Genuinely out-of-sample: predictions recorded before matches were played, graded against real outcomes. Generated 2026-07-13 15:49 UTC. Click any matchup row for a feature-level breakdown of why the model favors that side.
Live from Polymarket US
Pulled directly from Polymarket US's public API when you click refresh —
genuinely live, not a stored snapshot. Model probabilities are the model's current read on
each match (refreshed every record_predictions run, matched by event id) —
available even for matches too far out to be in the forward-test betting window yet; matches
with no estimate at all show market price only. Decision support only — nothing here is
placed automatically.
Recorded
10
Scored
3
Pending
7
Model log loss
1.3004
▲ trails market by 0.4695 log loss
Market log loss
0.8309
Model Brier
0.4649
Total staked
$108.29
Total P&L
-$75.55
▼ -69.8% ROI on 2 bets
Snapshot — bets to consider (7)
Not yet resolved, soonest first. Snapshotted once each match fell into the betting window (see `paper_trading.snapshot_window_min_hours`/`snapshot_window_max_hours`) — a proxy for the price at actual bet-placement time, captured the last time record_predictions ran, not a live feed (see "Live from Polymarket US" above for that). Decision support only, nothing here is placed automatically.
Start (ET)
League
Matchup
Model P(A)
Market P(A)
Edge
Stake
Full Kelly %
2026-07-15 05:00
LCK
Gen.G vs Karmine Corp
0.857
0.842
+0.016
$37.31 on Gen.G
7.5%
Feature
Raw diff (A − B)
Pull toward Gen.G
/ Karmine Corp
Log-odds contribution
Recent win rate
+0.40
+0.346
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
n/a (imputed)
+0.000
Rest advantage
+15.97
+0.006
Player Elo rating
+109.69
+0.884
Momentum
-0.07
-0.019
Region strength
+134.61
+0.478
Lane matchups won
+3.00
+0.101
Feature sum (log-odds) → implied P(Gen.G) per game
+1.796 → 85.7%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
Gen.G avg 1755
Chovy
1868
Canyon
1812
Ruler
1765
Kiin
1705
Duro
1628
Karmine Corp avg 1646
Yike
1761
Busio
1674
Caliste
1666
kyeahoo
1564
Canna
1564
2026-07-15 05:00
EWC
Team Secret vs Sentinels
0.469
0.515
-0.046
$37.82 on Sentinels
7.6%
Feature
Raw diff (A − B)
Pull toward Team Secret
/ Sentinels
Log-odds contribution
Recent win rate
+0.00
+0.000
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
n/a (imputed)
+0.000
Rest advantage
+10.42
+0.004
Player Elo rating
+48.82
+0.393
Momentum
+0.67
+0.194
Region strength
-211.30
-0.750
Lane matchups won
+1.00
+0.034
Feature sum (log-odds) → implied P(Team Secret ) per game
-0.125 → 46.9%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
Team Secret avg 1630
Bie
1749
Eddie
1632
Hizto
1594
Dire
1594
Pun
1579
Sentinels avg 1581
Rahel
1615
HamBak
1604
Impact
1597
DARKWINGS
1547
huhi
1540
2026-07-15 06:10
MSI
T1 vs GAM Esports
0.912
0.891
+0.021
$84.87 on T1
17.0%
Feature
Raw diff (A − B)
Pull toward T1
/ GAM Esports
Log-odds contribution
Recent win rate
+0.00
+0.000
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
n/a (imputed)
+0.000
Rest advantage
-31.85
-0.013
Player Elo rating
+146.24
+1.178
Momentum
-0.33
-0.097
Region strength
+311.61
+1.106
Lane matchups won
+5.00
+0.169
Feature sum (log-odds) → implied P(T1) per game
+2.344 → 91.2%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
T1 avg 1726
Doran
1778
Faker
1741
Keria
1715
Peyz
1702
Oner
1693
GAM Esports avg 1580
Kiaya
1729
Taki
1636
Artemis
1530
Aress
1519
Draktharr
1483
2026-07-15 06:10
MSI
Bilibili Gaming vs Movistar KOI
0.755
0.802
-0.047
$26.37 on Movistar KOI
5.3%
Feature
Raw diff (A − B)
Pull toward Bilibili Gaming
/ Movistar KOI
Log-odds contribution
Recent win rate
+0.40
+0.346
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
+0.50
+0.130
Rest advantage
-32.74
-0.013
Player Elo rating
+33.34
+0.269
Momentum
+0.27
+0.078
Region strength
+42.72
+0.152
Lane matchups won
+5.00
+0.169
Feature sum (log-odds) → implied P(Bilibili Gaming) per game
+1.130 → 75.5%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
Bilibili Gaming avg 1646
Viper
1713
Bin
1663
Knight
1639
ON
1621
Xun
1595
Movistar KOI avg 1613
Myrwn
1646
Supa
1618
Alvaro
1617
Jojopyun
1605
Elyoya
1577
2026-07-15 07:20
MSI
G2 Esports vs FURIA Esports
0.858
0.762
+0.096
—
—
Feature
Raw diff (A − B)
Pull toward G2 Esports
/ FURIA Esports
Log-odds contribution
Recent win rate
n/a (imputed)
+0.000
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
n/a (imputed)
+0.000
Rest advantage
n/a (imputed)
+0.000
Player Elo rating
+181.48
+1.462
Momentum
n/a (imputed)
+0.000
Region strength
+96.30
+0.342
Lane matchups won
n/a (imputed)
+0.000
Feature sum (log-odds) → implied P(G2 Esports) per game
+1.804 → 85.8%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
G2 Esports avg 1681
Hans Sama
1730
Caps
1727
BrokenBlade
1682
SkewMond
1661
Labrov
1608
FURIA Esports
No roster on record.
