Interview PrepCareerGenAIPrompt EnggChatGPTLLMLangchainRAGAI AgentsMachine LearningDeep LearningGenAI ToolsLLMOpsPythonNLPSQLAIML Projects

Reading list

Machine Learning Basics for a Newbie
6 Steps of Machine learning Lifecycle Introduction to Predictive Modeling
Introduction to Exploratory Data Analysis & Data Insights Descriptive Statistics Inferential Statistics How to Understand Population Distributions?
Reading Data Files into Python Different Variable Datatypes
Probability for Data Science Basic Concepts of Probability Axioms of Probability Conditional Probability
Central Tendencies for Continuous Variables Spread of Data KDE plots for Continuous Variable Overview of Distribution for Continuous variables Normal Distribution Skewed Distribution Skeweness and Kurtosis Distribution for Continuous Variable
Central Tendencies for Categorical Variables Understanding Discrete Distributions Performing EDA on Categorical Variables
Dealing with Missing Values Understanding Outliers Identifying Outliers in Data Outlier Detection in Python Outliers Detection Using IQR, Z-score, LOF and DBSCAN
Sample and Population Central Limit Theorem Confidence Interval and Margin of Error
Bivariate Analysis Introduction
Covariance Pearson Correlation Spearman's Correlation & Kendall's Tau Correlation versus Causation Tabular and Graphical methods for Bivariate Analysis Performing Bivariate Analysis on Continuous-Continuous Variables
Tabular and Graphical methods for Continuous-Categorical Variables Introduction to Hypothesis Testing P-value Two sample Z-test T-test T-test vs Z-test Performing Bivariate Analysis on Continuous-Catagorical variables
Chi-Squares Test Bivariate Analysis on Categorical Categorical Variables
Multivariate Analysis A Comprehensive Guide to Data Exploration The Data Science behind IPL
Supervised Learning vs Unsupervised Learning Reinforcement Learning Generative and Descriminative Models Parametric and Non Parametric model
Machine Learning Pipeline Preparing Dataset Build a Benchmark Model: Regression Build a Benchmark Model: Classification
Evaluation Metrics for Machine Learning Everyone should know Confusion Matrix Accuracy Precision and Recall AUC-ROC Log Loss R2 and Adjusted R2
Dealing with Missing Values Replacing Missing Values Imputing Missing Values in Data Working with Categorical Variables Working with Outliers Preprocessing Data for Model Building
Understanding Cost Function Understanding Gradient Descent Math Behind Gradient Descent Assumptions of Linear Regression Implement Linear Regression from Scratch Train Linear Regression in Python Implementing Linear Regression in R Diagnosing Residual Plots in Linear Regression Models Generalized Linear Models Introduction to Logistic Regression Odds Ratio Implementing Logistic Regression from Scratch Introduction to Scikit-learn in Python Train Logistic Regression in python Multiclass using Logistic Regression How to use Multinomial and Ordinal Logistic Regression in R ? Challenges with Linear Regression Introduction to Regularisation Implementing Regularisation Ridge Regression Lasso Regression
Introduction to K Nearest Neighbours Determining the Right Value of K in KNN Implement KNN from Scratch Implement KNN in Python
Bias Variance Tradeoff Introduction to Overfitting and Underfitting Visualizing Overfitting and Underfitting Selecting the Right Model What is Validation? Hold-Out Validation Understanding K Fold Cross Validation
Introduction to Feature Selection Feature Selection Algorithms Missing Value Ratio Low Variance Filter High Correlation Filter Backward Feature Elimination Forward Feature Selection Implement Feature Selection in Python Implement Feature Selection in R
Introduction to Decision Tree Purity in Decision Tree Terminologies Related to Decision Tree How to Select Best Split Point in Decision Tree? Chi-Squares Information Gain Reduction in Variance Optimizing Performance of Decision Tree Train Decision Tree using Scikit Learn Pruning of Decision Trees
Introduction to Feature Engineering Feature Transformation Feature Scaling Feature Engineering Frequency Encoding Automated Feature Engineering: Feature Tools
Introduction to Naive Bayes Conditional Probability and Bayes Theorem Introduction to Bayesian Adjustment Rating: The Incredible Concept Behind Online Ratings! Working of Naive Bayes Math behind Naive Bayes Types of Naive Bayes Implementation of Naive Bayes
Understanding how to solve Multiclass and Multilabled Classification Problem Evaluation Metrics: Multi Class Classification
Introduction to Ensemble Techniques Basic Ensemble Techniques Implementing Basic Ensemble Techniques Finding Optimal Weights of Ensemble Learner using Neural Network Why Ensemble Models Work well?
Introduction to Stacking Implementing Stacking Variants of Stacking Implementing Variants of Stacking Introduction to Blending Bootstrap Sampling Introduction to Random Sampling Hyper-parameters of Random Forest Implementing Random Forest Out-of-Bag (OOB) Score in the Random Forest IPL Team Win Prediction Project Using Machine Learning Introduction to Boosting Gradient Boosting Algorithm Math behind GBM Implementing GBM in python Regularized Greedy Forests Extreme Gradient Boosting Implementing XGBM in python Tuning Hyperparameters of XGBoost in Python Implement XGBM in R/H2O Adaptive Boosting Implementing Adaptive Boosing LightGBM Implementing LightGBM in Python Catboost Implementing Catboost in Python
Different Hyperparameter Tuning methods Implementing Different Hyperparameter Tuning methods GridsearchCV RandomizedsearchCV Bayesian Optimization for Hyperparameter Tuning Hyperopt
Understanding SVM Algorithm SVM Kernels In-depth Intuition and Practical Implementation SVM Kernel Tricks Kernels and Hyperparameters in SVM Implementing SVM from Scratch in Python and R
Introduction to Principal Component Analysis Steps to Perform Principal Compound Analysis Computation of Covariance Matrix Finding Eigenvectors and Eigenvalues Implementing PCA in python Visualizing PCA A Brief Introduction to Linear Discriminant Analysis Introduction to Factor Analysis
Introduction to Clustering Applications of Clustering Evaluation Metrics for Clustering Understanding K-Means Implementation of K-Means in Python Implementation of K-Means in R Choosing Right Value for K Profiling Market Segments using K-Means Clustering Hierarchical Clustering Implementation of Hierarchial Clustering DBSCAN Defining Similarity between clusters Build Better and Accurate Clusters with Gaussian Mixture Models
Understand Basics of Recommendation Engine with Case Study
8 Ways to Improve Accuracy of Machine Learning Models
Introduction to Dask Working with CuML
Introduction to Machine Learning Interpretability Framework and Interpretable Models model Agnostic Methods for Interpretability Implementing Interpretable Model Understanding SHAP Out-of-Core ML Introduction to Interpretable Machine Learning Models Model Agnostic Methods for Interpretability Game Theory & Shapley Values
Introduction to Machine Learning Interpretability Framework and Interpretable Models model Agnostic Methods for Interpretability Implementing Interpretable Model Understanding SHAP Out-of-Core ML Introduction to Interpretable Machine Learning Models Model Agnostic Methods for Interpretability Game Theory & Shapley Values
Introduction to AutoML Implementation of MLBox Introduction to PyCaret TPOT Auto-Sklearn EvalML
Pickle and Joblib Introduction to Model Deployment
Deploying Machine Learning Model using Streamlit Deploying ML Models in Docker Deploy Using Streamlit Deploy on Heroku Deploy Using Netlify Introduction to Amazon Sagemaker Setting up Amazon SageMaker Using SageMaker Endpoint to Generate Inference Deploy on Microsoft Azure Cloud Introduction to Flask for Model Deploying ML model using Flask
Model Deployment in Android Model Deployment in Iphone

