This dataset provides simulated data for predicting the final score of a cricket team in a T20 match based on various match and environmental factors. It is intended for beginner to intermediate machine learning practitioners who want to experiment with regression models in sports analytics. The dataset includes common factors that can influence a cricket match score, such as overs played, wickets lost, run rate, opponent strength, pitch conditions, and weather.

Each row represents a unique T20 match scenario, and the target variable is the predicted total score for the batting team at the end of their innings.

Dataset Structure:

Match ID: Unique identifier for each match. Overs Played: Overs completed (between 1 and 20, as per T20 format). Wickets Lost: Number of wickets lost by the batting team. Run Rate: Runs per over at the current state of the match. Home/Away: Whether the match is played at the team’s home ground or away. Opponent Strength: Rating from 1 to 10, indicating the opponent’s bowling strength. Pitch Condition: Type of pitch – Batting-friendly, Bowling-friendly, or Balanced. Weather: Weather conditions during the match – Sunny, Cloudy, or Overcast. Predicted Score: Target variable. Predicted total score for the team based on match conditions.

Here are a few use cases for this project:

1. Sports Training & Improvement: Coaches or players can use the images from the "Cricket" model to study cricket playing styles, strategies, and techniques. The model can identify cricket equipment, players, and positions helping sportspersons analyze game practices.

2. Sports Journalism & Broadcasting: The model can be used by sports broadcasting networks to automatically analyze and tag certain moments of a cricket match, such as a player's stance, delivery style, or field settings. This can provide real-time insights and stats during live broadcast.

3. E-commerce: Online sports retailers can use this model to create more accurate items' descriptions, tag their cricket product images for easier searchability, and improve user experience.

4. Gaming and Virtual Reality: Computer game developers can use this model to create more realistic and detailed cricket games. The AI model can help model the movements of players, the trajectory of the cricket ball, and other nuances of the sport.

5. Security and Surveillance: In stadiums or sports facilities, the model can be used to monitor crowd behavior during a cricket match assisting security personnel's activities. It can detect any potential unauthorized field intrusions or unwanted activities.

Note: Please consider the data example given, it mentions a blurry image of a group of fish, which doesn't align with the described use cases. It seems like it belongs to a different dataset. Please verify and provide correct data samples.

The data was downloaded from the extensive cricket data website cricsheet.org in JSON format. I used the pandas Python library to transform the match data into ball-by-ball data with several relevant fields. This allows for the data to be used to train regression models etc

This dataset was created as part of a project where I created metrics to rank players for T20 Internationals and the Indian Premier League (IPL). The entire project materials can be found at https://github.com/jamiewelsh25/Cricket_Data_Project/

Notebooks can be found below where I delve into predicting second innings chase success as well as first innings scores. Furthermore, I build a model to evaluate batters, bowlers and all-rounders using a Runs Added Over Average Player metric.

Cricket Dataset V1

## Overview

Cricket Dataset V1 is a dataset for object detection tasks - it contains Game States annotations for 1,920 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Cricket Tracking

## Overview

Cricket Tracking is a dataset for object detection tasks - it contains Person Ball annotations for 1,140 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

This dataset contains data from the first innings of T20 International Cricket Matches. It contains the records of more than 400 T20I matches that include each and every ball bowled in the first innings.

There are a total of 17 parameters that have been used:

1) Batting_team: Batting Team

2) Bowling_team: Bowling Team

3)City: Venue

4)Current_score: Runs Scored

5)pp: Powerplay

6)Balls_left: Deliveries left to bowl

7)Wickets_left: Wickets left

8)Crr: Current Run Rate

9)Top_Order: Whether the top order batsmen are batting or not

10)Middle Order: Whether the middle order batsmen are batting or not

11)Lower Order: Whether the lower order batsmen are batting or not

12)Tail: Whether the tail-enders i.e. the bowlers have come to bat or not

13)Pressure: If the batting team is under pressure by scoring at a run rate of 7 or below or not

14)Last_five: Runs scored in the last 5 overs

15)Aggression_Mode: Whether the batting team is batting aggressively by scoring 45 or more runs in the last 5 overs

16)Death_Overs: Overs after over no. 15 are death overs

17)Runs_x: Final Score made by the batting team by the end of the innings

Context

Cricket is a worldwide attraction sport. In some of the regions of the Commonwealth, like India or Australia, it is very close to a National Sport.

Content

Tweets about Cricket, one of the most loved team sports in the world.

Acknowledgements

The content is from Twitter; tweets were collected using Twitter API and tweepy.

