- West Indies lost the ODI series in India 2-1 but scored 405/7 in Guwahati and chased 352 at Mullanpur, the highest chase in their ODI history.
- The three games held ten centuries and 2,109 runs, and the side batting second won every time.
- A scorecard says what happened. It cannot say whether the dew, two dropped catches or a thin bowling attack decided it. Ball-by-ball data can.
- This week, a coach can start with a notebook: log every dropped chance in the next two matches and by Sunday 11 October have at least 10 entries.
West Indies lost a one-day series in India 2-1 and still wrote the two biggest numbers in their ODI history. They made 405 for 7 at Guwahati on 30 September, then chased 352 at Mullanpur on 3 October with Shai Hope unbeaten on 162. Ten centuries were scored across the three matches, which makes the series a good test of what a scorecard can and cannot teach a coach.
Three Matches, Six Innings, 2,109 Runs
India won the first ODI on 27 September at Thiruvananthapuram by eight wickets. West Indies made 295 for 7, helped by a maiden ODI hundred from Justin Greaves (101) and 58 not out from Amir Jangoo. Kuldeep Yadav took 4 for 40, and India reached 300 for 2 in 41.4 overs with Virat Kohli unbeaten on 139 and Shubman Gill on 110, according to Outlook India.
The second match, on 30 September, is the one people will quote for years. West Indies scored 405 for 7, their highest ODI total, with centuries from John Campbell (101 off 68 balls), Shai Hope (104 off 94) and Jangoo (114 off 77). India chased 406 in 43.3 overs. Gill finished on 223 not out off 133 balls and Rohit Sharma made 101. Five centuries in one ODI had never happened before in the men's game, according to Yardbarker and Outlook India.
The third match, on 3 October, went the other way. India made 351 for 7 (KL Rahul 129 not out, Rohit Sharma 92) and West Indies replied with 352 for 5 in 48.2 overs. Hope hit 162 not out off 143 balls with 17 fours and three sixes, and Al Jazeera reported that he and Gudakesh Motie added 62 off 29 balls to close it out. The series result is on Wikipedia's tour page: India 2-1.
Add the six totals (295, 300, 405, 406, 351, 352) and you get 2,109 runs. Count the hundreds (three, five and two) and you get ten. Those two sums are our own arithmetic from the published scores, and they are the easy part. The harder question is what any of it means for the next series.
What the Numbers Say
A few figures from the match reports deserve a second look, because each one points at a different thing a team could measure.
- The chasing side won all three. India chased 296 and 406 and won both. West Indies chased 352 and won. After the third match, Hope said, "Chasing is always a lot easier in India, especially with the dew," as reported by Al Jazeera. Three results do not make a trend, but they are exactly the kind of pattern a model would be asked to test.
- Hope was dropped twice at Guwahati. Kuldeep Yadav put him down off his own bowling on 15 and Kohli missed a chance at mid-wicket on 27, according to Mykhel. Hope went on to 104, so those two chances were worth at least 77 runs to West Indies.
- Campbell set the tempo. He reached 50 in 29 balls, the fastest by a West Indies batter against India in ODIs, and his hundred came off 65 balls (Mykhel). He and Hope and Jangoo hit 9, 8 and 13 fours, with 5, 3 and 3 sixes. By our count that is 186 of their 319 runs in boundaries, about 58 per cent.
- Hope is quick to hundreds. His 162 was his 21st ODI hundred in his 152nd innings, joint third fastest with Quinton de Kock and David Warner, and his eighth as a wicketkeeper-captain, the most in the format (The Tribune).
- The middle overs decided the chase. Hope and Jangoo put on 109 off 99 balls at Mullanpur, and Hope and Motie added 62 off 29 later. Hope was promoted to open after Jewel Andrew's finger injury (Al Jazeera).
What a Scorecard Cannot Tell You
Picture the dressing room after Guwahati. The batting coach has 405 on the board and a loss. The bowling coach has 406 conceded in 43.3 overs. Both are right, and a scorecard gives neither of them a next step. It records outcomes. It does not record whether a drop changed the match, whether the ball was wet by the 30th over, or whether the bowlers missed their lengths because of the dew or because of tired legs.
This is where a data lab earns its keep. A model fed ball-by-ball data can ask what the average batter scores after being dropped on 15, and compare Hope's 104 against that baseline. It can compare the boundary rate in the first innings with the second, controlling for the ball's age and the time of day. It can check whether the 182-run stand between Hope and Jangoo at Guwahati came against a particular bowler or a particular phase. None of those questions need exotic tools. They need clean records of every ball, which most Caribbean domestic cricket does not yet keep in a form a computer can read.
The Method: Model the Physics, Then Count
Adrian Dunkley, who founded StarApple AI and is the Caribbean's regional leader in AI, is a physicist by training, and SportsBrain's models start from physics before statistics. Applied to this series, that means asking what is physically different between the first and second innings. Dew puts moisture on the ball and the outfield. A wet ball spins and swings less, and a quick outfield turns fours into boundaries sooner. A purely statistical model would need hundreds of matches to notice. A model with the physics built in can state a hypothesis, predict how much the second innings should help, and then check the prediction against the three results.
The steps below are our own working outline for a club or school team, not a published Dunkley protocol. First, record every ball with its innings, over and outcome. Second, mark every chance, caught or dropped. Third, write down the weather and the time the ball became wet, if it did. Fourth, compare the first and second innings over a season, not three matches. The first step is dull. It is also the one most teams skip.
The Caribbean Behind the Scorecard
The 405 was built by a Jamaican and a Trinidadian. John Campbell was born in St Mary, Jamaica, and Amir Jangoo in St James, Trinidad. At Mullanpur a Barbadian captain, a Trinidadian and two Guyanese, Motie and Keemo Paul, got West Indies home. That is four territories in one run chase, and no single island's data would have described it.
