This page is a live audit trail — each row is a timestamped prediction paired with the real outcome. Explore the record by league, filter for hits or misses, and judge calibration for yourself.
PREDICTIONS SCORED
387,257
Completed matches with known outcomes
PEAK ACCURACY (HIGH)
78.5%
When the model is most confident · TT-Elite
EARLIEST RECORD
Feb 19, 2026, 7:29 AM
Start of tracked history
LAST UPDATED
Nov 8, 2026, 11:11 AM
Most recent match processed
METHODOLOGY · EVERY SPORT · STATED VS. REALIZED
How this number is measured
"HIGH confidence" means the model assigned a probability above 80% to the outcome it predicted. "Stated" is the average of those predicted probabilities; "Realized" is how often that prediction actually won. A well-calibrated model keeps the two close. Predictions are scored only after the match is final and the outcome is known — no prediction is graded against a result it could have seen in advance. Each sport reports the accuracy of a single named model revision (never a blend of retired and current models), and a draw counts as a miss against a team-win call in every three-way market. Figures below cover every prediction in the record, not a filtered subset.
SPORT
STATED
REALIZED
GAP
HIGH-TIER SAMPLE
Table Tennis
82.9%
77.3%
−5.6pp
2,075
Tennis
87.9%
74.8%
−13.1pp
2,007
Soccer
83.9%
83.8%
−0.1pp
594
NFL
87.4%
88.9%
+1.5pp
36
NBA
85.8%
92.9%
+7.1pp
42
GAP = STATED − REALIZED · A LARGE GAP MEANS OVERCONFIDENCE, NOT AN EDGE
FILTER LEAGUE
CONFIDENCE TIERS · TT-Elite · BEST-PERFORMING LEAGUE
Calibration check
HIGH ≥ 80% · MED 60–80% · LOW 50–60%
A well-calibrated model wins more often when it is most certain. High-confidence forecasts should consistently outperform medium and low tiers.
HIGH CONFIDENCE
78.5%
1,126 of 1,435 correct
MEDIUM CONFIDENCE
66.3%
16,294 of 24,588 correct
LOW CONFIDENCE
54.6%
17,437 of 31,965 correct
High = model >80% sure · Medium = 60–80% · Low = 50–60%
ACCURACY OVER TIME · 7-DAY ROLLING · BY CONFIDENCE TIER
Rolling accuracy
HIGH ≥ 80% · MED 60–80% · LOW 50–60%
7-day rolling accuracy by confidence tier. A widening gap between High and Low confirms the model is well-calibrated — it wins more when it is most certain.
ROLLING ACCURACY · BY MONTH · FORWARD-CHAINED
Accuracy by month
OVERALL AVG · 61.6%
56%FEB 26
58%APR 26
59%MAY 26
60%JUL 26
62%AUG 26
63%SEP 26
62%OCT 26
64%NOV 26
MONTHLY OVERALL ACCURACY · BARS IN GREEN ≥ 60% · LAST 12 MONTHS
ACCURACY · BY LEAGUE AND MODEL
League breakdown
Each row shows how accurately each model predicted match winners for that league. The top-scoring model per league is highlighted — different models lead in different competitions.
