Prediction of NBA talent and quality from non-professional tracking data

Patent metadata

Field	Value
Publication number	US-12307767-B2
Application number	US-202318336400-A
Country	US
Kind code	B2
Filing date	Jun 16, 2023
Priority date	Oct 1, 2020
Publication date	May 20, 2025
Grant date	May 20, 2025

Abstract

A computing system identifies broadcast video for a plurality of games in a first league. The broadcast video includes a plurality of video frames. The computing system generates tracking data for each game from the broadcast video of a corresponding game. The computing system enriches the tracking data. The enriching includes merging play-by-play data for the game with the tracking data of the corresponding game. The computing system generates padded tracking data based on the tracking data. The computing system projects player performance in a second league for each player based on the tracking data and the padded tracking data.

First claim

The invention claimed is:

1. A method, comprising:
receiving, by a computing system, broadcast video data for a plurality of games from a tracking system, wherein the broadcast video data includes one or more players in a first league;
generating, by the computing system, tracking data for each of the plurality of games, the tracking data comprising coordinates of one or more player locations in the plurality of games;
receiving, by the computing system, play-by-play data for each of the plurality of games from a play-by-play feed corresponding to the broadcast video data, the play-by-play data describing one or more events that occur within the plurality of games;
merging, by the computing system, the play-by-play data for the plurality of games with the tracking data of the plurality of games to generate an enriched data set; and
projecting, by the computing system, player performance in a second league for the one or more players based on the enriched data set, wherein the projecting includes classifying each of the one or more players.

2. The method of claim 1 , further comprising:
identifying, by the computing system, events in the tracking data using a neural network.

3. The method of claim 2 , further comprising:
supplementing, by the computing system, the events with contextual information derived from the tracking data.

4. The method of claim 1 , wherein projecting, by the computing system, the player performance in the second league for the first player based on the enriched data set comprises:
projecting a draft position for the first player.

5. The method of claim 4 , wherein projecting the draft position for the first player comprises:
classifying the first player into a bin of a plurality of bins, each bin representing a range of draft positions.

6. The method of claim 1 , wherein merging, by the computing system, the play-by-play data for the plurality of games with the tracking data of the plurality of games to generate the enriched data set comprises:
combining the play-by-play data with optical character recognition data, the coordinates of the player locations, and ball position data using a fuzzy matching algorithm.

7. The method of claim 1 , further comprising:
reducing, by the computing system, random noise in the enriched data set by creating new player representations using mean-regression.

8. A non-transitory computer readable medium comprising one or more sequences of instructions, which, when executed by a processor, causes a computing system to perform operations comprising:
receiving, by a computing system, broadcast video data for a plurality of games from a tracking system, wherein the broadcast video data includes one or more players in a first league;
generating, by the computing system, tracking data for each of the plurality of games, the tracking data comprising coordinates of one or more player locations in the plurality of games;
receiving, by the computing system, play-by-play data for each of the plurality of games from a play-by-play feed corresponding to the broadcast video data, the play-by-play data describing one or more events that occur within the plurality of games;
merging, by the computing system, the play-by-play data for the plurality of games with the tracking data of the plurality of games to generate an enriched data set; and
projecting, by the computing system, player performance in a second league for the one or more players based on the enriched data set, wherein the projecting includes classifying each of the one or more players.

9. The non-transitory computer readable medium of claim 8 , further comprising:
identifying, by the computing system, events in the tracking data using a neural network.

10. The non-transitory computer readable medium of claim 9 , further comprising:
supplementing, by the computing system, the events with contextual information derived from the tracking data.

11. The non-transitory computer readable medium of claim 8 , wherein projecting, by the computing system, the player performance in the second league for the first player based on the enriched data set comprises:
projecting a draft position for the first player.

12. The non-transitory computer readable medium of claim 11 , wherein projecting the draft position for the first player comprises:
classifying the first player into a bin of a plurality of bins, each bin representing a range of draft positions.

13. The non-transitory computer readable medium of claim 8 , wherein merging, by the computing system, the play-by-play data for the plurality of games with the tracking data of the plurality of games to generate the enriched data set comprises:
combining the play-by-play data with optical character recognition data, the coordinates of the player locations, and ball position data using a fuzzy matching algorithm.

14. The non-transitory computer readable medium of claim 8 , further comprising:
reducing, by the computing system, random noise in the enriched data set by creating new player representations using mean-regression.

15. A system comprising:
a processor; and
a memory having programming instructions stored thereon, which, when executed by the processor, causes the system to perform operations comprising:
receiving broadcast video data for a plurality of games from a tracking system, wherein the broadcast video data includes one or more players in a first league;
generating tracking data for each of the plurality of games, the tracking data comprising coordinates of one or more player locations in the plurality of games;
receiving play-by-play data for each of the plurality of games from a play-by-play feed corresponding to the broadcast video data, the play-by-play data describing one or more events that occur within the plurality of games;
merging the play-by-play data for the plurality of games with the tracking data of the plurality of games to generate an enriched data set; and
projecting player performance in a second league for the one or more players based on the enriched data set, wherein the projecting includes classifying each of the one or more players.

16. The system of claim 15 , wherein the operations further comprise:
identifying events in the tracking data using a neural network.

17. The system of claim 16 , wherein the operations further comprise:
supplementing the events with contextual information derived from the tracking data.

18. The system of claim 15 , wherein projecting the player performance in the second league for the first player based on the enriched data set comprises:
projecting a draft position for the first player.

19. The system of claim 18 , wherein projecting the draft position for the first player comprises:
classifying the first player into a bin of a plurality of bins, each bin representing a range of draft positions.

20. The system of claim 15 , wherein merging the play-by-play data for the plurality of games with the tracking data of the plurality of games to generate the enriched data set comprises:
combining the play-by-play data with optical character recognition data, the coordinates of the player locations, and ball position data using a fuzzy matching algorithm.

Assignees

• Stats Llc

Inventors

• Patton Andrew
• Walker Nathan
• Scott Matthew
• Ottenwess Alex

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06V20/44

  Event detection · CPC title

• G06T7/292

  Multi-camera tracking · CPC title

• G06N5/042

  Backward inferencing · CPC title