Saturday, March 27th Recap

 

ThunderTalk (TT) vs OMG (Net: +0.5 units)

TOP Esports vs Team WE (Net: -2.22 units)

DAMWON vs KT Rolster (Net: +0.82 units)

Nongshim vs Brion (Net: -2.96 units)

 

G2 Esports vs Schalke (Net: +1.105 units (+3.445, -2.,34 live)

TSM vs Evil Geniuses (Net: pending)

Short on time this weekend, these written recaps will have to be added at a later date.

 

LPL Net Total: -1.72 units

LCK Net Total: -2.14 units

LEC Net Total: +1.105 units

LCS Net Total: -2.55 units

 

Daily Net Total: -8.75 units

 

Current Week (March 22nd-28th):  (pending)

 

Last Week (March 15th-21st):  -16.94 units (-11.17% ROI)


 

LOL Pro League (China)

Week 10 – Day 7

 

 

LGD Gaming +1115 (+1.5 maps @ +300, -1.5 @ +2800)

vs

Royal Never Give Up -5000 (-1.5 maps @ -435)

 

 

Map ML Price: LGD +612 / RNG -1111

Total Maps Played: 2.5 maps (over +327 / under -476)

Kill Total: 26.5 (over -115 / under -112)

Kill Spread: +11.5 @ -111 / -11.5 @ -116

Team Kill Totals: 7.5 / 18.5

Time Total: 30:00 (over +115 / under -149)

 

 (Individual player model positional ranking is a “living” number comparing individual performance against other players at their position)

Royal Never GU Tale of the Tape LGD
-112.6 Gold Diff @ 10 min -512.7
779.6 Gold Diff @ 15 min -1395.9
1116.1 Gold Diff @ 20 min -892.8
-2.5 GPM first 20 min vs Avg -135.1
66.7 Gold Diff / min first 20 -101.2
554.9 Gold Diff / min Rest of Game 434.4
1904.3 Gold /min (GPM) 1710.2
99.9 Gold / min vs Avg -94.2
195.5 Gold Diff / min -171.4
2.7 Gold Percent Rating (GPR) -2.3
1677.3 Kill Agnostic GPM 1565.8
115.1 Kill Agnostic Gold Diff/min -104.3
1985.8 GPM in wins 1889.8
344.7 Gold Diff per min in wins 233.8
1629.3 GPM in losses 1620.4
-308.0 Gold Diff per min in losses -374.1
193.5 Adjusted Gold Diff / min -173.4
20.8 Win-Adjusted GPM -75.1
17.6 Win-Adjusted Gold Dif/min -93.3
59.0 Dragon Control % 45.9
48.6 Herald Control % 34.8
75.0 Baron Control % 36.4
9.0 Quality Wins? 4.0
33.3 % of wins as Quality 36.4

 

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

 

Suggested Model Plays (in order of strength of edge):

LGD +1.5 maps @ +300 (VERY strong)

OVER 2.5 maps @ +327 (strong)

LGD map moneyline @ +612 (strong)

LGD series moneyline @ +1115 (moderate)

 

Unfortunately for LGD, RNG still have something to play for here and that’s the #1 seed for side selection. Now, you could make a case that it’s not a huge deal and that maybe RNG don’t care about it that much but that hasn’t appeared to be the case in the matches since RNG have locked top two. I expect RNG to be here for this series.

That said, this number is completely ridiculous and is basically saying “you’re not allowed to bet RNG straight up” …. but we’ll find other ways if we want to.

I will say, that the “correct” play here is to just play LGD and don’t worry about the result because you’re getting a professional team against another professional team and we’ve seen LGD take a game off of a good RareAtom team this season so this isn’t completely unprecedented. There’s also a strong chance we see something saucy in the last game from LGD that could just throw RNG for a loop so it’s definitely not outside the realm of possibility and certainly more likely than this number.

 

Other Markets:

 

Kill Total Model Projections:

Odds-Weighted: 27.502

Time-Implied: 26.994

Underdog Win: 27.094

“Gelati” Total Projection: 27.267 kills

Combined % of Games OVER Kill Total: 39.7% (RNG 40% / LGD 39.39%)

Volatility Rating:  RNG 0.29055 / LGD 0.33094 (League Avg: 0.3046)

Suggested Model Plays (in order of strength of edge):

UNDER 26.5 @ -112 (strong)

(alt) UNDER 27.5 @ -143 (moderate)

(alt) UNDER 24.5 @ +125 (moderate)

—-

Team Totals:

RNG team total UNDER 18.5 @ -109 (VERY strong)

LGD team total OVER 7.5 @ -130 (light)

 

RNG League Average LGD
Combined Kills / game 27.616 26.45 27.084
Combined Kills / min 0.862 0.88 0.859
Kills per win 18.979 18.24 17.969
Kills per loss 9.125 9.13 8.585
Deaths per win 8.30 8.20 8.73
Deaths per loss 18.38 18.00 16.86
Average Margin of Victory (AMOV) 10.11 9.57 8.55
Average Margin of Defeat (AMOD) 9.13 9.61 9.82

 

I’m going to be playing the LGD kill spread here even though I’ve been absolutely roasted on these this season. RNG don’t win in as dominant a fashion as some of the other favorites in large part because of how they play but 11.5 is also just a ridiculous number and LGD are somewhat competent. You could also play the RNG under here as LGD tend to not give up as many kills in losses anyway. I like the RNG under slightly more. If you really wanted to go bananas you could take RNG under, game total under, and the kill spread but I’ll stick to my 1.5x dip on the RNG under.

