22 152
Prognosen checked after the result.
Period: last 120 days · Glicko-2 and math models (no live signals)AI scoreboard
Intelligent AI scoreboard for sports analytics
The AI scoreboard shows automated Spiel analytics, Ergebnis probabilities, live signals and post-result model checks.
First we look at strong Ergebnisse and the Markt's Übereinstimmung with the model; exact Ergebnis remains a reference Markt.
Öffentlich metric
PropickAI-Prognose accuracy
Available to everyone. Informational. The model is calibrated to the Markt; this is Prognose accuracy, not betting profitability.
59.15%
Overall hit-rate on the main Ergebnis.
13 102 hits out of 22 152
Model Sicherheit
Average Prognose
Actual (hit-rate)
Prognosen
50–60%
53.0%
58.5%
11 613
60–70%
64.6%
65.3%
3 966
70–80%
74.5%
71.2%
2 080
80%+
84.3%
88.3%
1 193
Calibration: the higher the model Sicherheit on the left, the higher the real hit-rate on the right. This means the model's probabilities Spiel reality.
Performance by sport
Hit-rate is the share of Ergebnisse correctly predicted by the model. ROI is the historical return on the model's Value-Tipps (flat 1 unit, over the period). Informational, backtest, not a profit guarantee.
Sport
Prognosen
Hit-rate
Value Tipps ROI
Tennis
4 626
65.6%
n/a
Football
3 868
46.1%
n/a
Basketball
1 083
66.6%
n/a
Hockey
339
55.8%
n/a
Volleyball
251
67.7%
n/a
A hit-rate above the Markt does not guarantee positive ROI: at low Quoten even high accuracy can go negative (e.g. tennis). Profit holds where the model agrees with the Markt (basketball).
Method
Wie unsere KI funktioniert
PropickAI compares sports statistics, Quoten, Glicko-2, live signals and the results of completed Spiele. The AI scoreboard shows which models give a stable signal, where the Ergebnis Wahrscheinlichkeit is confirmed by the Markt and how Prognosen pass the post-game check.