Importance of actionable insights in analytics (with case from ICC Cricket World Cup)

Kunal Jain
Kunal Jain Last Updated : 13 Apr, 2015
5 min read

Have you ever been to a meeting, where everyone in the room has good stats to share, but no one knows how to use them? I am sure, you have! Don’t worry, you are not alone and by end of this article, you will have some tips to make your analysis better.

In this article, I will challenge how analytics is being showcased during the cricket matches. While the display is a good way to evangelize use of analytics on grand scale, it is missing the very point it should be proving!

The message I want to leave through this article is far broader than just application to cricket. Example of cricket is being used only as a case study.

P.S. If you don’t follow cricket, you can still follow the article for the points I am trying to tell. Here is a beginner’s guide to Cricket. 

 

Example of insights shown during a cricket match:

Here is an example of what spectators glued to the screen would typically see at the start of the match:

  • Keys to success for Team A:
    • Team A wins 85% of matches when Player X scores more than 80 runs
    • If there are 2 partnerships of more than 50 runs, team A wins in 72% matches
    • Team A wins 70% matches, if bowler Y picks up 4 wickets

There will be similar set of so called “insights” for the second team as well. The analysts performing this analysis have also attached probabilities with each event!

Here is a live example, which was shown in a recent match (South Africa vs. New Zealand):

key_to_success

These look good and exciting! What is the problem with this? Can you spot problems with these so called keys to success?