 

Time Total Projection:

Combined AVG Game Time: 31.366 minutes

Odds Weighted / Rating Weighted: 30.84 / 31.06

Combined % of Games OVER Time Total: 54.33% (RNG 57.14% / LGD 51.52%)

Volatility Rating: RNG 0.16007 / LGD 0.18016 (League Avg: 0.16106)

Suggested Model Plays (in order of strength of edge):

OVER 30:00 @ +115 (strong)

 

RNG LGD
Average Game Time 32.09 30.64
Avg Game Time (in wins) 32.35 34.52
Avg Game Time (in losses) 31.202 28.702

Just take the over. I know we’ll probably end up getting two unders here but RNG haven’t exactly been speed running people this season with more than 57% of their games going over this total.

 

Other Derivatives:

Suggested Model Plays (in order of strength of edge):

LGD first dragon @ +104 (VERY strong)

OVER 1.5 barons @ +194 (strong)

OVER 4.5 dragons @ +132 (strong)

** denotes advantage in a category that the opponent is also strong in, must be considered

These three are all decent bets from a value perspective but also sort of play into the time total over that I’m already on so I’ll simply pass as over 30 minutes is a statistically better wager than these overs.

 

My Picks:

 

Map Spread: LGD +1.5 maps @ +300 (0.5 units)

Moneyline: LGD +1115 (0.1 units)

Map Spread: LGD -1.5 maps @ +2800 (0.1 units)

Kill Total (team): Map 1 RNG UNDER 18.5 @ -109 (1.635 units)

Kill Total (team): Map 2 RNG UNDER 18.5 @ -108 (1.62 units)

Kill Total (team): Map 3 RNG UNDER 18.5 @ -115 (1.725 units)

Time Total: Map 1 OVER 30:00 @ +115 (1 unit)

Time Total: Map 2 OVER 30:00 @ +110 (1 unit)

Time Total: Map 3 OVER 30:00 @ +109 (1 unit)

 


RareAtom +174 (+1.5 maps @ -159, -1.5 @ +462)

vs

EDward Gaming -227 (-1.5 maps @ +123, +1.5 @ -769)

 

 

Map ML Price: RA +147 / EDG -189

Total Maps Played: 2.5 maps (over +113 / under -145)

Kill Total: 25.5 (over -114 / under -114)

Kill Spread: +5.5 @ -105 / -5.5 @ -123

Team Kill Totals: 9.5 / 15.5

Time Total: 34:00 (over +146 / under -192)

 (Individual player model positional ranking is a “living” number comparing individual performance against other players at their position)

EDG Tale of the Tape RA
201.7 Gold Diff @ 10 min -259.0
333.2 Gold Diff @ 15 min 222.7
172.3 Gold Diff @ 20 min -1.9
19.7 GPM first 20 min vs Avg 41.1
57.2 Gold Diff / min first 20 27.8
458.6 Gold Diff / min Rest of Game 335.8
1863.2 Gold /min (GPM) 1839.2
58.8 Gold / min vs Avg 34.8
163.6 Gold Diff / min 91.6
2.3 Gold Percent Rating (GPR) 1.3
1663.2 Kill Agnostic GPM 1656.6
92.6 Kill Agnostic Gold Diff/min 72.4
1937.7 GPM in wins 1944.8
308.8 Gold Diff per min in wins 314.6
1639.6 GPM in losses 1668.5
-272.0 Gold Diff per min in losses -268.7
161.7 Adjusted Gold Diff / min 89.6
-27.2 Win-Adjusted GPM -20.1
-18.3 Win-Adjusted Gold Dif/min -12.5
61.8 Dragon Control % 49.4
50.8 Herald Control % 46.3
56.8 Baron Control % 58.8
7.0 Quality Wins? 5.0
29.2 % of wins as Quality 23.8

 

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

 

Suggested Model Plays (in order of strength of edge):

RareAtom +1.5 maps @ -159 (light)

OVER 2.5 maps @ +113 (miniscule)

 

With a series win RareAtom could surpass Team WE for the #6 seed but cannot catch FPX at #5. This would give them a playoff bye. EDG could potentially get the #1 seed but would need an RNG loss in the previous match which isn’t likely. I think both of these teams will be game for this match but RareAtom do get a slight motivational edge from me because the difference between a bye or not is huge in this bracket and EDG are really only battling for side selection and have already earned their byes. Regardless this should be a serious match with a playoff atmosphere.

On paper these two teams are eerily similar even down to their win-adjusted and kill-agnostic numbers. EDG have slightly better neutral objective control and that’s the primary difference. Both are scaling focused teams but are capable of snowballing early leads and playing more tempo-centric compositions.

I think EDG are a better team but I think this match is much closer to a 50/50 than my model thinks it is. This is essentially a very slightly better version of the same team and that’s not enough to justify this moneyline, I’ll take the dogs.

 

Other Markets:

 

Kill Total Model Projections:

Odds-Weighted: 27.745

Time-Implied: 24.886

Underdog Win: 25.207

“Gelati” Total Projection: 25.812 kills

Combined % of Games OVER Kill Total: 43.75% (EDG 37.5% / RA 50%)

Volatility Rating:  EDG 0.26599 / RA 0.32775 (League Avg: 0.3046)

Suggested Model Plays (in order of strength of edge):

UNDER 25.5 @ -114 (moderate-light)

(alt) UNDER 24.5 @ +105 (moderate-light)

(alt) UNDER 26.5 @ -143 (light)

—-

Team Totals:

RareAtom team total OVER 9.5 @ -125 (VERY strong)

 

EDG League Average RA
Combined Kills / game 22.500 26.45 27.151
Combined Kills / min 0.712 0.88 0.803
Kills per win 16.802 18.24 16.863
Kills per loss 8.344 9.13 12.514
Deaths per win 7.67 8.20 8.67
Deaths per loss 15.50 18.00 18.69
Average Margin of Victory (AMOV) 8.63 9.57 7.81
Average Margin of Defeat (AMOD) 6.63 9.61 9.54

RA team total is one of my favorite positions of the weekend because RA are more than capable of winning these games outright but they also average 12.514 kills PER LOSS and have eclipsed this number in 6 out of their 13 losses this season. If I didn’t think RA had a reasonable chance to win this outright I wouldn’t back this position because EDG have been squeaky clean in most of their wins and have a VERY LOW deaths per win number.

