With IPL, It’s an Exciting Inning for Augmented Analytics in Cricket – a MECBot Exclusive

With over one billion fans across the world, Cricket is easily one of the most popular competitive sports globally. Not surprisingly, the Indian sub-continent makes up nearly 90% of the global fan base of Cricket. The passionate, thrilling and nail-biting entertainment offered by the T20 format of the sport was catapulted to a new level over a decade ago, when the Indian Premier League (IPL) emerged in the global Cricket landscape. Touted as the brain-child of the then BCCI Vice President, Mr. Lalit Modi, the IPL opened to rave reviews, quickly gaining ground with the masses.

By incorporating city-based teams, the IPL made Cricket as the new face of India’s regional diversity (e.g. KKR for Kolkata, CSK for Chennai, and so on). It democratized Cricket talent through open auctions on national & international players. Further, the power-packed celebrity endorsement for franchises (like Shah Rukh Khan rooting for KKR or Preity Zinta taking controlling stake in Kings XI Punjab) sealed the deal for IPL’s successful run over the years.

But, did you know?

When you take a peek behind the scenes, it is impossible to miss the role played by data analytics as one of the key enablers of the incredible viewership experience offered by IPL (or Cricket in general). Back when IPL first entered the stage, the audience was still crunching basic statistics like runs scored, wickets taken, strike rate, run rate and so on. Flash forward to 2019, we have a whole new lingo spoken effortlessly by the new generation of Cricket lovers, like:

  • What’s the probability of Royal Challengers Bangalore (RCB) winning against KKR in the Eden Gardens?
  • How does MS Dhoni’s 360o Batting Profile Analysis look like this year?
  • How is Late-Order Hitting working out for Sunrisers Hyderabad (SRH)?

Apart from the audience, the key stakeholders higher up in the Cricket data value chain like coaches, sponsors, players and media have also taken the deep dive into analytics. This has been reflected in smarter team selections, data-driven performance improvement of athletes, informed choices for batting order decisions and optimizing corporate sponsorships.

Let us consider the following interesting example of data-driven team selection from IPL 2018:

“A replacement statistic ascertains a player’s value based on whether he/she helps the team win. In essence, it is the measurement of just how much better is a player than the lowest level of players that will be paid by the team.

For example, Chennai Super Kings (CSK) has Tamil Nadu wicketkeeper-batsman N. Jagadeesan in its squad this year. He was acquired for Rs 20 lakh as a backup for skipper M.S. Dhoni. While the Chennai captain is 75 times (Rs. 15 crores) more valuable than Jagadeesan, the latter is expected to provide a level of production, if Dhoni were to get injured, at the league minimum salary.”

In reality, analyzing the massive and disparate pool of Cricket data and gleaning insights in near real-time is no easy task. Let’s take a quick glimpse at the most critical challenges faced by Cricket Analytics today.

Top 3 Challenges in Cricket Analytics:

  1. Cricket Data is Complex, Multivariate and Uneven, Coupled With High Volume & Velocity: To put things into perspective here, we are talking about “the ball-by-ball information of 5,31,253 Cricket players in close to 5,40,290 cricket matches at 11,960 cricket grounds across the world – and this is just the tip of the iceberg. Cricket data gets generated at a blinding speed, accumulates rapidly, and makes it near impossible to unify uneven data types into a single form.
  2. Number of Sources of Cricket Data Are Rapidly Growing in Quantity & Variety: As if diversity in the data itself was not enough, Cricket data today emanates from an ever-expanding plethora of sources – thanks to digital explosion. Apart from on-the-pitch data, there’s also a vast pool of off-the-pitch data like player fitness indicators, expert opinions, fan reactions and reviews, performance data from wearable technology, search history of cricket fans, lifetime value of players, look-alike analysis of players, weather predictions, informed predictions made by Cricket experts, and much more. Much of this data is unstructured, sitting in silos that don’t talk to each other.
  3. Insights Extracted from Cricket Data Need to be in Near Real-Time and in Human Language: Analyzing this gigantic data pool and making the insight platter available in a format that people can understand and interact with in near real-time is the third challenge. Let’s say that you are watching a live match and Virat Kohli breaks Chris Gayle’s current record of scoring the maximum number of centuries in IPL. This insight would be dead beat if not cued to you in real-time. Now, let’s say that after getting this update, you now want to know how many centuries the ‘Universe Boss’ has actually scored in IPL so far. The catch? Your analytics platform only understands ‘Chris Gayle’ – not ‘Universe Boss.’ Thus, if analytics cannot speak the language of Cricket, its usability is reduced drastically.

This is where Augmented Analytics comes into the picture. But first, let us understand what Augmented Analytics is and why it is important.

