Domestic scouting in the IPL

Creating success in a cost and time efficient way

If you are interested in the information I can provide your organisation (which can be a lot more detailed and advanced than my Substack content), please feel free to get in touch via email.

T20 franchise leagues in cricket have a completely different dynamic to, for example, Premier League football teams. Because of this, an IPL team cannot try to mirror the approaches of a Premier League team in terms of player development.

In the Premier League, you have clear benefits to creating a strong youth system that can produce players for the first team, and this has now taken a step further to hyper-specialization from coaches with resources being allocated to try and develop rare roles and skillsets (e.g. 20+ goal a season strikers) which are generally far more scarce, and therefore more expensive, than other roles and skillsets.

The benefits to developing high quality young players are multi-fold. First of all, a team gets to incorporate them into their first team for minimal development cost (far less than any transfer fee). Furthermore, these home-grown players often have greater affinity and loyalty to the club and fanbase. Most importantly, though, you’ve developed a saleable asset. Even if a player isn’t perceived to be good enough by a top Premier League team, they could be sold to a lower level club or the development club retains a sell-on percentage in case they made a mistake and the player turns out to be really good eventually.

Franchise cricket teams cannot develop saleable assets. Maybe they will in the future, as the ecosystem evolves, but they can’t right now. Probably the most successful T20 team worldwide for developing players has been the Lahore Qalanders in the PSL - they’ve evolved from being a joke franchise in the early years of the PSL to being the flag bearer for player development, particularly for young pace bowlers.

But what benefit do Lahore have from doing so? Yes, they can find the next Shaheen Afridi maybe, but the draft dynamics mean that it is very difficult for them to either 1) hold onto all their developed talent and 2) cannot yield any income from selling them. There will be situations where Qalanders will develop a player, and just as they are getting more experience and towards peak level, cannot retain them and they go off and play for, say, Multan Sultans, with no compensation for that development being paid.

Likewise, in the IPL, look at RCB. They’ve run trials for women’s domestic pacers who can bowl 130+ kph - however, given that very few women’s pacers in world cricket can currently hit this speed, I’m not entirely sure about the logic of it and therefore am sceptical as to whether will succeed. As a poster on Reddit commented, it’s about the equivalent of a men’s team running trials looking for a 165kph bowler!

A lot has been made of RCB working on ‘Hinterland Scouting’, where they try and look for unearthed talent and even has an AI partner assisting them in the process. Avinash Singh is one player who was apparently found via such methodology. However, he’s still to make his debut this season despite RCB using 20 players from their roster so far in the tournament.

And this leads to a problem. Finding a player via such methodology clearly isn’t particularly cheap. Say a player that an IPL franchise finds via this scouting is an extremely raw talent (a likely scenario) who needs to be polished before making his debut. He’s on a low salary, but his development can go two ways.

First of all, he debuts and suddenly creates a big impact. Great - the franchise now have a player on their hands. Then the mega auction comes along. The player, hypothetically, decides that instead of 20 lakh they first accepted when they joined they now want 6 Crore to be retained, quite justifiably because they are now one of the leading players in the IPL.

Then this puts the franchise into a situation where they might not be able to keep the player in their roster, because of limited retention spots or because they don’t have the finances (using the RCB example, they pay Kohli, Maxwell, Hasaranga, Harshal and Siraj a big percentage of their budget and are unlikely to be keen to let any of them leave). They have no choice but to release the player and he goes to a rival team - all that development, patience, investment now benefits a rival. You can see why some IPL teams want to move away from the mega auction model.

The other development could be that the player simply doesn’t make it at the franchise. He is too raw to yield quick results, or doesn’t develop in the way that the franchise anticipates, so they get released. In the case of a player who is too raw and needs long-term patience to develop into an asset, they’ve just released a player who they’ve spent a lot of money and time on, and suddenly in a year or two’s time, a rival team benefits from that effort when it all clicks for that player.

At this stage of the IPL’s revolution, teams are even reticent to get involved in mid-season trading, and post-season trading still doesn’t happen that much. So a fully evolved transfer market like football is some way off in franchise cricket - but it would be a great way of rewarding teams who put a lot of time, effort and money into developing talent.

Given this, wouldn’t it make more sense for teams to do their domestic scouting in a far more cost-efficient and time-efficient manner? Data available for free in the public domain, or at low cost, can be an excellent predictor of future success for players at lower levels. That’s a key element of data analytics - using past performances to create competitive advantages by assisting future decisions (and being evidence-based, plus accountable) in an industry (like most industries) where perfect information doesn’t exist.