 

Time Total Projection:

Combined AVG Game Time: 32.836 minutes

Odds Weighted / Rating Weighted: 33.13 / 33.12

Combined % of Games OVER Time Total: 42.28% (EDG 37.5% / RA 47.06%)

Volatility Rating: EDG 0.14019 / RA 0.16490 (League Avg: 0.16106)

Suggested Model Plays (in order of strength of edge):

OVER 34:00 @ +146 (light)

 

EDG RA
Average Game Time 32.33 33.34
Avg Game Time (in wins) 32.25 32.46
Avg Game Time (in losses) 32.590 34.759

There’s some value on this over but I’ll be looking at this from a different perspective that’s better EV (see below).

 

Other Derivatives:

Suggested Model Plays (in order of strength of edge):

UNDER 12.5 towers @ -172 (VERY strong)

OVER 4.5 dragons @ -110 (strong)

OVER 1.5 barons @ +138 (moderate)

RA first blood @ -102 (moderate)**

RA first dragon @ +106 (moderate)**

EDG first dragon @ -137 (moderate)**

EDG first tower @ -154 (light)

** denotes advantage in a category that the opponent is also strong in, must be considered

This is about as cheap a price as you’re going to get on the under 12.5 towers in the LPL. I’ll be taking it as I project 11.35 to go down in this game and only a combined 21.3% of games between these two have eclipsed that total.

Instead of taking the plus money over 34 time total which isn’t a bad look I’d rather take the over dragons or barons which can hit more frequently in a sub 34 minute context. They’re better EV bets here.

 

My Picks:

 

Map Spread: RareAtom +1.5 maps @ -159 (1.59 units)

Moneyline: RareAtom +177 (0.75 units)

Map Spread: RareAtom -1.5 maps @ +462 (0.25 units)

Kill Total (team): Map 1 RA OVER 9.5 kills @ -125 (1.25 units)

Kill Total (team): Map 2 RA OVER 9.5 kills @ -127 (1.27 units)

Kill Total (team): Map 3 RA OVER 10.5 kills @ -108 (1.08 units)

Prop: Map 1 UNDER 12.5 towers @ -172 (1.72 units)

Prop: Map 2 UNDER 12.5 towers @ -172 (1.72 units)

Prop: Map 3 UNDER 12.5 towers @ -167 (1.67 units)

Prop: Map 1 OVER 4.5 dragons @ -110 (1.1 units)

Prop: Map 2 OVER 4.5 dragons @ -102 (1.02 units)

Prop: Map 3 OVER 4.5 dragons @ -109 (1.09 units)

 

 


LOL Champions Korea (LCK)

Week 10 – Day 4

 

 

 

Hanwha Life Esports +124 (+1.5 maps @ -227, -1.5 @ +353)

vs

T1 -161 (-1.5 maps @ +172, +1.5 @ -526)

 

 

Map ML Price: HLE +122 / T1 -156

Total Maps Played: 2.5 maps (over +106 / under -135)

Kill Total: 24.5 (over -109 / under -119)

Kill Spread: +2.5 @ -105 / -2.5 @ -122

Team Kill Totals: 10.5 / 13.5

Time Total: 33:00 (over -106 / under -122)

 (Individual player model positional ranking is a “living” number comparing individual performance against other players at their position)

T1 Tale of the Tape HLE
255.2 Gold Diff @ 10 min -109.6
1162.9 Gold Diff @ 15 min -265.6
1418.4 Gold Diff @ 20 min -564.4
53.9 GPM first 20 min vs Avg 12.3
89.5 Gold Diff / min first 20 20.2
135.4 Gold Diff / min Rest of Game 190.4
1829.8 Gold /min (GPM) 1833.9
46.4 Gold / min vs Avg 50.6
94.0 Gold Diff / min 80.1
1.3 Gold Percent Rating (GPR) 1.1
1644.5 Kill Agnostic GPM 1634.9
60.8 Kill Agnostic Gold Diff/min 47.4
1952.0 GPM in wins 1959.8
337.6 Gold Diff per min in wins 343.6
1650.1 GPM in losses 1645.2
-264.3 Gold Diff per min in losses -315.2
98.7 Adjusted Gold Diff / min 84.8
25.3 Win-Adjusted GPM 33.1
52.2 Win-Adjusted Gold Dif/min 58.2
51.3 Dragon Control % 58.7
54.3 Herald Control % 43.6
53.6 Baron Control % 55.4
17.0 Quality Wins? 8.0
68.0 % of wins as Quality 33.3

 

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

 

Suggested Model Plays (in order of strength of edge):

OVER 2.5 maps @ +106 (very light)

HLE +1.5 maps @ -227 (miniscule)

HLE map moneyline @ +122 (miniscule)

 

This match has no effect on the standings for either team since DRX can no longer catch T1, T1 can’t catch HLE, and HLE can’t catch DRX. This is purely for momentum, pride, whatever you want to point to. The reason I say this is that there’s a decent chance that despite this being a match that’s taken seriously we see these players a little bit looser than they’d normally be.

