Sourcing Domestic Players for the IPL
Digging into the numbers behind player sourcing & exploring biases, opportunities, and scouting wins
This is the second in a two-part series. The first part covers the sourcing of overseas players.
In the discussion on sourcing overseas players, we saw how there is a significant bias towards bringing in familiar players who have either been highly visible in the international scene or in the Australian cricketing circles. While there are some situations, especially with bowling where overseas players with international experience have an edge over those without, these biases are largely unwarranted for batting. The aussie bias is almost completely unsupported by data, although it has a human explanation — familiarity and if I am critical; camaraderie, if I am generous; chemistry. In fact, if anything, there may be room for more caribbean bias on the batting side. We also saw how English cricketers, cricketers from the British Isles really, are severely under-represented given their overall T20 batting performances in recent times.
With all the talk around how the IPL is discovering domestic talent and how it is feeding Indian cricket across formats, I wanted to explore the make-up of the domestic player pool at the IPL and how much of a role guerrilla scouting; with scouts going to lesser known tournaments and fast-tracking players based on acute technical observations might be playing.
Spoiler: The big takeaway of this piece is that fewer new players are able to make it to the IPL. It isn’t necessarily the case that the best are waiting in the wings. But, the not so good are playing on, failing to make way for the better and that is where the opportunity to optimize lies.
As we saw in the other piece, 151 players played at least one game in IPL 2020. 96 of these were domestic players, while 55 were overseas. I used country of birth to decide whether a player was domestic or not, so there may be a couple that I miscounted.
43% of the debutants in IPL 2020 were overseas players although overseas players only make up roughly a third of a squad.
Exposure for domestic talent
The IPL is purported to be a tournament that gives critical exposure to Indian talent by giving them the platform to play with and against the best in the world. There were 23 players who made their IPL debuts in 2020. 13 of them were Indians, 10 were overseas players. Given a squad size limit of about 24, and a quota of 4 in the playing 11, there should be ~33–36% of each squad that is made of overseas talent. So the 43% debutant ratio for overseas players seems a tad high. And this is lower than it has been recently.
I am not questioning the ratio of domestic players getting a look in or the current state of the balance between domestic and overseas prioritization. I am curious about whether this is intentional and whether it serves the purposes of the endeavor. I am saying, “Here is where we are. Does this make sense?” The IPL is what it is, in no small part because of the quality of cricket that is on display and overseas players as well as Indian international stars are a big contributing factor to this. Without this, we may as well stick to the Syed Mushtaq Ali trophy (we will see in a bit, indicators on the difference in quality between the SMA & the IPL). Also, overseas players getting a look in doesn’t always cannibalize domestic talent because they have a separate quota. However, teams may not want to risk getting too many debutants on board in a given year and this may play a bit of a cannibalizing role after all.
Now, there have been several changes to squad rules over the years that make the above numbers a bit tricky to interpret. I wish these were easier to track, but even given the few that I was able to pull up, the number of new domestic faces that get a look in is lower than I’d expected. Here are some of the changes in rules that I could find:
- The squad size limits and overseas quotas have changed multiple times over the years. 30 players including 10 overseas players were allowed until 2010, this was increased in 2011 and reduced to 27(9) in 2014, but teams didn’t utilize the full quota which has subsequently been dropped to 24–25(8)
- In (until?) 2014, U19 players could only be picked if they had played first-class or list-A cricket.
- In 2014, uncapped domestic players were auctioned first & teams were allowed to retain 5 domestic players. However, this is also when the least number of Domestic debuts happened.
Next, let’s dig into the background experience of the folks who do get picked in the IPL so that we can then see if there are any blindspots there and also to evaluate the competition for that pool of players.
IPL is a stepping stone to the Indian senior side
Of the domestic players that played in the IPL between 2018–2020, only 13% had played senior intl. cricket before their IPL debut. On the other hand, there were ~50 active Indian international players in this period (2018–2020) and all but one of them — Pujara, played in the IPL in that time frame.
