Thanks for showing interest in taking part in our Co-learning activity. We assure you the journey full of learning by working on live sports analytics project. To get into the Data Science field, the first and the last thing you need is the portfolio where you have solved a problem using data science. Ask any good Data Scientist and they will tell you the same.
In this project, we will create statistical solution for predicting fantasy team for each and every cricket match. This project is live project being tested on IPL 2021.
Any solution needs to be tested well before deployment. Here we are trying to create a solution to predict fantasy team for cricket matches irrespective of tournament. This would help users to minimize their loss probability. IPL is one of the most loved competitive form of cricket tournament. It has a mix of players from various nations coming to play together, this makes it even more competitive in every sense. IPL is an annual series and thus ideally suited for testing of this live project. A perfect use case is already in hand, we just need to use it wisely :)
- Need to be passionate about cricket as domain knowledge plays a critical role in success of any project.
- Working proficiency in tools like EXcel, Python, etc.
- Participant should be passionate about data and should know tricks to make data speak efficiently.
Goal of this project is to achieve an accuracy of more than 8 players prediction in Fanstasy team for IPL 2021 matches.
How to achieve higher prediction accuracy? :
- Analyze the data and decide on segregation factors.
- Approach to accomodate all rounder in the process.
- Consideration for venue and other external factors.
- Analysis of any other important series which can add value for new players and uncapped players assesment.
- Prepare statistical approach for player selection considering all above factors.
- Add head to head player statistics to above match to add intutive insight in terms of data.
- Test the above model on previous matches played by creating team as per dreams 11 point system.
- Try out Monte Carlo simulator or any other technique to improve accuracy.
- Test it on ongoing IPL 2021 matches by creaing your own dream team based on model.
- Tweak the model as per requirement to achieve as high accuracy as possible (benchmark accuracy is 8 players).
- Preparing a process document and business report to be submitted along with code at the completion of project.
For enrollment into live project a one on one discussion session would be scheduled with our team where our team will understand your aspiration for joining this project and your understanding about field of Data Science and Cricket.
Mentor driven project
Mentor would have alternate day check-in to keep track of the progress and also discuss approach taken by the team. Since, this is time bound project thus a team of 4 each would be created who all will have a designated task and need to co-ordinate for task completion.
Team task
Approach would be discussed along with mentor and same needs to be implemented in stipulated timeframe in co-ordination with other team members. Time commitment is the key to success for this project completion.
The whole idea is to mentor and guide you well in this learning journey. Like others, we don't want to give you False hopes and then fail at making a successful career transition. Also, once you join this live project and after finishing you can showcase this in your profile and apply with sports analytics company as well scale the same approach for other sports as well.
Right data is essence but selection of right data happens only with proper analysis. Datasets ahve been collated by our team to make this process easier for all the participants.Datasets collected are keeping in mind below points:
- All IPL season data matchwise.
- Ball by Ball delivery data for all IPL matches.
- Venue details for all matches along with scorecard.
- Bowling style deatils for bowlers and all rounders.
- Other 20-20 format series details.
- Recent and career statistics performance of all players.
To delve into Data Science, the first step is to identify and analyze our dataset in order to ascertain different patterns, detect missing values and anomalies and identify different characteristics of our data. In other words, we attempt to understand what the data is trying to tell us.
Analyze the data based on different aspects :
- Categorise players based on their strength.
- Categorise venues based on pitch and boundary capability.
- Come up with new features based on game format like hard hiiting capability, Short per Index, etc.
- Categorize players based on Dreams 11 point system.
- Analyze fitness level of player based on recent and career statistics.
- Analyze player to player combination.
- Analyze toss winning aspect for team and player as well to asses the performance.
- Segreate the players who behave like an outlier or for whom performance pattern can't be detected.
In the field of sports analytics variables are highly correlated. We know that for any model to perform well correlation needs to be handled.
Approaches to handle correlation:
- AHP technique to assign weights to variables.
- Assigning weights to recent and career statitics columns seperately.
- Using Dreams 11 poin system to calculate scores.
- Using Monte Carlo simulator to generate scores for each player.
- Using ELO scores for players.
Sports Analytics is more about statistical approach than Machine Learning Approach alone. So based on analysis model needs to be created to predict fantasy team players for each match of the series. Without testing no model is complete, so a thorough round of testing needs to be performed.
- Test the model performance on 2nd half of IPL 2020 matches.
- Test the model on 1st half of IPL 2021 matches.
- Tweak the model (if required) and test on 2nd half of IPL 2021 matches.
A developed model needs to be evaluated for it’s performance before being actually deployed in real time environment.
- Create fantasy team as per model for each and every match.
- Record the accuracy of model by checking how many predicted players made to Dream Team.
- Goal is to achieve an accuracy of more than 8 players correct prediction in fantasy team.
Concluding a project and presenting to stakeholders is an importanat step for closure.
- Detailed approach needs to be documented.
- All dream team and predicted team should be maintained.
- Final business presentation needs to be created mentioning project outcome.
- All code needs to be alligned and complied in a folder properly.
- You’ll get to understand the entire Cricket Analytics pipeline.
- You’ll get to learn about additional techniques such as Monte Carlo Simulation & ELO Score Calculation.
- You’ll learn methods to handle Correlation and methods to assign weights to variables accordingly.
- You’ll be getting hands on live project experience for cricket analystics.
- You’ll be able to intract with mentors and last year IPL project team.
- You’ll get to test your model on real time data of IPL 2021.
- You’ll be building a strong LinkedIn profile visible to sports analytics companies.
- We will teach you to create the content and increase LinkedIn presence to build your self-brand during this project(If you are interested).
- This project should give enough confidence to handle projects in Sports Analytics Domain.
- Get your Resume and GitHub reviewed by experts. [Additional service at cheaper price]
- Once you have undertaken and completed the project, you will get a project completion certificate and full-fledged support from our mentors from the community for any technical help, guidance, etc.
- As most of the company prefer giving assignment to the candidate for ML role, we can help you to mentor for same.
- As we have mentored and we know your skills and achievement, we will refer you for any AI/ML job in sports analytics companies which fits your profile.
- As we know LinkedIn is the platform to catch recruiters' attention, we will shout out your achievements, help to boost your work on LinkedIn to get visibility.
- You will be eligible for future sports analytics projects.
- Since the project time-lines are fixed, you will have to submit the assigned task within a given time.
- Please join the project based on your available bandwidth. Project learning fees is non-refundable.
- Project meeting will happen every alternate day.
- Time commitment required would be atleast 3 hours per day without fail till project wrap up (approx 3 months).
If you feel you are qualified and are up for the learning ride, please send your profile (Resume, LinkedIn, GitHub) to colearninglounge@gmail.com with subject line "IPL 2021 - Live Project"
While applying, do let us know:-
- Why do you want to be a part of this project.
- What are your expectations from this project?
- Course starts on 25th March 2021 till 10th June 2021.
- To maintain the quality of project maximum 12 person in batch is allowed forming 3 teams of 4 each.
- Last day to apply is 22nd March 2021.
- For any query email us to colearninglounge@gmail.com.
During the span of the project, if you help us in creating content (learning material) for the project, then based on your contribution we will consider you for further projects as mentor/assistant mentor.