LEAGUE
MODEL ACCURACY
CONFIDENCE TIERS (BEACON)
MATCHES
ANCHOR SIGNAL
EDGE SIGNAL
MOMENTUM SIGNAL
BEACON CONSENSUS
HIGH
MED
LOW
ATP Tour
61.0%
66.1%
65.9%
64.0%
79.6%
13,506 picks
65.4%
47,262 picks
53.8%
27,075 picks
87,843
Setka Cup
56.2%
56.6%
56.5%
56.4%
89.2%
333 picks
62.9%
14,886 picks
53.8%
42,959 picks
58,178
TT-Elite
59.8%
60.0%
59.2%
60.1%
78.5%
1,435 picks
66.3%
24,588 picks
54.6%
31,965 picks
57,988
WTA Tour
64.0%
67.1%
67.4%
66.5%
81.4%
12,043 picks
68.0%
24,408 picks
56.2%
20,699 picks
57,150
TT-Cup
56.1%
54.3%
55.0%
55.9%
60.0%
325 picks
61.4%
4,818 picks
54.9%
27,652 picks
32,795
Czech-Liga-Pro
56.7%
57.2%
57.0%
57.3%
—
65.0%
5,350 picks
54.8%
16,020 picks
21,370
EFL League One
41.5%
41.7%
41.8%
41.7%
44.9%
98 picks
47.3%
1,659 picks
39.9%
5,207 picks
6,964
EFL Championship
44.1%
44.3%
44.4%
44.3%
46.4%
112 picks
51.7%
1,421 picks
42.3%
5,419 picks
6,952
EFL League Two
41.5%
41.7%
41.6%
41.6%
48.2%
193 picks
42.3%
1,611 picks
41.1%
4,970 picks
6,774
La Liga
47.7%
47.7%
47.9%
47.8%
75.9%
299 picks
56.0%
1,315 picks
42.2%
3,383 picks
4,997
Premier League
46.7%
47.0%
47.0%
47.0%
60.5%
243 picks
52.7%
1,538 picks
43.2%
3,203 picks
4,984
La Liga 2
42.6%
42.6%
42.5%
42.4%
50.0%
8 picks
53.8%
725 picks
40.1%
3,543 picks
4,276
Ligue 1
44.7%
45.0%
45.0%
44.8%
59.2%
267 picks
50.5%
1,387 picks
40.3%
2,604 picks
4,258
Serie A
48.2%
48.2%
48.2%
48.2%
69.9%
276 picks
56.2%
1,327 picks
41.6%
2,520 picks
4,123
2. Bundesliga
45.1%
45.3%
45.3%
45.3%
65.2%
23 picks
53.3%
687 picks
43.4%
3,200 picks
3,910
Eredivisie
49.8%
49.9%
49.9%
49.8%
79.1%
344 picks
57.7%
1,304 picks
40.6%
2,200 picks
3,848
1. Bundesliga
46.7%
47.0%
46.9%
46.9%
69.0%
281 picks
54.7%
1,145 picks
40.7%
2,422 picks
3,848
Ligue 2
43.3%
43.5%
43.6%
43.6%
63.0%
73 picks
48.9%
705 picks
41.7%
2,860 picks
3,638
Turkish Super Lig
54.1%
54.3%
54.1%
54.2%
80.4%
317 picks
63.9%
980 picks
46.5%
2,317 picks
3,614
Primeira Liga
52.7%
52.9%
52.9%
52.8%
80.9%
477 picks
58.3%
835 picks
44.8%
2,236 picks
3,548
Belgian Pro League
48.3%
48.7%
48.5%
48.7%
78.4%
134 picks
60.7%
807 picks
42.7%
2,271 picks
3,212
Serie B
42.9%
43.0%
43.0%
42.8%
54.2%
72 picks
46.6%
556 picks
41.7%
2,583 picks
3,211
Greek Super League
57.4%
57.7%
57.8%
57.6%
83.7%
215 picks
69.8%
777 picks
47.7%
1,510 picks
2,502
Scottish Premiership
50.8%
51.1%
50.9%
51.1%
83.0%
165 picks
64.1%
665 picks
41.9%
1,526 picks
2,356
Scottish Championship
46.7%
46.7%
46.7%
46.8%
53.8%
13 picks
60.4%
396 picks
43.5%
1,695 picks
2,104
Scottish League Two
41.4%
41.7%
41.7%
41.5%
59.3%
27 picks
45.5%
336 picks
40.0%
1,199 picks
1,562
Scottish League One
43.6%
44.2%
44.0%
43.9%
43.8%
64 picks
46.7%
364 picks
43.0%
1,129 picks
1,557
USA MLS
47.3%
47.3%
47.3%
47.3%
—
57.6%
125 picks
46.4%
1,432 picks
1,557
National League
47.4%
47.7%
47.6%
47.6%
—
70.7%