The tendency to pick international cricketers that we saw in the case of the overseas players doesn’t seem to carry to the domestic side. However, legacy international players do tend to stick around and contribute to the lower debut rate for domestic players in the IPL. Of all the folks who did play intl. cricket before the IPL, only one started in the IPL in the past 12 yrs. Almost everyone started in 2008!!! Since this was the first season for the IPL, we can safely say that almost no one gets into the national team without playing the IPL first. Even Virat Kohli made his IPL debut before his international debut. Even the likes of Saha & Rahane who are not seen as T20 fits, made their IPL debuts before breaking into the national team.
There are 16 Indian players who played in IPL 2020 who have been playing since the first edition 13yrs ago!
Players who are picked for national teams and do well on the international circuit often end up playing for several years for their country. It’s what we call a career. The players on a national team play with each other throughout the year, across the world and there is a lot of value in growing careers, and investing in building teams over multiple years. The demands of the international setup also warrant less churn.
There is a hangover of this process in the IPL. Moreover, T20 is a different game and most international cricket is not T20 cricket; it’s the longer formats that are more prevalent.
Franchise cricket doesn’t need to be run the same way as a national team. This doesn’t apply to Indian players yet, thanks to the BCCI, but players playing in T20 franchises need to learn to gel with different team mates and fit into different team strategies for every tournament they play in. Tournaments are hectic, with less time to react, but they are also relatively shorter than most international tours have traditionally been.
All this, to say that there are 16 Indian players who played in IPL 2020 who have been playing since the first edition 13yrs ago! Can they all still be the best possible picks? It is very likely that there are better replacements waiting in the wings. But, the hangover from how national teams are run means that stability over multiple years is over valued. There are aspects of the auction process that also contribute to this.
But these players are international stars. And this is part of how the IPL is designed, isn’t it? It is meant to give opportunities to local talent. Getting a bunch of international stars to create competition and to get the quality up, leads to more eyeballs and better training for domestic players. The presence of Indian international stars works as a double edged sword — it blocks potential talent from the exposure, but it also brings in excitement and eyeballs.
Experience before the IPL
We have seen that most players don’t play international cricket before the IPL. So what are the platforms that teams look at for domestic talent? Well, hold off striking out the dependence on the international circuit for vetting players.
More than half of the domestic players at the IPL have some form of international exposure before they make it to the IPL
If we include U19 internationals for domestic cricketers, 53% (a jump from 13%) of them played internationals before their IPL debuts! So these guys weren’t as raw as we had initially thought after all. For overseas players, including U19 takes the proportion up to 93% (from 88%).
The Syed Mushtaq Ali trophy (SMA) is considered to be the breeding ground for the IPL. While this is true, teams do tend to look at players who have had more than just SMA experience. Including international and non-international cricketers who played in IPL 2018–20, ~2/3rd played domestic first-class cricket (e.g. Ranji trophy, Duleep trophy, Irani trophy), domestic list-A games e.g(Deodhar trophy, Vijay Hazare trophy, Challenger series) as well as the SMA.
Let’s focus on these three to create a candidate pool of domestic players for IPL 2020 for our evaluation of opportunities.
Who is playing the SMA?
612 players played more than 3 SMA games in 2019 (there are 5 league games that every team is guaranteed to play, so a threshold of > 3 games suggests a reasonably active player). Of the 612, 148 had played in the IPL before. Out of the other 464, 40 played U19 for India before. Additionally, there were 65 international cricketers who have also played in the IPL that played in the SMA. So, more than a third of the players who are playing in the SMA have been vetted in some forum of relevant significance already. This is great for the SMA & the quality of cricket played. It is also a testimony to the value that these players, especially the ones who have already made it to the IPL, see in SMA experience. This is analogous to Test stalwarts playing in the Ranji, Duleep & Irani trophies.
more than a third of the players who are playing in the SMA have been vetted in the IPL, U19 internationals or senior internationals already
On the other hand, there are 10 India players who played less than 4 games in the SMA. This included the likes of Dhoni, Kohli, Jadeja, Sharma etc. It also included pace bowlers like Shami, Bhuvi (2) & Bumrah which is understandable if they were watching their workload. Notably, Ishant Sharma was not one of them (played 7). Yadav played 4.