2026-07-15 08:30
MSI
LYON vs JD Gaming
0.554
0.455
+0.098
$84.56 on LYON
16.9%
Feature
Raw diff (A − B)
Pull toward LYON
/ JD Gaming
Log-odds contribution
Recent win rate
+0.10
+0.087
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
-0.50
-0.130
Rest advantage
-111.76
-0.044
Player Elo rating
+25.49
+0.205
Momentum
+0.23
+0.068
Region strength
-1.05
-0.004
Lane matchups won
+1.00
+0.034
Feature sum (log-odds) → implied P(LYON) per game
+0.216 → 55.4%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
LYON avg 1631
Inspired
1738
Dhokla
1663
Saint
1618
Berserker
1573
Isles
1562
JD Gaming avg 1605
JunJia
1726
HongQ
1648
GALA
1637
Xiaoxu
1514
Vampire
1500
2026-07-15 08:30
MSI
Hanwha Life Esports vs MIBR.LOS
0.911
0.901
+0.010
—
—
Feature
Raw diff (A − B)
Pull toward Hanwha Life Esports
/ MIBR.LOS
Log-odds contribution
Recent win rate
n/a (imputed)
+0.000
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
n/a (imputed)
+0.000
Rest advantage
n/a (imputed)
+0.000
Player Elo rating
+187.23
+1.509
Momentum
n/a (imputed)
+0.000
Region strength
+230.91
+0.819
Lane matchups won
n/a (imputed)
+0.000
Feature sum (log-odds) → implied P(Hanwha Life Esports) per game
+2.328 → 91.1%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
Hanwha Life Esports avg 1687
Zeus
1724
Gumayusi
1706
Zeka
1706
Delight
1685
Kanavi
1615
MIBR.LOS
No roster on record.
Model vs. market log loss over time
ModelMarket
By league (min. 3 scored matches)
Model wins (lower log loss)Market wins
Simulated P&L (fractional Kelly, decision support only — no real money)
Scored — graded against real outcomes (3)
Start (ET)
League
Matchup
Model P(A)
Market P(A)
Edge
Stake
Full Kelly %
Result
P&L
2026-07-11 04:00
MSI
Hanwha Life Esports vs LYON
0.821
0.865
-0.044
—
—
Hanwha Life Esports
—
Feature
Raw diff (A − B)
Pull toward Hanwha Life Esports
/ LYON
Log-odds contribution
Recent win rate
+0.00
+0.000
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
n/a (imputed)
+0.000
Rest advantage
+0.96
+0.000
Player Elo rating
+59.25
+0.478
Momentum
-0.67
-0.193
Region strength
+114.96
+0.409
Lane matchups won
+3.00
+0.101
Feature sum (log-odds) → implied P(Hanwha Life Esports) per game
+0.795 → 68.8%
Best-of-5 series adjustment (Polymarket US prices the series, not one game) → P(Hanwha Life Esports wins series)
82.1%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
Hanwha Life Esports avg 1688
Zeus
1724
Zeka
1707
Gumayusi
1707
Delight
1687
Kanavi
1617
LYON avg 1629
Inspired
1736
Dhokla
1661
Saint
1616
Berserker
1572
Isles
1561
2026-07-10 04:00
MSI
LYON vs G2 Esports
0.265
0.248
+0.018
$8.29 on LYON
1.7%
LYON
+24.45
Feature
Raw diff (A − B)
Pull toward LYON
/ G2 Esports
Log-odds contribution
Recent win rate
+0.10
+0.093
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
n/a (imputed)
+0.000
Rest advantage
+0.17
+0.000
Player Elo rating
-63.52
-0.635
Momentum
+0.23
+0.069
Region strength
-18.10
-0.065
Lane matchups won
n/a (imputed)
+0.000
Feature sum (log-odds) → implied P(LYON) per game
-0.538 → 36.9%
Best-of-5 series adjustment (Polymarket US prices the series, not one game) → P(LYON wins series)
26.5%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.
LYON avg 1624
Inspired
1730
Dhokla
1656
Saint
1611
Berserker
1566
Isles
1555
G2 Esports avg 1687
Hans Sama
1736
Caps
1733
BrokenBlade
1687
SkewMond
1667
Labrov
1613
2026-07-09 04:00
MSI
Bilibili Gaming vs Hanwha Life Esports
0.093
0.386
-0.293
$100.00 on Hanwha Life Esports
75.5%
Bilibili Gaming
-100.00
Feature
Raw diff (A − B)
Pull toward Bilibili Gaming
/ Hanwha Life Esports
Log-odds contribution
Recent win rate
-0.10
-0.093
Blue/red side win rate
n/a (imputed)
+0.000
Head-to-head record
+0.50
+0.135
Rest advantage
-0.85
-0.000
Player Elo rating
-54.89
-0.549
Momentum
+0.10
+0.030
Region strength
-189.71
-0.682
Lane matchups won
n/a (imputed)
+0.000
Feature sum (log-odds) → implied P(Bilibili Gaming) per game
-1.159 → 23.9%
Best-of-5 series adjustment (Polymarket US prices the series, not one game) → P(Bilibili Gaming wins series)
9.3%
Side (blue/red) isn't known before a match, so the model's real ~53% historical
blue-side edge is averaged over both teams here rather than credited to whichever one happens to be
listed first — otherwise it would silently read as a home-side advantage for an arbitrary label.