← Back to Today's Picks

Model Track Record

Transparent performance data across 3606 NBA regular-season games (3 seasons)

64.5%

Overall Accuracy

2325/3606

75.4%

High Confidence

1193/1583 games (>65%)

57.4%

Medium Confidence

600/1045 games (55-65%)

68.0%

Last 50 Games

34/50

Monthly Trend

47.8%

10/22

64.4%

11/22

57.7%

12/22

59.9%

01/23

60.1%

02/23

64.6%

03/23

60.0%

04/23

51.9%

10/23

60.3%

11/23

68.1%

12/23

68.4%

01/24

63.8%

02/24

67.0%

03/24

64.3%

04/24

63.4%

10/24

67.4%

11/24

68.3%

12/24

62.2%

01/25

69.9%

02/25

70.2%

03/25

67.6%

04/25

≥65% 55-64% <55% 50% = coin flip

Last 20 Predictions

Date	Matchup	Score	Pick	Conf	Result
04-13	PHX @ SAC	98-109	PHX	50.7%	✗
04-13	LAL @ POR	81-109	POR	51.8%	✓
04-13	LAC @ GSW	124-119	LAC	60.8%	✓
04-13	TOR @ SAS	118-125	TOR	66.0%	✗
04-13	OKC @ NOP	115-100	OKC	97.1%	✓
04-13	UTA @ MIN	105-116	MIN	91.3%	✓
04-13	DAL @ MEM	97-132	MEM	69.6%	✓
04-13	DEN @ HOU	126-111	HOU	62.5%	✗
04-13	DET @ MIL	133-140	MIL	55.7%	✓
04-13	CHI @ PHI	122-102	CHI	87.1%	✓
04-13	WAS @ MIA	119-118	MIA	80.0%	✗
04-13	IND @ CLE	126-118	CLE	51.0%	✗
04-13	NYK @ BKN	113-105	NYK	71.1%	✓
04-13	CHA @ BOS	86-93	BOS	91.3%	✓
04-13	ORL @ ATL	105-117	ATL	51.8%	✓
04-11	HOU @ LAL	109-140	HOU	70.4%	✗
04-11	LAC @ SAC	101-100	LAC	80.9%	✓
04-11	GSW @ POR	103-86	GSW	58.4%	✓
04-11	SAS @ PHX	98-117	PHX	51.8%	✓
04-11	OKC @ UTA	145-111	OKC	97.1%	✓

Methodology

Our model uses XGBoost trained on 3 seasons of NBA data. The primary feature is the net rating differential between home and visitor teams, computed as a rolling 20-game average of point differentials.

Predictions are evaluated on a straight-up winner basis (did the model pick the winning team?). Only regular-season games where both teams had at least 5 games of history are included.

Confidence tiers: High (>65%), Medium (55-65%), Low (<55%).

Get Daily AI-Powered Picks

Join 500+ bettors receiving our model's top plays every morning.