If you isolate Hanwha Life’s season to just their matches against T1, DWG, and Gen.G (the other good teams) they have a 4-9 game record and the following numbers:

Combined kills per game (per minute): 27.54 (0.858 per min)

Gold Differential at 10 / 15 / 20: +251 / -343 / -374

Gold per minute / differential per minute: 1748 / -146

Opponent Gold per minute: 1894

Kill Agnostic Gold Differential per minute: -114

If we compare these to their “per loss” averages most of them are actually better which suggests that Hanwha elevate or have at least had good games against the good teams. Hanwha are averaging -315 gold differential per minute in losses but -146 against the good teams. Pretty wild. Now, there’s two ways you could look at this. First, Hanwha really do elevate against the good teams. It makes some sense, we see it in sports and esports all the time “playing up / down” to your competition. Second, it could suggest that we see some regression. I think it’s a little of both.

Hanwha play an inherently higher variance game reliant on kills but they’re very consistent in getting opponents to play their game. If we remember back to the week one contest between these teams the Canna, Ellim, Clozer, Gumayusi, Keria version of T1 shut Hanwha out in game one, got rolled in game two, and won a high kill but never really close affair in game three. Obviously that was a long time ago but we’ve seen the veteran version of T1 is a significantly stronger lineup overall.

I like the T1 side a little more than my model does here especially because we’ve seen strong performances by Hanwha Life against the big dogs and they’ve still only made it across the finish line four times. I’ll be on T1 here.

Other Markets:

 

Kill Total Model Projections:

Odds-Weighted: 26.012

Time-Implied: 26.119

Underdog Win: 25.167

“Gelati” Total Projection: 26.147 kills

Combined % of Games OVER Kill Total: 56.25% (T1 50% / HLE 62.5%)

Volatility Rating:  T1 0.29671 / HLE 0.28689 (League Avg: 0.3104)

Suggested Model Plays (in order of strength of edge):

OVER 24.5 @ -109 (moderate)

(alt) OVER 23.5 @ -137 (light)

—-

Team Totals:

HLE team total OVER 10.5 @ -118 (VERY strong)

 

T1 League Average HLE
Combined Kills / game 26.629 24.67 26.069
Combined Kills / min 0.813 0.76 0.787
Kills per win 17.794 16.86 17.448
Kills per loss 7.719 8.59 8.844
Deaths per win 7.56 7.81 8.38
Deaths per loss 17.24 16.89 17.75
Average Margin of Victory (AMOV) 9.84 9.04 9.96
Average Margin of Defeat (AMOD) 9.24 9.03 8.81

The model likes the over based primarily on frequency but I think this number is just about right. That said, there’s nothing really on the line here and I mentioned that these two teams might be a little looser than they’d normally be so I could see the over being the play. I’m going to wait for lineups and play this if Clozer is in and pass otherwise.

 

Time Total Projection:

Combined AVG Game Time: 32.651 minutes

Odds Weighted / Rating Weighted: 32.55 / 32.57

Combined % of Games OVER Time Total: 47.44% (T1 52.38% / HLE 42.5%)

Volatility Rating: T1 0.14576 / HLE 0.10263 (League Avg: 0.15498)

Suggested Model Plays (in order of strength of edge):

none

 

T1 HLE
Average Game Time 32.95 32.35
Avg Game Time (in wins) 32.80 32.83
Avg Game Time (in losses) 33.159 31.635

I’d lean over but just like the model I’m passing here.

 

Other Derivatives:

Suggested Model Plays (in order of strength of edge):

T1 first tower @ -137 (VERY strong)**

HLE first tower @ +100 (strong)**

OVER 11.5 towers @ -116 (light)

T1 first blood @ -128 (light)

** denotes advantage in a category that the opponent is also strong in, must be considered

No plays here.

 

My Picks:

 

Moneyline: T1 -161 (1.61 units)

Map Spread: T1 -1.5 maps @ +172 (0.5 units)

 


Liiv Sandbox +104 (+1.5 maps @ -286, -1.5 @ +286)

vs

DRX -133 (-1.5 maps @ +212, +1.5 @ -400)

 

 

Map ML Price: LSB +101 / DRX -128

Total Maps Played: 2.5 maps (over +104 / under -132)

Kill Total: 24.5 (over -120 / under -108)

Kill Spread:  +2.5 @ -116 /  -2.5 @ -111

Team Kill Totals: 11.5 / 12.5

Time Total: 33:00 (over -114 / under -114)

 (Individual player model positional ranking is a “living” number comparing individual performance against other players at their position)

DRX Tale of the Tape LSB
-535.7 Gold Diff @ 10 min 353.3
-1571.6 Gold Diff @ 15 min 1362.1
-1642.3 Gold Diff @ 20 min 2317.3
-63.6 GPM first 20 min vs Avg 31.2
-54.5 Gold Diff / min first 20 44.4
-92.5 Gold Diff / min Rest of Game -154.4
1762.5 Gold /min (GPM) 1772.8
-20.8 Gold / min vs Avg -10.6
-36.5 Gold Diff / min -51.2
-0.5 Gold Percent Rating (GPR) -0.7
1619.9 Kill Agnostic GPM 1602.5
-5.7 Kill Agnostic Gold Diff/min -44.2
1906.4 GPM in wins 1976.5
249.7 Gold Diff per min in wins 339.2
1631.7 GPM in losses 1652.4
-296.8 Gold Diff per min in losses -281.9
-31.8 Adjusted Gold Diff / min -46.5
-20.3 Win-Adjusted GPM 49.8
-35.7 Win-Adjusted Gold Dif/min 53.7
41.2 Dragon Control % 45.6
31.7 Herald Control % 53.0
50.8 Baron Control % 33.3
2.0 Quality Wins? 12.0
10.0 % of wins as Quality 50.0

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

 

Suggested Model Plays (in order of strength of edge):

LSB series moneyline @ +104 (VERY strong)

LSB map moneyline @ +101 (VERY strong)

LSB -1.5 maps @ +286 (VERY strong)

LSB +1.5 maps @ -286 (VERY strong)

 

So… this requires a little bit of an explanation. The model really REALLY likes Sandbox and really REALLY dislikes DRX.