123 picks
45.6%
1,418 picks
1,541
Argentina Liga Profesional
43.1%
43.2%
43.2%
43.2%
—
67.1%
82 picks
41.7%
1,287 picks
1,369
Brazil Serie A
50.5%
50.8%
50.9%
50.7%
100.0%
1 picks
68.9%
135 picks
48.1%
962 picks
1,098
Mexico Liga MX
52.4%
52.1%
52.0%
52.2%
100.0%
1 picks
70.5%
166 picks
48.5%
813 picks
980
Romania Superliga
48.6%
48.6%
48.6%
48.5%
66.7%
3 picks
65.4%
130 picks
45.6%
768 picks
901
Japan J1 League
46.3%
45.7%
46.3%
46.3%
—
60.2%
103 picks
44.5%
787 picks
890
Poland Ekstraklasa
46.6%
46.2%
46.4%
46.5%
—
60.2%
88 picks
44.9%
777 picks
865
China Super League
53.8%
53.5%
53.8%
53.7%
100.0%
11 picks
68.6%
156 picks
48.4%
545 picks
712
Sweden Allsvenskan
48.3%
47.6%
47.6%
47.4%
50.0%
4 picks
62.6%
123 picks
44.1%
571 picks
698
Norway Eliteserien
53.4%
54.0%
54.0%
53.8%
87.5%
8 picks
68.9%
119 picks
50.1%
555 picks
682
Switzerland Super League
47.1%
46.6%
46.5%
46.5%
0.0%
1 picks
68.2%
85 picks
43.3%
564 picks
650
Russia Premier League
53.7%
53.9%
53.9%
53.9%
100.0%
7 picks
74.2%
132 picks
47.3%
457 picks
596
Denmark Superliga
50.6%
50.3%
50.6%
50.3%
100.0%
1 picks
60.9%
87 picks
48.2%
465 picks
553
Austria Bundesliga
47.9%
47.5%
47.7%
48.1%
75.0%
8 picks
58.7%
104 picks
45.0%
431 picks
543
Ireland Premier Division
43.1%
43.1%
43.1%
42.9%
—
58.3%
60 picks
41.0%
481 picks
541
Finland Veikkausliiga
51.7%
51.7%
51.5%
51.7%
75.0%
4 picks
66.3%
95 picks
48.2%
423 picks
522
Champions League
58.7%
59.2%
58.2%
59.2%
—
75.0%
40 picks
55.5%
173 picks
213
Argentina Copa De La Liga Profesional
41.9%
41.4%
41.4%
41.4%
—
53.8%
13 picks
40.5%
190 picks
203
NFL
67.7%
0.0%
0.0%
67.7%
88.9%
36 picks
68.7%
99 picks
54.0%
63 picks
198
NBA
67.9%
0.0%
0.0%
67.9%
92.9%
42 picks
67.0%
94 picks
51.7%
60 picks
196
UEFA Euro
49.7%
50.3%
49.7%
50.3%
—
53.3%
30 picks
49.6%
131 picks
161
NWSL
50.5%
50.5%
50.5%
50.5%
—
72.7%
11 picks
47.9%
96 picks
107
African Cup of Nations
43.3%
43.3%
44.3%
44.3%
—
77.8%
9 picks
40.9%
88 picks
97
FIFA World Cup
67.5%
68.7%
68.7%
67.5%
—
87.5%
16 picks
62.7%
67 picks
83
Copa America
59.4%
56.3%
59.4%
59.4%
—
66.7%
12 picks
55.0%
20 picks
32
FA Women's Super League
63.6%
54.5%
63.6%
63.6%
66.7%
3 picks
33.3%
3 picks
80.0%
5 picks
11
Frauen Bundesliga
83.3%
83.3%
83.3%
83.3%
100.0%
2 picks
100.0%
1 picks
66.7%
3 picks
6
UEFA Europa League
66.7%
66.7%
66.7%
66.7%
—
—
66.7%
3 picks
3
Copa del Rey
66.7%
66.7%
66.7%
66.7%
—
100.0%
1 picks
50.0%
2 picks
3
Liga Profesional
50.0%
50.0%
50.0%
50.0%
—
—
50.0%
2 picks
2
FIFA U20 World Cup
100.0%
100.0%
100.0%
100.0%
—
100.0%
1 picks
—
1
North American League
100.0%
100.0%
100.0%
100.0%
—
—
100.0%
1 picks
1
RAW LOG · EVERY PREDICTION RECORD
Prediction audit trail
Every row is one match. "Fav %" is the probability assigned to the predicted winner. "Source" is the model version that produced the forecast.