Inspiration

Few suggestions: - Perform data cleaning on the multitude of spam tweets using #Cricket hashtag to ride on the popularity of this sport; - Perform sentiment analysis on tweets from cricket fans across the World;

samhitmantrala/cricket dataset hosted on Hugging Face and contributed by the HF Datasets community

Dataset Card for "llama-cricket-dataset"

More Information needed

Flat Cricket

## Overview

Flat Cricket is a dataset for object detection tasks - it contains Ball annotations for 663 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Context

I was working on a college assignment that made me process cricket matches data. After ending that assignment I came up with the idea to integrate these matches' data with the players' data that have played in each of these matches. When I found a website that hosted this data I tried and web scraped all the data possible including those cricketers who have even not played in the matches of which I had the data.

Content

I collected the data by web scraping the website Cricket Country The website provided extensive data about many cricketers. I processed the data to remove noise and the final dataset has full name along with place, country, date of birth of the players, the teams for which they have played and the batting/bowling style.

Acknowledgements

I owe thanks to the website owners for hosting such extensive data.

bhuvaneshprasad/odi-cricket-dataset-1971-2014 dataset hosted on Hugging Face and contributed by the HF Datasets community

1) Data Introduction • The Cricket Player Debuts & Last Matches (4 Nations) Dataset is a comprehensive collection of debut times and final matches for cricketers from four countries in India, Australia, Pakistan, and Afghanistan, structuring career data by format, including Test/ODI/T20I/IPL.

2) Data Utilization (1) Cricket Player Debuts & Last Matches (4 Nations) Dataset has characteristics that: • This dataset, along with player name, team, and player ID, provides the debut opponent, date, location, and last match opponent, date and location for each format (Test/ODI/T20I/IPL) in columns. • It was collected in a programming manner through the Cricbuzz API and focuses on player information from four countries considering data completion. (2) Cricket Player Debuts & Last Matches (4 Nations) Dataset can be used to: • Player career pattern analysis: A career management strategy can be derived by comparing the age of debut by format, duration of activity, and frequency of format transitions (e.g., Test→T20) by country. • Cricket Distribution Network Dashboard: Visualization of career history for each player can be used for scouting, fan service, media content planning, and more.

Context

This data is the denormalized ball-by-ball data from cricsheet for cricket matches from 2004-2021

https://cricsheet.org/downloads/all_male_json.zip

Acknowledgements

This dataset is based on https://cricsheet.org/

To quote their about page

WHO IS BEHIND THIS? That would be Stephen Rushe. He has written all of the code which extracts and validates the data, as well as developing the website. He also finds it strange to be writing about himself in the third-person.

WHAT IS CRICSHEET?

Cricsheet is a collection of projects which collectively provide data for various aspects of cricket. The current projects provide ball-by-ball match data for Men’s and Women’s Test Matches, One-day internationals, Twenty20 Internationals, some other international T20s, various club competitions such as the Indian Premier League, Women's T20 Challenge, Big Bash League, T20 Blast, Pakistan Super League, The Hundred, Charlotte Edwards Cup, and Rachael Heyhoe Flint Trophy matches; as well as County Championship and Sheffield Shield matches, and a registry of people linking the identifiers used for them on various sites.

CricketShotClassification Dataset

  Dataset Description

This dataset is designed for video classification of cricket shots. It contains labeled videos of ten different cricket shots, making it suitable for training and evaluating machine learning models for cricket action recognition.

  Dataset Structure

The dataset contains videos of ten cricket shots:

Shot Name Label Class ID

Cover Drive cover 0

Defense Shot defense 1

Flick Shot flick 2

Hook Shot… See the full description on the dataset page: https://huggingface.co/datasets/rokmr/cricket-shot.

Analysis of ‘Cricket data’ provided by Analyst-2 (analyst-2.ai), based on source dataset retrieved from https://www.kaggle.com/mahendran1/icc-cricket on 30 September 2021.

--- Dataset description provided by original source is as follows ---

Context

Any aspiring datascientist will look everything in view of data. Even when chilling with friends, watching cricket live and cheering for the favorite team.

Content

It includes ODI, Test, t20 statistics of all the players in all the three category (batting ,bowling and fielding).

Acknowledgements

We wouldn't be here without the help of cricket. Thank you for all the great cricketers for the wonderful contribution.

--- Original source retains full ownership of the source dataset ---

This dataset contains information about all national cricket teams. It includes data on the players, matches played, win/loss ratio, coach, and captain for each team. The data was collected from various sources and compiled to create a comprehensive dataset for analysis and research purposes

The dataset contains information on 18 national cricket teams. It includes 8 variables, including the team name, ICC ranking, captain, coach, number of players, matches played, win/loss ratio, and the format of the match (Test, ODI, or T20). The data is in CSV format and has 18 rows and 8 columns. There are no missing or incomplete values in the dataset

nitv/cricket-data dataset hosted on Hugging Face and contributed by the HF Datasets community

Dataset

This dataset was created by Sachin Adnaik

Released under Other (specified in description)

Contents