Team Power # Rank
DWG 1.27850502 1
GEG 1.010480608 2
T1 0.6974068866 3
HLE 0.421228439 4
LSB 0.2027350113 5
AF -0.2069204987 6
NS -0.7201704757 7
DRX -0.764387683 8
KT -0.8150948121 9
BRO -1.103782496 10

Why does the model like Sandbox so much? Well, they have a tremendous early game and great herald control while DRX are below average in literally every measure except for baron control. (see above Tale of the Tape)

Look, DRX aren’t going to have flashy stats, it’s not the type of team they are and I get that. I also understand that Sandbox can throw as hard as Randy Johnson but there’s just no way I’m not backing Sandbox here. Sure, they had their miracle run crushed and they might be a little demoralized in this spot but DRX have nothing to play for either.

This is a clash of styles. As many of you know I favor early game snowballing teams the longer the season goes on as they get cleaner and cleaner and more difficult to defend against. If you’re going to play the scaling team I need to see that you’re very good at damage control. DRX are decent at best at it. DRX are good team fighting team that benefitted from a simplified drafting approach by a good coaching staff early this season. That’s not enough at this point.

I’m not as bullish as the model is because I know the modus operandi for both of these teams and the numbers don’t tell the full story, but they sure as hell tell a decent chunk of it. Give me the dogs.

Other Markets:

 

Kill Total Model Projections:

Odds-Weighted: 24.818

Time-Implied: 26.394

Underdog Win: 25.535

“Gelati” Total Projection: 25.712 kills

Combined % of Games OVER Kill Total: 45.714% (DRX 42.86% / LSB 48.57%)

Volatility Rating:  DRX 0.34749 / LSB 0.26865 (League Avg: 0.3104)

Suggested Model Plays (in order of strength of edge):

(alt) UNDER 25.5 @ -123 (moderate-light)

UNDER 24.5 @ -108 (light)

(alt) UNDER 23.5 @ +113 (miniscule)

—-

Team Totals:

Sandbox team total OVER 10.5 @ -104 (strong)

DRX team total UNDER 13.5 @ -123 (moderate)

DRX League Average LSB
Combined Kills / game 25.155 24.67 26.129
Combined Kills / min 0.776 0.76 0.817
Kills per win 15.638 16.86 17.776
Kills per loss 7.759 8.59 8.545
Deaths per win 7.20 7.81 7.00
Deaths per loss 17.91 16.89 17.05
Average Margin of Victory (AMOV) 7.70 9.04 10.62
Average Margin of Defeat (AMOD) 10.45 9.03 7.95

 

 

Time Total Projection:

Combined AVG Game Time: 33.132 minutes

Odds Weighted / Rating Weighted: 32.98 / 32.35

Combined % of Games OVER Time Total: 49.05% (DRX 52.38% / LSB 45.71%)

Volatility Rating: DRX 0.16694 / LSB 0.16190 (League Avg: 0.15498)

Suggested Model Plays (in order of strength of edge):

none

 

DRX LSB
Average Game Time 33.56 32.71
Avg Game Time (in wins) 34.57 30.12
Avg Game Time (in losses) 32.631 34.239

I’d lean to the under here given that I anticipate a Sandbox win but I’m already playing that through different means so pass.

 

Other Derivatives:

Suggested Model Plays (in order of strength of edge):

OVER 11.5 towers @ -116 (moderate-strong)

LSB first herald @ -143 (moderate-strong)

OVER 4.5 dragons @ -116 (moderate)

OVER 1.5 barons @ +129 (light)

** denotes advantage in a category that the opponent is also strong in, must be considered

I like the LSB first herald quite a bit here.

 

My Picks:

 

Moneyline: Sandbox +104 (1.5 units)

Map Spread: Sandbox -1.5 maps @ +286 (0.75 units)

 

 


LOL European Championship (LEC)

Spring Playoffs – Round 1 – Day 3

 

 

Rogue -400 (-1.5 maps @ -167, -2.5 @ +206, +1.5 @ -1250)

vs

MAD Lions +303 (+1.5 maps @ +129, +2.5 @ -278, -1.5 @ +629)

 

 

Map ML Price: ROG -256 / MAD +196

Total Maps Played: 3.5 maps (over -227 / under +172), 4.5 maps (over +216 / under -294)

Kill Total: 25.5 (over -109 / under -120)

Kill Spread: -7.5 @ -111 / +7.5 @ -116

Team Kill Totals: 16.5 / 8.5

Time Total: 32:00 (over -118 / under -111)

 

 (Individual player model positional ranking is a “living” number comparing individual performance against other players at their position)

ROG Tale of the Tape MAD
928.7 Gold Diff @ 10 min 411.7
1338.6 Gold Diff @ 15 min 1061.0
867.8 Gold Diff @ 20 min 1203.1
78.7 GPM first 20 min vs Avg 85.5
132.8 Gold Diff / min first 20 83.1
434.6 Gold Diff / min Rest of Game 43.7
1899.0 Gold /min (GPM) 1827.5
110.1 Gold / min vs Avg 38.6
232.7 Gold Diff / min 63.0
3.3 Gold Percent Rating (GPR) 1.0
1682.1 Kill Agnostic GPM 1629.9
138.2 Kill Agnostic Gold Diff/min 61.2
1959.4 GPM in wins 1954.2
363.5 Gold Diff per min in wins 377.3
1687.7 GPM in losses 1685.0
-225.0 Gold Diff per min in losses -290.7
232.7 Adjusted Gold Diff / min 62.9
23.4 Win-Adjusted GPM 18.2
60.1 Win-Adjusted Gold Dif/min 73.9
57.1 Dragon Control % 57.7
45.2 Herald Control % 78.1
57.7 Baron Control % 52.0
7.0 Quality Wins? 5.0
87.5 % of wins as Quality 55.6

 

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

 

Suggested Model Plays (in order of strength of edge):

MAD series moneyline @ +303 (VERY strong)

MAD map moneyline @ +196 (strong)

MAD +1.5 maps @ +129 (strong)

OVER 4.5 maps @ +216 (light)

 

 

Quantitative Analysis:

 

My model is more bullish on Rogue than anybody I know (including me) and still thinks this number is out of line and tremendously disrespectful to MAD Lions. I’ll touch more on why I happen to agree in the qualitative section below but let’s first take a look at the numbers which tell a bit of a mixed story.