In my e-book ‘Strategies for Success in the Indian Premier League’ written in 2020, I used a number of examples of this. First of all, I described the structure of Indian domestic cricket as ‘I look at the main Indian Domestic T20 Structure as a three-tier pyramid, with the Indian Premier League at the top, then followed by the Syed Mushtaq Ali Trophy and then both the Karnataka Premier League and Tamil Nadu Premier League, which aren’t massively dissimilar based on my perceived standard.

Perhaps this isn’t hugely dissimilar to the structure of football in England, with the Indian Premier League potentially the equivalent to the English Premier League in football. The Syed Mushtaq Ali Trophy might be the English Championship, while the Karnataka and Tamil Nadu Premier Leagues could represent the English League One.’

In addition:

‘Teams certainly need to be mindful of how performances are expected to translate from those leagues to the Indian Premier League, as well as being also mindful of data sample size. In many cases, purchasing a player who has only played in these leagues for a high price might be viewed as quite a risky proposition.

Indian Premier League teams will have scouts actively looking at these leagues trying to find unearthed gems, but my research shows that - as is often the case with the eye test - this practice can be rather hit or miss. Players are recruited for large sums but frequently have little impact, while there also look like some players still in these leagues which data makes a solid case for being good enough to step up to the Indian Premier League.

With so many matches in these leagues, all of varying standards, it is very difficult, if not impossible, for a visual scout to cover all the players in these competitions while also efficiently accounting for quality of opposition. However, data can do exactly that, as well as doing so in a time-efficient manner, allowing us to highlight some players who look to have high current ability or potential.’

There is an excellent article on The Athletic website called ‘Does scouting in football work?’ Here’s some interesting quotes from an academic study performed with scouts and coaches:-

“As far as statistical tests were concerned, the researchers conceded, “we found no evidence that observing soccer players in games hurt or helped validity”. They couldn’t prove that it mattered whether scouts watched players at all.

In baseball, where sophisticated new technology can do parts of a scouting department’s job more cheaply and reliably, some teams have done just that.

Evan Drellich’s new book Winning Fixes Everything tells how the Houston Astros drastically cut their scouting operation after finding that adding scouts’ opinions to the data they had on pro players made their evaluations “about two per cent better”.”

When I wrote the book, as examples, I ascertained at the time that only three batters - Rishabh Pant, Ishan Kishan and Devdutt Padikkal - were able to hit 300+ runs in the SMAT at a strike rate of 150+ while being 21 years or below. A pretty good starting point for working out potential superstars. Padikkal currently looks to me to be a ‘distressed asset’ who could thrive in a different role at a different team.

You can do similar with the Vijay Hazare Trophy, 300+ runs at SR 100+ at a young age. Finds you Ruturaj Gaikwad, Virat Singh, Ishan Kishan, Prithvi Shaw and Yashashvi Jaiswal. With the exception of Virat Singh, all would have a current or previous expected valuation of 10+ Cr at auction. Imagine if you could pick up these guys on a conveyor belt year on year for 2-3 Cr - the risk/reward ratio is insanely good.

Not just this, but there are repeated instances of current superstars having superb lower level data before their IPL debut. The filters I used in the previous examples are obviously basic, but much more advanced metrics to compare players and take into account all sorts of nuances can be utilised too, in order to create another further level of due diligence.

Concluding, in short, it is my view that it is very difficult, if not virtually impossible for teams to know the ability of every player in the Indian domestic player pool without using predictive analytics.

While there are few guarantees with respect to predicting a player’s future career, identifying a group of players through data who have the potential to yield high rewards from a low risk makes a lot of sense, and likewise, it also makes sense to utilise data to help teams predict with confidence which players are highly unlikely to be good enough.

Can visually scouting a few matches and the variance of a trial environment suffice in terms of thorough analysis before purchase? I don’t think so at all. That’s not to say there’s not a role for visual scouting, but it’s tough to make a case for them being cost-efficient or time-efficient. If going to a match with a blank notebook and no preconceived ideas on players, they need to focus on identification of players and then if that player’s data, in the eyes of a trained recruitment analyst, is good enough then the player is shortlisted.

While I firmly believe that a team recruited solely via data would be a lot better than a team solely recruited via the eye test, fortunately teams don’t have to constrain themselves to one approach only. My ideal approach would be to generate a shortlist of players via thorough data analysis to then ask the visual scouts to watch further.

Ideally, this would create a cost-effective and time-effective ‘double-tick’ method whereby there is agreement between data and the eye test that a player is either good enough, or not good enough - and let’s not forget, swerving bad deals is as important as obtaining good deals in the marketplace.

If you are interested in the information I can provide your organisation (which can be a lot more detailed and advanced than my Substack content), please feel free to get in touch via email.