IPL teams have to pick 30% of contending domestic players. This is much less competition than the 9% for overseas players
For folks who had played U19 cricket for India before their IPL debut (2018–20), the median number of games played in domestic first class, domestic List A & the SMA were 5, 9 & 6 respectively. The median for U19 games themselves was 12. For those who didn’t play U19 cricket for India, these numbers were 9, 12 & 9.5 respectively. Using these as guidance, if we apply a uniform filter of more than 20 total games, and at least (5 U19 games), 5 SMA, 3 first-class & 8 List-A games on the 600 odd players, we have 213 non-U19 and 16 U-19 players in contention for an IPL debut in 2020.
Overall, this is ~380 (213 + 16 + 148 who have already played in the IPL) players vying for ~110 spots across 8 teams i.e. ~30% as opposed to 9% for overseas players. Even if we consider all of the ~600 players as contenders, ~18% will play. So there is much less competition for domestic players as compared to overseas players as I had alluded to previously.
Are teams picking the best domestic options?
Let’s examine the performance of the available options as we did for overseas players. Unfortunately, I only have ball-by-ball data for about a 3rd — 103 of the 380 players for batting and 72 for bowling.
It’s less about who didn’t play and more about who did
Batting
Again looking at global percentiles of survival & scoring rate over expectation in the last 3 years before the 2020 auction for those with available data, we see that most of the best players actually played in the 2020 edition of the IPL. Raina & Parthiv Patel weren’t really options for different reasons. So Manan Vohra was the only player in the top 20 percentile for scoring rate who didn’t play in 2020. For an opening or one drop bat that he plays as, Vohra’s stats don’t seem to really warrant an argument here. However, there are several others who did play in 2020 that show up as sub-par on the charts. For players that haven’t played much, we should assume that they are playing not on reputation, but on the basis of potential seen by the coaching and scouting staff e.g. Riyan Parag. For folks that have been playing for a while, I am not sure what a sound argument would be for them to play on, but the hope is that the reason isn’t legacy or reputation.
In order to help us take a look at the others for which we don’t have ball-by-ball data, I did the following:
- Find the relationship between (R/W)PO in the SMA & in the IPL for folks that have played at least 200 balls of each in the 5yrs before the Dec 2019 auction. (For everyone including folks for whom we have ball-by-ball data)
- Find number of balls needed for a players estimate to settle.
- Calculate projected IPL strike rate for batters who have played balls beyond the threshold determined in #2.
Since we are using a simple linear translation between SMA & IPL stats, one at a time, we aren’t really going to get a different relative view between players that is different from using just the SMA stats on their own without translation. However, we are alleviating sample size concerns by using the threshold from #2 and we are ensuring that there really is a relationship between SMA performance and IPL performance by doing #1.
Notably, we are not looking at situation based expectations and hence we may end up giving undue advantage to folks who played in more conducive situations e.g. openers & late order batters.
The IPL RPO for players is in a very tight range even as their SMA RPO varies between 6 and 10. WPO on the other hand is equivalent in a lot of cases. I couldn’t help but gawk at Pandya’s RPO difference between the two tournaments there.
For these players, it looks like the difference between RPO calculated at x balls played to the final RPO for SMA goes under 0.25 runs at about 120 balls.
