Model Accuracy & Track Record

Every prediction frozen at registration time. Every result shown — including where the market wins.

Pre-match model

Team-signal predictions, frozen at registration time — published uncalibrated and resolved automatically

League	Month	N	Acc %	Brier(lower = better)	Mkt Brier
CBLOL	Oct 2026	7	71.4%	0.1866	0.0413✓
Demacia Cup Global Invitational	Oct 2026	26	34.6%	0.2623	0.1820✓
EMEA Masters	Oct 2026	33	63.6%	0.2240	0.1702✓
LCS	Oct 2026	9	66.7%	0.2399	—
Worlds	Oct 2026	22	40.9%	0.2729✓	0.2930
CBLOL	Sep 2026	9	55.6%	0.2207	0.2202✓
EMEA Masters	Sep 2026	98	50.0%	0.2748	0.2005✓
LCK	Sep 2026	8	62.5%	0.2356	0.1347✓
LCS	Sep 2026	12	50.0%	0.2535	0.0726✓
LEC	Sep 2026	8	37.5%	0.2411	0.2099✓
LPL	Sep 2026	13	53.9%	0.2468	0.1999✓
North American Challengers League	Sep 2026	6	66.7%	0.2338	0.0916✓
World Star Challengers Invitational	Sep 2026	51	54.9%	0.2383	0.1786✓
CBLOL	Aug 2026	24	79.2%	0.1953	0.1008✓
LCK	Aug 2026	39	35.9%	0.2611✓	0.2909
LCP	Aug 2026	17	70.6%	0.2307	0.1583✓
LCS	Aug 2026	20	80.0%	0.1816	0.1556✓
LEC	Aug 2026	37	59.5%	0.2249	0.1740✓
LES	Aug 2026	13	92.3%	0.1412	0.0929✓
LFL	Aug 2026	25	72.0%	0.2183	—
LPL	Aug 2026	51	68.6%	0.2202✓	0.2242
North American Challengers League	Aug 2026	22	59.1%	0.2396	0.1563✓
Prime League 1st Division	Aug 2026	16	56.3%	0.2299	0.2225✓
LCK	Jul 2026	6	50.0%	0.2416✓	0.3731
LCP	Jul 2026	8	62.5%	0.2091✓	0.2501
LEC	Jul 2026	8	87.5%	0.1656	0.1378✓
LES	Jul 2026	16	81.3%	0.1709	0.0749✓
LFL	Jul 2026	30	50.0%	0.2680	0.2270✓
LPL	Jul 2026	20	75.0%	0.2108	0.1699✓
MSI	Jul 2026	14	57.1%	0.2423	0.1828✓
North American Challengers League	Jul 2026	5	80.0%	0.2309✓	0.2775
Prime League 1st Division	Jul 2026	49	61.2%	0.2294✓	0.2346
EMEA Masters	Jun 2026	13	53.9%	0.2508	0.1806✓
MSI	Jun 2026	5	40.0%	0.2266	0.0030✓
Prime League 1st Division	Jun 2026	5	40.0%	0.3119✓	0.3192
Overall totals		768	58.7%	0.2352	0.1934

At-draft AI agent (Hermes)

Predicts at draft lock-in, compared against Polymarket odds captured at the same moment. Rows where the market wins are shown — that transparency is the credibility.

League	N	Acc %	Agent Brier(lower = better)	Mkt Brier
EM	246	69.1%	0.1919	0.1865✓
LCK	219	62.6%	0.2287	0.2266✓
LEC	165	63.0%	0.2278	0.2180✓
North Amer	146	62.3%	0.2363	0.2216✓
LCS	145	55.2%	0.2485✓	0.2493
Prime Leag	144	63.2%	0.2260	0.2160✓
CBLOL	140	64.3%	0.2265	0.2215✓
LCK CL	139	56.8%	0.2405	0.2368✓
LPL	137	64.2%	0.2312✓	0.2366
LCP	112	67.9%	0.2103	0.2054✓
LES	110	66.4%	0.2060	0.1896✓
LFL	106	67.0%	0.2149	0.1921✓
TCL	83	69.9%	0.2102	0.2048✓
LJL	76	61.8%	0.2363	0.2094✓
DCGI	71	66.2%	0.2080✓	0.2107
WSCI	69	63.8%	0.2043	0.1920✓
MSI	65	66.2%	0.2052	0.1978✓
VCS	59	54.2%	0.2524✓	0.2671
Worlds	38	57.9%	0.2217✓	0.2298
LPLOL	12	75.0%	0.1202✓	0.1294
EWC	9	44.4%	0.3067	0.2723✓
KeSPA Cup	7	57.1%	0.1991	0.1594✓
All leagues	2301	63.5%	0.2222	0.2152

✓ ★ agent Brier lower✓ ★ market Brier lower0.25 = coin flip

Methodology

The pre-match model uses team-level signals only: Elo ratings, season win rate, H2H record, and recent form — no champion data, frozen at prediction time. Hermes (the at-draft agent) adds seven signals including champion win rates and duo synergies. Brier score measures probabilistic accuracy: 0 = perfect, 1 = maximally wrong, 0.25 = equivalent to a 50/50 coin flip. Lower is better. Predictions are pre-registered — they cannot be edited after creation.

About the model·Our prediction game's append-only public ledger