing a higher variance form of League right now to toss another wrench in our plans here. I’m not in love with this play by any stretch but I do think there is enough value on Hanwha in this spot to back them at standard weight.
Other Markets:
Kill Total Model Projections:
Odds-Weighted: 27.363
Time-Implied: 24.906
Underdog Win: 23.475
“Gelati” Total Projection: 25.669 kills
Combined % of Games OVER Kill Total: 55.1587% (HLE 71.43% / NS 38.89%)
Volatility Rating: HLE 0.30276 / NS 0.35757 (League AVG: 0.3184)
Suggested Model Play: Under 22.5 (small edge)
My first thought for this series was OVER OVER OVER and after seeing today’s games that was amplified. Of course, there is a pretty strong chance Hanwha just run this series over but given the aforementioned higher variance, kill-reliant approach I think overs are more appealing than my model is ever going to think. This is a matchup where you have a very low combined kill team and the highest but I’m expecting Nongshim to come out with a gameplan for this matchup and keep this competitive. I’d expect a lot of back-and-forth and like the over quite a bit more than my model does here.
Time Total Projection:
Combined AVG Game Time: 31.689 minutes
Combined % of Games OVER Time Total: 69.44% (HLE 66.67% / NS 72.22%)
Volatility Rating: HLE 0.08219 / NS 0.18082 (League AVG: 0.14566)
Suggested Model Play: OVER 32:00 @ +106 (double digit edge on market)
Only a third of Hanwha’s games have gone under this total and even less for Nongshim (27.88%). Sure, this is expected to be lopsided but Hanwha haven’t exactly been super clean in closing games out and I’d expect Nongshim to have some kind of game plan in place here to hide Bay for as long as possible. I doubt they’re going to rough and tumble with HLE but Hanwha have dragged a lot of teams into that this season. I also think this will likely be more competitive than it looks.
The other derivative markets don’t offer a lot to work with here. Nongshim first dragon presented the best value. I like the idea of playing OVER 4.5 dragons slain given how I think this series could play out but these teams are only averaging over that in 43% of outings combined (33.33% / 52.63%).
My Picks:
(This moved to Hanwha -244 during writing, even better)
Moneyline: Hanwha -244 (2.44 units)
Map Spread: Hanwha -1.5 maps @ +147 (0.5 units)
Time Total: Map 1 OVER 32:00 @ +106 (1 unit)
Time Total: Map 2 OVER 32:00 @ +106 (1 unit)
Time Total: Map 3 OVER 32:00 2 +106 (1 unit)
Kill Total: Map 1 OVER 23.5 @ +100 (1 unit)
Kill totals on maps 2 and 3 are way overpriced and adjusted upward. Passing on those I’ll just take game one.
Afreeca Freecs +344 (+1.5 maps @ -120, -1.5 @ +1062)
vs
Gen.G -455 (-1.5 maps @ -105, +1.5 @ -3333)
Total Maps Played: 2.5 (over +111 / under -141)
Kill Total: 22.5 (over -116 / under -112)
Kill Spread: +6.5 @ -108 / -6.5 @ -120
Team Kill Totals: 8.5 / 14.5
Time Total: 32:00 (over -123 / under -105)
(Individual player model positional ranking is a “living” number comparing individual performance against other players at their position)
After a bit of a bizarre opening to the season Afreeca appear to be piecing things together. Even though they’ve lost their past two matches to KT and Hanwha Life, they were much more competitive than in previous series and prior to that they swept Nongshim and T1Clozer. It makes a lot of sense if you think about it. New bottom lane comes in, takes some time to integrate and get their footing and since then they’ve looked a lot better. Afreeca aren’t exactly dominating people but I do expect them to continue improving as the season goes on an settle in as one of the mid-table teams.
By most measures Gen.G have had a rough start. This is their worst four week stretch since 2019 and Bdd had an absolutely brutal start to the season but has been recovering nicely improving series-to-series.
They’re still leaps and bounds ahead of most teams in most metrics. That includes Afreeca.