Augmented Analytics: An Overview

Augmented Analytics encompasses ingestion of high volume of data from internal and external data sources, cleaning and massaging the ingested data for noise-reduction and pre-processing, and seamless unification of heterogeneous data sources into an extensible,  semantically enriched data fabric. Graph data model plays a pivotal role in capturing and preserving the inherent relationships among entities, which makes it the most comprehensive approach for creation of smart data fabric. This smart data fabric is then modeled to turbo-charge automated pattern detection, smart data discovery, free form queries with NLP and easy deployment of purpose-built AI algorithms without any need for users to code manually. It is continuously updated with new data, insights and data relationships such that it can be compressed and decompressed to any degree of granularity.

Augmented Analytics in Cricket: Introducing MECBot

In Cricket, Augmented Analytics goes beyond questions like ‘How many runs did you score?’ or ‘How many wickets did you take?’, and asks questions that truly matter, like: ‘How valuable are you to your team?’

With IPL at hand, this question has become of paramount importance. With MECBot, the #1 Augmented Data Management Platform, it is just a matter of a few minutes to extract meaningful player and team insights in order to identify the best strategy for the team in terms of player selection as well as other crucial decisions. For example, MVPI, or the Most Valuable Player Index, can be used to identify the best players performing in a tournament for a particular team. Hence, when analyzing batsmen, the metrics for MVPI analysis could include parameters like Hard Hitting Ability, Finishing Ability, Fast Scoring Ability, Consistency and Running Between Wickets. Learn more about MVPI here.

Use-Case: How MECBot Transformed User Experience for a Leading Sports Analytics Company Using Augmented Analytics

For an acclaimed Fortune 1000 sports analytics company, the major challenge was to engage and retain their audience – such as Cricket fans, team managers, sponsors, media houses and sports editors – for a longer duration on their portal. The company was sitting on the enormous Cricket data it had compiled during the last century, but was unable to operationalize it in an engaging format in order to hold users’ attention span and give them an interactive, dynamic and meaningful experience. MECBot was deployed to develop and design a comprehensive Cricket analytics platform that provides dynamic insights, real-time updates, natural language query and intelligent analytics using Cricket terminology, fueled by cutting edge AI at its core.

Natural language query & narration E.g. “Fastest 100s in ODI” or “Kohli against Aus in Aus in ODI”
360-degree player analysis Current performance compared against seasonal averages, weak positions, strong positions, current opposition in current pitch, etc.
Look-alike analysis A player is analysed against look-alike (similar players) to derive data-driven strategy for a match
Specific, intelligent comparison “Warne in bouncy pitches” or “Lara against Lee in bouncy pitches”
Near real-time coverage Alerts sent before, during and after  the match
Team performance analysis Prediction of team performance in the current match based on historical data, current environmental data like bouncy pitch, moisture conditions, etc.
Multiple user definitions Different views of data, e.g. Managers & fans see different dashboards each

 

Here’s how MECBot boosted the Key Performance Indicators (KPIs) of the sports analytics company:

How MECBot Works

MECBot automates the entire data unification process envisaging all forms of data at scale, and delivers unprecedented business results. To accomplish this, MECBot first structures the unstructured data contextually using domain specific business ontologies and marries it with structured transactional data in near real time. This creates a comprehensive Data Graph for an enterprise (Smart Enterprise Graph or Data Fabric).

This Smart Enterprise Graph is accessible through “Free Form Vertical Search” and APIs. Plus, it is always hydrated through schedules and can be transformed into any shape required by the downward analytics or data application layers. MECBot takes user inputs on the data relationships and his preferences to extract logical datasets that unravel these connections. These logical datasets or the context lattices can be accessed through MECBot APIs. MECBot is now available to the user to slice and dice the data. You can say goodbye to multi-op queries and SQL Queries with our free-form search, explore data with descriptive analytics tools, access the free-form query feature in English language and visualize the data with MECBot’s own visualization tools or using third-party tools like Tableau, Qlik etc.

Conclusion

By adopting Augmented Data analytics in Cricket, you can navigate fluidly across disparate data sources and infrastructure types. With a single, consolidated framework to manage, enhance and connect all data powered by mobility across multiple isolated decision centers, you can leverage infrastructure solutions that align with your business needs without worrying about compatibility, integrity or security. IPL is here to say – and so is Augmented Analytics. It is time to take the leap forward towards differentiated user experience in sports with MECBot.

Excited? Want to know more? Just drop a line at contactus@formcept.com to know about our sports analytics solutions in depth. Don’t forget to like and share this post with other Cricket fans and sports analytics enthusiasts in your network!

 

References:

  1. https://economictimes.indiatimes.com/news/sports/india-constitutes-90-percent-of-one-billion-cricket-fans-icc-research/articleshow/64760726.cms
  2. https://sportstar.thehindu.com/cricket/ipl/in-ipl-its-a-for-analytics/article23408804.ece#
  3. http://www.affineanalytics.com/blog/howstat-application-of-data-science-in-cricket/
MECBot
  • Posted on April 11, 2019

Leave a Reply

Your email address will not be published. Required fields are marked *

Request a Demo

Please fill in the details below to get a demo.

OR