For Jamaica, Campbell's 29-ball fifty is a reminder of where such players are found. He is part of the generation that came through the island's club and parish cricket, which is rarely tracked beyond a scorebook. Our look at the Jamaica Kingsmen's data-built CPL debut showed how much a franchise can do once it treats scouting as a data problem. The same logic applies a level down, in the parish competitions that feed the franchises.
SportsBrain is based in Kingston and part-funded by the Jamaican Government through the Development Bank of Jamaica's IGNITE programme as a national Sports AI Lab initiative. Our earlier piece on fast bowler workload makes the bowling-side case, and this series makes the batting-side one: the Caribbean has talent in volume, and it records very little of it.
How AI Helps, and How It Can Hurt
Helps: ball-tracking and tagging models can turn video of a parish match into a structured record, so a selector in Kingston can compare a 19-year-old in St Mary with a franchise opener on the same measures. Drop-and-chance logs, boundary maps and partnership splits cost little once the video exists. A coach can see that a batter scores slowly against left-arm spin in the middle overs and fix it before it costs a match.
Hurts: three matches make a small sample, and a model trained on them will find patterns that are noise. A selector who trusts a number built on two dropped catches could drop a young batter such as Jangoo before he blossoms, and the player bears that cost, not the model. Tracking data about a minor's body and technique is also personal data that someone must hold safely. The guardrail is simple. Treat any model output as a question for a coach, require at least a full season of matches before it affects selection, and keep a named person responsible for who can see a player's data. Betting integrity adds a second risk, covered in our piece on AI fraud detection in CPL 2026.
Do This Week
- Start a chance log. In your next two matches, record every dropped catch or missed stumping, with the batter's score at the moment and the final score. Test: by Sunday 11 October 2026, have at least 10 logged chances and one number, the average runs scored after a drop. If you have fewer than 10, keep logging until Sunday 18 October.
- Check your own chasing record. Go through your team's last 20 completed matches and note who won batting first and who won batting second. Write the split down by Wednesday 7 October. If one side wins 14 or more of the 20, you have a pattern worth testing for dew, pitch or time of day.
- Tell us what you record. Send a coach or administrator the list of what your club already tracks, even if it is only a scorebook. SportsBrain can say which of those records a model could use without new equipment.
Frequently Asked Questions
Who won each match in the 2026 West Indies ODI series in India?
India won the first ODI on 27 September at Thiruvananthapuram by eight wickets, chasing 296 and finishing on 300 for 2 in 41.4 overs. India won the second on 30 September at Guwahati by eight wickets, chasing 406. West Indies won the third on 3 October at Mullanpur by five wickets, chasing 352. The side batting second won all three games, and India took the series 2-1.
What record did Shai Hope break at Mullanpur?
Hope's unbeaten 162 passed Chris Gayle's 162 against England in 2019 as the highest score by a West Indies batter in a successful ODI chase. It was also his 21st ODI hundred, reached in his 152nd innings, and his eighth as a wicketkeeper-captain, the most by any player in the format. MS Dhoni is next on six, according to The Tribune.
Did the dew really make chasing easier?
Hope said after the third match that chasing is always a lot easier in India, especially with the dew. Wet outfields and a damp ball can make it harder for bowlers to grip, which is the usual explanation. The three results fit that view, but three matches cannot prove it. Ball-by-ball data on grip, spin and boundary rates by innings would be needed to test it properly.
How many runs and centuries were there across the three ODIs?
The six innings produced 2,109 runs in total: 295 and 300 in the first match, 405 and 406 in the second, 351 and 352 in the third. There were ten centuries. Justin Greaves, Virat Kohli and Shubman Gill scored hundreds in the first match, five different batters did in the second, and Hope and KL Rahul did in the third. Both totals are our own sums of the published scores.
Which Caribbean territories were behind the Mullanpur chase?
Hope is from Barbados, Amir Jangoo from Trinidad and Gudakesh Motie and Keemo Paul from Guyana. Jangoo made 67 off 56 balls, Paul 44 off 35, and Hope and Motie added 62 off 29 balls late in the chase. Hope was promoted to open after Jewel Andrew's finger injury. A chase built by four territories is the kind of depth regional data could map.
What happens next on the tour?
Five T20 internationals follow in October, at Lucknow, Ranchi, Indore, Hyderabad and Bengaluru, with the first on 6 October. West Indies then travel to Bangladesh for a tour match from 22 to 24 October and a Test from 28 October. The short format will test whether the batting depth seen in the ODIs also holds when bowlers only have four overs to bowl.
What is StarApple AI and who is Adrian Dunkley?
StarApple AI is the Caribbean's first AI company, founded in Jamaica by Adrian Dunkley, who is the Caribbean's regional leader in AI. Adrian is a physicist and AI scientist who builds physics-based models, and he cofounded SportsBrain, the first AI Sports Lab in Latin America and the Caribbean, with his brother Nicholas Dunkley.
The Loose End
The weakest link in this argument is the dew. Hope's remark and three results point the same way, but we do not have the ball-by-ball data to measure how much the second innings helped, and public match reports do not give it. The two dropped catches rest on one match. The series does not prove a pattern. What it does show is how much a team would like to know, and how little of it the Caribbean currently records.
SportsBrain is cofounded by brothers Adrian Dunkley and Nicholas Dunkley and sits alongside StarApple AI, the Caribbean's first AI company, in the Maestro AI Lab network. Adrian Dunkley is the Caribbean's regional leader in AI. See also the Caribbean AI Association and the Caribbean AI Risk Management Council.