 

The problem with current insights:

While these insights are good to see, they do not help the teams play better. Let me explain – an insight saying that team A wins 85% of matches when Player X scores more than 80 runs is useless to the team coach and the captain.

It is an insight, but not an actionable insight!

The team management would want analytics to do much more than just pulling out this insight and then praying that Player X has a good outing in every match!

The current insights provide no help to the team management to utilize this information.

If I was the analyst running this model – I would go further and say what is the best strategy to make sure Player X goes on to make a big score – Which position should he bat at? What kind of areas and which bowlers should he target? Which bowlers should he negotiate?

Let’s take another example – analysts have mapped out the strong zone and the weak zone for each batsman. You would think this is clearly actionable. The bowler just needs to ball in the right areas.

But it isn’t! Why? Strong zone and weak zone for a batsman would change from bowler to bowler, with different field settings and different whether conditions. It would also depend on the current form of both – the bowler and the batsman.

Other way to look at these insights is this – they have a lot of numbers and stats, but don’t really tell what they mean.

 

Ways to improve insights (in Cricket matches):

As shown already, the current level of analysis shown during cricket matches is rudimentary at best! There are tons of ways to improve this analysis. I’ll share a few high level thoughts, which, when implemented would surely provide better use of analytics:

  • Right level of granularity: Some of the insights shown in these matches are very high level in nature to be useful. The analyst who created the key to success in above case study has not done the analysis at the right level of granularity. For example, if you look at the keys to success for South Africa – the first point mentioned is that they should score more than 280 runs! Let’s look at the granularity here – all matches played outside Asia since 2011? Even a lay man would say that the pitches prepared for World Cup have favored batsman and 300 has been a par score, which is usually not the case. The analyst did not look at the right level of granularity to take into account the right nature of pitch.
  • Focus on creating actionable insights from user perspective: Like any other piece of analysis, the focus of the analysis has to be the end user. If the analysis is for team management, then we need to bring out insights to improve performance of the players. If it is the audience watching on the TV, then it has to answer the questions which would be going on in mind of the audience. For example, there can be widgets showing the probabilities of win for each team.
  • Additional features / variables should be considered: Sadly, in almost all these insights, there are a lot of features which are missing. For example: current form of the players, the way in which a batsman paces his innings are missing from the current analysis. Needless to say, the more feature we add, the more insightful and helpful the analysis would be.
  • What if scenarios and alternate strategies: Finally, in order to make this analysis more useful and fun, there could be displays of what-if scenarios. So, what do the analysts predict, if bowler X comes up to bowl vs. if bowler Y comes to bowl. Similarly for batting as well – how does the scenario likely to emerge if the captain decides to send player X instead of Y on fall of next wicket!

There are many more ways to improve, but this should hopefully help you to understand what I mean by actionable insights.

 

End Notes:

The idea behind this article was to bring out some ways in which you can improve your analysis and to show case them through a real life case study. I have learnt some of these practices the hard way over time, but you don’t need to do that! A single minded focus on yielding actionable insights for your users can completely change the way analytics can add value to them. On the other hand, a sketchy job can lead to wrong outcomes and mis-guided views. All the best for the next piece of analysis you do.

 

Disclaimer: I have shared some of the gaps on the analysis showcased to the audience during recent cricket matches. I am sure individual teams in the tournament would be taking help of analysts for more sophisticated analysis. I do not have access to that analysis and hence do not know, how many of the shortcomings mentioned here get covered through those pieces of analytics.

 

If you like what you just read & want to continue your analytics learning, subscribe to our emails, follow us on twitter or like our facebook page.

actionable insightscricketsports analystsSports analytics
Kunal Jain
Kunal Jain

Kunal Jain is the Founder and CEO of Analytics Vidhya, one of the world's leading communities of AI professionals.

With over 17 years of experience in the field, Kunal has been instrumental in shaping the global AI landscape. His expertise spans diverse markets, from developed economies like the UK to emerging ones like India, where he has successfully led and delivered complex data-driven solutions. As a recognized thought leader, Kunal has empowered countless individuals to realize their AI ambitions through his visionary approach to AI education and community building.

Before founding Analytics Vidhya, Kunal earned both his undergraduate and postgraduate degrees from IIT Bombay and held key roles at Capital One and Aviva Life Insurance across multiple geographies. His passion lies at the intersection of analytics, AI, and fostering a thriving community of data science professionals.