Rogue are the best team in the West when it comes to kill-agnostic metrics and hold very strong win-adjusted economy metrics as well. MAD have the best objective control (statistically, their film maybe disagrees with this) in the LEC besides G2. Rogue are #1 by a decent margin and MAD #2 in early game economy rating.

This is the two strongest early game teams in the league but one of them is significantly better AFTER the early game while also being good in the early game (Rogue) while the other lacks in late game measures like rest of game gold differential and baron control.

From a statistical perspective it makes sense that the model has these two as close as it does.

Conclusion: Slight edge Rogue overall

Series Outcomes
wins Losses Probability
3 1 22.429%
3 2 20.206%
3 0 16.599%
2 3 16.561%
1 3 15.067%
0 3 9.139%
(Series Win): 59.234%

 

 (Rogue win% outcomes)

 

Qualitative Analysis:

While these two look very similar on paper, their actual play speaks very differently, at least to me. MAD Lions tend to feast when they do win but that’s not to discredit their numbers either since they’re actually one of the stronger losers in the LEC. That’s actually something both of these teams have in common to some extent they just do it in different ways. Both of these teams are very difficult to beat. Rogue make it challenging in a similar fashion to G2, always maintaining their economy despite the chaos of the game and engineering ways to accrue more resources for themselves. MAD are a great punch back team and when they do get a lead have a tendency to completely blow it out of the water with aggressive use of their advantages, sometimes to a fault.

MAD are more consistent than I tend to let on but they’re still a much higher variance team than the seemingly iron-clad Rogue (except against G2 apparently) who are ridiculously consistent. Even with the entire league knowing what they tend to do nobody can stop them with any kind of efficiency. They’re just good.

I think Rogue are the better team here, it’s no question, but this number is simply way too disrespectful toward MAD Lions. Rogue should win this series off the back of their consistency in combination with MAD’s inconsistencies (the bottom lane currently) but I make this maybe -250 at best even after adjustments in favor of Rogue. I’m maybe the most bullish person you know on this Rogue lineup and even I’m taking the MAD Lions here.

Other Markets:

 

Kill Total Model Projections:

Odds-Weighted: 29.873

Time-Implied: 29.073

Underdog Win: 27.094

“Gelati” Total Projection: 29.24 kills

Combined % of Games OVER Kill Total: 51.634% (ROG 44.44% / MAD 58.82%)

Volatility Rating:  ROG 0.30197 / MAD 0.29408 (League Avg: 0.3045)

Suggested Model Plays (in order of strength of edge):

(alt) OVER 26.5 @ +112 (moderate)

—-

Team Totals:

MAD team total OVER 8.5 @ -111 (VERY VERY strong)

 

ROG League Average MAD
Combined Kills / game 27.917 27.73 30.702
Combined Kills / min 0.818 0.85 0.936
Kills per win 20.147 17.84 18.094
Kills per loss 9.000 9.46 11.031
Deaths per win 7.50 8.42 7.44
Deaths per loss 16.75 18.42 21.88
Average Margin of Victory (AMOV) 10.86 9.61 10.56
Average Margin of Defeat (AMOD) 7.75 9.83 11.75

Agree with the model on the MAD Lions team total over here which feels criminally low and like it could get there even in a Rogue win. I have absolutely no idea why this is so low when Rogue have averaged 11 kills a loss and eclipsed this number almost 90% of the time. Rogue tend to give up very few kills per loss but this seems a little nutty to me.

 

Time Total Projection:

Combined AVG Game Time: 33.146 minutes

Odds Weighted / Rating Weighted: 33.53 / 33.64

Combined % of Games OVER Time Total: 48.2% (ROG 61.11% / MAD 35.29%)

Volatility Rating: ROG 0.17926 / MAD 0.11786 (League Avg: 0.14865)

Suggested Model Plays (in order of strength of edge):

none

 

ROG MAD
Average Game Time 34.17 32.12
Avg Game Time (in wins) 33.36 31.18
Avg Game Time (in losses) 37.004 33.185

 

I’d lean toward the unders in this series if anything given how strong the early game is for each of these teams but they’re both also very VERY stubborn to defeat compared to the league averages so perhaps the over is the play. Pass for me.

Other Derivatives:

Suggested Model Plays (in order of strength of edge):

MAD first herald @ +141 (VERY VERY strong)

MAD first blood @ +105 (VERY strong)**

MAD first tower @ +129 (VERY strong) **

ROG first tower @ -172 (VERY strong) **

ROG first dragon @ -179 (strong)**

MAD first dragon @ +134 (strong)**

UNDER 4.5 dragons @ -108 (light)

OVER 12.5 towers @ +102 (light)

UNDER 1.5 barons @ -187 (miniscule)

** denotes advantage in a category that the opponent is also strong in, must be considered

I’ll play MAD first herald but that’s it for these since Rogue are exceptional in each of the other categories just like MAD.