Gen.G are second only to DAMWON in kill agnostic gold per minute, are denying the second most gold to their opponents, and possess the best first 20 minute gold differential per minute in the LCK even ahead of DAMWON.
From a pure economy perspective Gen.G are a way WAY better team here and considering the “rough” start they’ve had that’s impressive. The model showed a substantial edge (double digits) on the Afreeca side of this matchup but let’s look at this with some more granularity.
Gen.G have been a great early game team. They’re #2 in gold differential at 10, #3 at 15, #2 at 20, and have by far the best rest of game (post 20) gold differential per minute in the LCK. Like DAMWON, they’ve been a herald-centric team preferring to jump start the economy rather than scaling. Afreeca have been the same overall profile, build early leads through herald snowballs albeit worse at it. This creates some volatile game states and it’s going to test who the more versatile team is. Gen.G haven’t really had to deviate from their approach regardless of the opponent but we’ve seen how volatile that can be.
TL:DR – Afreeca are a diet version of Gen.G which can be both an advantage and a disadvantage.
I’m going to take a small position on the underdogs here for a few reasons.
First, fighting fire with fire tends to create volatile situations. When it’s two stronger late game teams it tends to be more the case that the better team with better economy wins more often but when you’ve got two tempo oriented teams it’s more variant.
Second, I’m expecting improvement from both of these squads but much more from Afreeca than Gen.G who are already trending back toward their “normal.” In other words Afreeca have more room to grow and they’re still showing value here even with my concerns over their generally weak economy metrics.
Third, Afreeca are individually stronger at two positions and typically when you see underdogs of this magnitude they’re outclassed across the board.
Fourth, I think there is a reasonable chance we see Karis and/or Flawless again here for Gen.G (not putting much weight on this, just icing on the cake).
Fifth, preparation time probably serves the underdog slightly better here as well.
Other Markets:
Kill Total Model Projections:
Odds-Weighted: 27.401
Time-Implied: 25.437
Underdog Win: 23.949
“Gelati” Total Projection: 25.592 kills
Combined % of Games OVER Kill Total: 57.353% (GEG 50% / AF 64.71%)
Suggested Model Play: OVER 22.5 or 23.5 (small edge)
Model shows a very slight edge on the market price for the over and hates the under. I like time totals more but I’d lean toward the over.
Time Total Projection:
Combined AVG Game Time: 33.527 minutes
Combined % of Games OVER Time Total: 73.24% (GEG 70% / AF 76.47%)
Volatility Rating: GEG 0.17424 AF 0.15547 (League AVG: 0.14566)
Suggested Model Play: OVER 32:00 @ -123 (double digit edge on market)
This is a slam dunk over play on the time total especially given that I think this will be a surprisingly competitive series.
In terms of other derivatives, both of these teams have been stellar at over 50% in all four firsts which suggests some play on the underdog firsts here. The best value compared to market is AF first blood but it’s also a strength for Gen.G. I like taking first herald here as it’s the next biggest edge but you could take both herald and dragon each at plus money and unless Afreeca get steamrolled you’ve got a very high chance of hitting one or both.
The model also likes over 12.5 towers quite a bit here as well with a combined average of 43.53% of games with each of these teams going over that total (40% for AF / 47.06% for Gen.G)
My Picks:
Map Spread: Afreeca +1.5 maps @ -119 (1.19 units)
Moneyline: Afreeca +344 (0.25 units)
Map Spread: Afreeca -1.5 maps @ +1062 (0.1 units)
Time Total: Map 1 OVER 32:00 @ -123 (1.23 units)
Time Total: Map 2 OVER 32:00 @ -123 (1.23 units)
Time Total: Map 3 OVER 32:00 @ +100 (1 unit)
Prop: Map 1 Afreeca first herald @ +123 (0.5 units)
Prop: Map 2 Afreeca first herald @ +123 (0.5 units)
Prop: Map 3 Afreeca first herald @ +123 (0.5 units)
Prop: Map 1 OVER 12.5 towers @ +171 (0.5 units)
Prop: Map 2 OVER 12.5 towers @ +171 (0.5 units)
Prop: Map 3 OVER 12.5 towers @ +171 (0.5 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)