Free Courses

Generative AI 4.7

Generative AI - A Way of Life

Explore Generative AI for beginners: create text and images, use top AI tools, learn practical skills, and ethics.
Large Language Models 4.5

Getting Started with Large Language Models

Master Large Language Models (LLMs) with this course, offering clear guidance in NLP and model training made simple.
Prompt Engineering 4.6

Building LLM Applications using Prompt Engineering

This free course guides you on building LLM apps, mastering prompt engineering, and developing chatbots with enterprise data.
LlamaIndex 4.8

Improving Real World RAG Systems: Key Challenges & Practical Solutions

Explore practical solutions, advanced retrieval strategies, and agentic RAG systems to improve context, relevance, and accuracy in AI-driven applications.
RAG systems using LlamaIndex 4.7

Microsoft Excel: Formulas & Functions

Learn Retrieval-Augmented Generation (RAG): learn how it works, the RAG framework, and use LlamaIndex for advanced systems.

Recommended Articles

Responses From Readers

Clear

Sriram
Sriram

Your disclaimer sums it up very well. While the audience analysis needs no grain et all, I am sure there are many analysts and analytic tool helping support staff to help the playing eleven!

123

Kunal Jain
Kunal Jain

Sriram, I have a different view there. The audience analysis in current form is mis-leading. If you think there is no need to show the analysis, the media should totally skip it. However, if they do show the analysis, than it better be good. For example, a score of 280+, giving South Africa a 85% chance of success is wrong insight. I would have thought the number to be 320+, and that is what it would come to, if you normalize it for scores in World Cup. Regards, Kunal

123
Philip
Philip

Dear Kunal I agree with your sentiment about the statistics shared with the audience. It is interesting how these figures are both non-informative and skewed. A while ago, I had a conversation with a data analyst who was involved in sports analytics. He indicated to me that they do indeed evaluate batting/bowling partnerships and a selection of appropriately granular statistics. Moreover, they all purchase data from [optasports], which collects every imaginable detail. This is where the delivery pitch maps etc. come from. I believe that their "Keys to success" statements, albeit useless, is a valuable addition for audiences, but it may also be likely that these are geared for easy viewer access, in that it is simple and easy to understand. Since they do however have access to more granular data, it is unforgivable for them not to be doing a better job ad sharing these insights with the public. I do not have access to the appropriate level of information, nor could I at fist glance get an idea about the pricing associated but I did manage some statistics that spoke more accurately about the probabilities of the outcome of that particular match. For example, batting averages, number of hundreds, number of fifties, number of 30s etc., over the last 18 month for the likely teams (in terms of player line-up) that were to take the field. Also the average per innings score of each team, batting second, batting first. I did not however group by location (pitches), but you could normalise over all batsmen to get and indication of individual form. I also looked at bowler averages, strike rate, total wickets (experience) and so forth. South Africa scored 30 100s, 43 50s to NZ's 16 and 34 respectively prior to yesterday's game for example, which tells a the tell of our batsmen heavy line-up, and similarly our bowling economy over the last 18 months, taking into account the 12 or so likely players, was over 5 where, NZ's was under 5. Our bowling strike rate and average however was better. Considering how the match played out, with the match reduced to 43 overs, the statistic (Economy) that proved fatal relates to our inability to defend a total because of poor(ish) bowling economy. Given the data available, is was obvious that NZ's in form bowling line-up would be pinned against our in from batting line-up. This coupled with their hot/cold batting line-up and our leaky bowling attack saw the game play out almost exactly like that, in particular because of the rain, a black swan event. My stats seemed to suggest that SA might have the upper hand assuming a 50 over game, and I think it's likely, given the circumstances at 37 overs into the first innings. SA should have been wearing the favourites tag going into the change of innings, had the rain stayed away. Regards, Philip

123

prahlad
prahlad

hello kunal, Excellent article. Opta is a sports analytics company and icc have outsourced it to them . I wonder how they went wrong in cricket but, they do provide good stats in football.

123

Sriram Ranga
Sriram Ranga

While I agree with some of the improvement suggestions, I think this is probably the first time anything of this sort is analyzed and displayed to the viewers, however rudimentary it is. Yeah, it is no great shakes at this point and its probably in its Alpha version, which I am sure will evolve into a more meaningful add-on to the matches in the near future. Let us give the benefit of doubt to the team who is behind this and hope that more richer analysis & insights come out for an avid viewer to make some sense out of it. Lastly, I hate over-engineering and over-analyzing an art/skill as we might end-up enjoying the art/skill as much as we do today:)

123

Write for us

Write, captivate, and earn accolades and rewards for your work
  • Reach a Global Audience
  • Get Expert Feedback
  • Build Your Brand & Audience
  • Cash In on Your Knowledge
  • Join a Thriving Community
  • Level Up Your Data Science Game
imag

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details