 

 

My Picks:

 

Map Spread: MAD +2.5 maps @ -278 (2.78 units)

Map Spread: MAD +1.5 maps @ +129 (0.5 units)

Moneyline: MAD +303 (0.3 units)

Map Spread: MAD -1.5 maps @ +629 (0.1 units)

Map Spread: MAD -2.5 maps @ +1600 (0.1 units)

Kill Total (team): Map 1 MAD OVER 8.5 @ -111 (1.11 units)

Kill Total (team): Map 2 MAD OVER 8.5 @ -110 (1.1 units)

Kill Total (team): Map 3 MAD OVER 8.5 @ -109 (1.09 units)

Kill Total (team): Map 4 MAD OVER 8.5 @ -127 (1.27 units)

Kill Total (team): Map 5 MAD OVER 9.5 @ -106 (1.06 units)

Prop: Map 1 MAD first herald @ +141 (1 unit)

Prop: Map 2 MAD first herald @ +141 (1 unit)

Prop: Map 3 MAD first herald @ +141 (1 unit)

Prop: Map 4 MAD first herald @ +141 (1 unit)

Prop: Map 5 MAD first herald @ +141 (1 unit)

 


LOL Championship Series

Mid-Season Showdown Playoffs

Round 1 – Day 4

 

 

 

100 Thieves -189 (-1.5 maps @ +108, -2.5 @ +336, +1.5 @ -476)

vs

Dignitas +151 (+1.5 maps @ -139, +2.5 @ -500, -1.5 @ +323, -2.5 @ +822)

 

 

Map ML Price: 100T -164 / DIG +129

Total Maps Played: 3.5 maps (over -303 / under +223), 4.5 maps (over +174 / under -227)

Kill Total: 24.5 (over -105 / under -123)

Kill Spread: -4.5 @ -122 / +4.5 @ -106

Team Kill Totals: 13.5 / 10.5

Time Total: 32:00 (over -125 / under -104)

 (Individual player model positional ranking is a “living” number comparing individual performance against other players at their position)

100T Tale of the Tape DIG
-172.8 Gold Diff @ 10 min -1593.7
-189.9 Gold Diff @ 15 min -409.0
-233.5 Gold Diff @ 20 min -99.4
-9.7 GPM first 20 min vs Avg -20.9
5.5 Gold Diff / min first 20 -60.5
-80.3 Gold Diff / min Rest of Game 123.1
1762.1 Gold /min (GPM) 1786.8
-8.9 Gold / min vs Avg 15.8
-13.2 Gold Diff / min 10.6
-0.2 Gold Percent Rating (GPR) 0.1
1610.1 Kill Agnostic GPM 1597.1
10.6 Kill Agnostic Gold Diff/min -14.9
1895.0 GPM in wins 1914.1
256.7 Gold Diff per min in wins 230.2
1615.8 GPM in losses 1586.7
-310.0 Gold Diff per min in losses -334.5
-4.0 Adjusted Gold Diff / min 19.7
-21.4 Win-Adjusted GPM -2.3
-30.0 Win-Adjusted Gold Dif/min -56.5
53.7 Dragon Control % 52.9
59.5 Herald Control % 57.1
54.5 Baron Control % 52.0
3.0 Quality Wins? 6.0
27.3 % of wins as Quality 54.5

 

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

 

Suggested Model Plays (in order of strength of edge):

UNDER 3.5 maps @ +223 (miniscule)

Dignitas map moneyline @ +129 (miniscule)

 

 

Quantitative Analysis:

 

The model makes this right about on market with the slightest of edges to the Dignitas price (not within qualifying range for a bet).

Series Outcomes
Wins Losses Probability
3 1 23.009%
3 2 20.359%
3 0 17.337%
2 3 16.153%
1 3 14.484%
0 3 8.659%
(Series Win): 60.705%

(100 Thieves win% outcomes)

Taking a look into how these teams square up it’s actually fairly even across the board. Dignitas look slightly better in wins, 100 Thieves slightly better in losses but they actually grade out pretty badly in their wins (7th in gold differential per minute in wins and 8th in gold per minute in wins). Dignitas rule the win-adjusted categories and sit as the composite #4 team in those categories. They’re also better in opponent adjusted economy. 100T have a very slight edge in kill agnostic gold per minute. Both teams grade similarly in objective control.

The biggest thing holding Dignitas back from a statistical standpoint is their rough opening ten minutes which ends up leading to a rough overall early game total. I’ll discuss it more in the qualitative section but I think a lot of these problems can be remedied fairly easily and are more to do with odd decisions just from watching film but more on that below.

If Dignitas’ early game wasn’t as bad as it is from a statistical perspective then this would essentially be an even money series and we’d be looking at essentially the same teams mirroring each other. That’s how I’m treating this series. The underlying metrics (opponent adjusted, win-adjusted, kill agnostic) are mostly in Dignitas’ favor.

Conclusion: Dig’s early game holds their overall rating back quite a bit but otherwise these teams are essentially even.

 

 

Qualitative Analysis:

The regular season series was a 2-0 for 100 Thieves. One was a slugfest, the other a more clinical 100T victory from earlier in the season.

Dignitas have the advantage of extra preparation time for this series. While it didn’t completely pan out for Evil Geniuses yesterday you can see how much more competitive that series was than people thought and the impact that picks like Yone and Neeko provided with time to prepare them. They likely should have pushed that to a game five. Particularly for aggressive drafting teams and teams with rather large pools to pick from, this can be a huge advantage in best-of-five series. I think Dignitas get a ton of help from the extra prep time as they’ve shown they’re more willing to play outside the box concepts without sacrificing in consistency. 100 Thieves have not really strayed from what they’ve been doing all season (other than Sylas) which isn’t always a bad thing but when it comes to a series makes you very predictable and easier to game plan against.

To me, 100 Thieves are such an easy team to read right now and that’s a huge problem. With two weeks to prepare and the film from a series where Cloud 9 dismantled them to go over I think Dignitas can craft a game plan to make this series challenging for them. 100 Thieves have better players but it’s not by much and perhaps more importantly, 100 Thieves haven’t been playing in a way where that has really mattered anyway.