I went ahead and used an aggressive threshold of 100 balls. There is some indication here that teams are using the available raw info from SMA for their picks. Most of the players picked are high on the RPO and low on WPO. This also shows the importance of scouting, subjective observation and advanced methods to understand performance within the context of a situation. Krunal Pandya was one of the top performers based on survival factor and additional RPO relative to expectation, but you wouldn’t pick him based on his raw SMA numbers given the high WPO; context is important. He is the top purple dot here at 7.77 RPO, 0.47 WPO.
Bowling
Going through similar motions for bowling, we see that unlike a good number of batters, most domestic bowlers fail to make it to the top of the global rankings. Of the unpicked contenders, Pawan Negi, the SLA from Delhi, stands out amongst the ones who played and have been doing well.
Amongst the ones that played, some of the lower performers that stand out are Siraj (5% on RPO, 25% of WPO), Vijay Shankar (9% on RPO, 27% on WPO) & Ishant Sharma (48% on RPO, 5% on WPO) amongst seamers who are likelier to be picked based on a combination of performance and reputation in other formats and Piyush Chawla (6% on RPO, 9% on WPO) among spinners.
Again, let’s use the same methodology as we did for batting to project RPO & WPO for the IPL from SMA performance in the 5yrs prior. I changed the ball threshold to 100 instead of the 200 we used for batting.
It is evident that SMA RPO isn’t a great predictor of IPL RPO while SMA WPO is a decent predictor of IPL WPO. It turns out that restricting the RPO projection to seam bowlers helps. For spinners, our simple model still fails to give us good projections. This would be great to understand, but it would lead us down a whole new tangent. So I am going to table RPO projections for spinners here.
Given that the RPO & WPO for bowlers also seems to reasonably settle around the 100 ball mark, I am using that as the threshold here as well.
For bowlers, seam bowlers in particular since we are taking a restricted perspective here, again unlike batters, there is a potential opportunity in looking at players who aren’t playing but who do better than the ones that are. This isn’t unexpected, since, as we saw earlier, the competition is largely in the mid and bottom tiers. In general, competition at the top usually has tighter margins but is sparser in volume, while competition in the lower tiers tends to have more players and slightly wider margins.
This also provides some perspective on the value of overseas pace bowlers. They add massively to the quality of pace bowling on display at the IPL.
Scouting wins
On the domestic front, we can consider a few indicators of potential scouting wins. A very popular scouting victory is Jasprit Bumrah, but remember, Bumrah played 9 games in the Syed Mushtaq Ali before his IPL debut. He hadn’t played anything else, but 9 is close to the median (9.5) for non-U19 players that we saw earlier — half of all the players who played the IPL played 9.5 SMA games. Bumrah hadn’t played in any first-class or list-A games, though.
We saw earlier that ~2/3 of players played first-class, list-A and SMA before they got into the IPL. What about the rest?
Of the non-U19 lot, only 5 players did not play SMA before the IPL. 2 of them, Tewatia and Shreyas Gopal had only played first-class cricket; Tewatia had only played 2 Ranji trophy games before his IPL debut in 2013. Over the years, going in and out of the IPL, he got a look, but that initial pick was made without much signal in terms of numbers. This article from wisden captures Tewatia’s journey well.
Of U19 players, 15 hadn’t played the SMA before their IPL debuts. But of these, only 1 hadn’t played any domestic cricket of note at all — Shivam Mavi. He had played 13 U19 ODIs including 6 U19 world cup games and 2 U19 Tests. From what I can gather, his selection must have been based on his world cup show. Not exactly a scouting win, but definitely based more on what people saw than what he had delivered. Of the 2020 debutants, Nagarkoti, Kartik Tyagi & Yashaswi Jaiswal hadn’t played in the SMA. All of them had played some Vijay Hazare trophy games. Jaiswal brought the most signal having played Ranji & Deodhar as well.
A deep dive into how performance across these different stages translates to IPL performance would be an interesting effort to take up.
If you enjoyed this piece, check out more of my work at Boundary Line and follow along here & on twitter @amol_desai