Best-of-fives are all about versatility and adapting in series and I’ll give major props to the Dignitas coaching staff for doing a great job with that sort of thing this season. I don’t think it’s outside the realm of possibility for 100 Thieves to realize some of their shortcomings and switch it up here but they’ve been remarkably stubborn in changing things up strategically with both iterations of this team. 100 Thieves from Lock-In with a strong focus on tempo and dominating the early game to build a lead for themselves would completely decimate this Dignitas team but that was almost three months ago and this isn’t the same looking team. For that reason I actually think Dignitas are very live to take this series and I’d adjust toward them to make this close to a 50/50.

I also think the two weeks down time is more than enough to plan out early game scripts that aren’t a complete disaster for Dignitas and if that’s the case and that’s suddenly not a huge problem for them, then the model would likely have this as a 50/50 as well.

The fact of the matter is that these teams really aren’t that much different. One has very slightly better players, the other is more willing to be aggressive in the draft and I think the aggressive drafting is a better tool to have in a five game series when the gap between the players isn’t that large.

Conclusion: Qualitatively I think this is a coin flip match which means there’s value on the underdog.

 

Combined Conclusion:

After adjustments I’d make this like a 52-48 series for 100 Thieves which makes Dignitas a value so I’m taking Dignitas here. I think they can fix the few problems they do have with some good game planning and they’ve been much more willing to find value in the draft instead of playing it overly conservatively. They’ve also got the extra time to prepare working for them.

Other Markets:

 

Kill Total Model Projections:

Odds-Weighted: 28.14

Time-Implied: 26.33

Underdog Win: 26.54

“Gelati” Total Projection: 27.06 kills

Combined % of Games OVER Kill Total: 49.6% (100T 38.1% / DIG 61.11%)

Volatility Rating:  100T 0.33441 / DIG 0.34929 (League Avg: 0.28653)

Suggested Model Plays (in order of strength of edge):

none

—-

Team Totals:

Dignitas team total OVER 10.5 @ -102 (strong)

 

100T League Average DIG
Combined Kills / game 24.509 26.00 29.094
Combined Kills / min 0.727 0.76 0.814
Kills per win 17.009 16.98 19.423
Kills per loss 7.11 9.37 12.25
Deaths per win 11.64 8.57 10.82
Deaths per loss 16.40 17.26 15.14
Average Margin of Victory (AMOV) 4.82 8.36 8.09
Average Margin of Defeat (AMOD) 8.80 8.68 7.57

Game total OVERS have been following what you’d think in playoffs so far but we’ve only seen a few series. Overs have hit in 7 out of 11 games. The model doesn’t like either side here (weights heavily on frequency) but I think we’re more likely to see something like we saw with TSM vs EG yesterday, a slugfest. The projections here think this goes over, I think we’ll see this go over with both teams throwing the kitchen sink at this, and DIG have tended to drag most games to their level. With that in mind, these are the #1 and #3 most volatile teams in combined kills per game results. They also tend to play longer games (see below). If we anticipate a 33 minute game time on average for these two you’re still at 23.991 in a 100T game and 26.862 in your average DIG game. I think we’re way more likely to get variance to the over than the under in this case. I have a hard time seeing this series go quiet for multiple games. I love this over way more than the model does. DIG team total over isn’t a bad look either.

 

Time Total Projection:

Combined AVG Game Time: 34.18 minutes

Odds Weighted / Rating Weighted: 34.07 / 34.07

Combined % of Games OVER Time Total: 51.19% (100T 52.38% / DIG 50%)

Volatility Rating: 100T 0.21896 / DIG 0.14838 (League Avg: 0.1734)

Suggested Model Plays (in order of strength of edge):

none

 

100T DIG
Average Game Time 35.22 33.14
Avg Game Time (in wins) 36.34 33.84
Avg Game Time (in losses) 33.98 32.04

This number seems low but you have to remember the averages for these two teams is boosted big time by some outlier games this season. Still, the average time for these two with that in mind is still above the average. I’d lean to the over but no bet from me.

 

Other Derivatives:

Suggested Model Plays (in order of strength of edge):

Dignitas first herald @ +114 (VERY strong)**

OVER 4.5 dragons @ -130 (strong)

OVER 1.5 barons @ +147 (strong)

100T first herald @ -156 (moderate)**

100T first tower @ -147 (light)

UNDER 12.5 towers @ -213 (light)

** denotes advantage in a category that the opponent is also strong in, must be considered

If you like the over then you likely think this is going to be a long series and I think these markets are better for attacking that concept. The over dragons is the strongest neutral position on the board not far ahead of the over barons (due to price). I like the over dragons quite a bit in this spot.

 

My Picks:

 

Map Spread: Dignitas +1.5 maps @ -139 (1.39 units)

Moneyline: Dignitas +151 (0.5 units)

Map Spread: Dignitas -1.5 maps @ +323 (0.25 units)

Map Spread: Dignitas -2.5 maps @ +822 (0.1 units)

Kill Total: Map 1 OVER 24.5 @ -105 (1.05 units)

Kill Total: Map 2 OVER 24.5 @ -112 (1.12 units)

Kill Total: Map 3 OVER 24.5 @ -112 (1.12 units)

Kill Total: Map 4 OVER 24.5 @ -112 (1.12 units)

Kill Total: Map 5 OVER 24.5 @ -112 (1.12 units)

Prop: Map 1 OVER 4.5 dragons @ -130 (1.3 units)

Prop: Map 2 OVER 4.5 dragons @ -135 (1.35 units)

Time Total: Map 3 OVER 32:00 @ -125 (1.25 units)

Time Total: Map 4 OVER 32:00 @ -127 (1.27 units)

Time Total: Map 5 OVER 32:00 @ -125 (1.25 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)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Leave a Reply