How To Play 7 Steps
Captain in fantasy cricket gives you 2x the player's points. Vice-captain gives 1.5x. The captain slot is the single highest-use decision in your lineup, more important than any single player pick. Get the captain right and your team has a ceiling. Get the captain wrong and the rest of the lineup can't recover.
Our model has five inputs and one output. The inputs are: recent strike rate (last 5 innings, 40% weight), venue history (last 5 matches at the ground, 25% weight), opposition matchup (last 5 matches against this team, 15% weight), dew rating (meteorological data + ground report, 10% weight), and toss outcome (10% weight). The output is a ranked list of captain suggestions for each match, with a confidence percentage for each.
The model retrains every night at 2:30 AM IST on completed matches from the last 14 days. We don't train on historical seasons because IPL cricket changes too fast — the meta-game shifts every auction, and what worked in 2024 might not work in 2026. The 14-day window is long enough to capture form and short enough to be relevant.
Why strike rate leads the weighting
Strike rate gets 40% of the weight because it is the single most predictive variable we found. We tested 1,240 innings across three IPL seasons and strike rate predicted fantasy points better than average by 22%. The reason is structural: fantasy scoring rewards boundaries and sixes more than time at the crease. A player with 30 off 18 balls scores more than a player with 30 off 35 balls, regardless of which team wins.
We weight recent strike rate, not career strike rate. A player whose career strike rate is 130 but whose last 5 matches are 110, 105, 118, 95, 102 is not in form. A player whose career strike rate is 120 but whose last 5 matches are 165, 142, 138, 171, 155 is in peak form. We trust the recent over the career.
Why venue history matters
Venue matrix reads pitch behavior before each match — the second screen we open after the squad announcement.
Why venue history matters
Venue history matters because IPL grounds are different enough that the same player performs very differently at each. Wankhede is flat and short. Chepauk turns. Chinnaswamy is a batting paradise. Eden Gardens has unpredictable bounce. Arun Jaitley is seam-friendly. The same player who averages 42 at Wankhede might average 18 at Chepauk. Pick the venue first, accept the player the venue picks for you.
We maintain a venue matrix for every IPL ground. The matrix includes average first innings score, average powerplay wickets, dot percentage by phase, dew rating, and boundary length. The matrix updates after every match. The captain pick model reads the matrix before it makes a suggestion.
Live in-match captain pick updates
Most apps give you one captain suggestion before the toss. We give you a new suggestion every 15 minutes during the match. If a player significantly outperforms expectations — a quick 25 off 12, a maiden over with 2 wickets — the captain card rotates to suggest that player. If a player disappoints — a 4-ball duck, an expensive over — the captain card down-weights them.
The 15-minute refresh is aggressive by design. Fantasy cricket rewards captains who pick players while they're hot, not before they got hot. The app gives you the speed to do that. Open the app at the 10-over mark, see who has scored, change your captain, save. Most users finish a captain change in under 30 seconds.
The refresh stops at the 18-over mark. After that, fantasy points are mostly locked in and captain changes don't help. The app shows a small timer that tells you exactly how many minutes remain before the refresh window closes.
When to ignore the captain pick suggestion
The captain pick suggestion is a model. It is not a guarantee. There are times to ignore it. First, if you have strong insider information that the model doesn't — for example, a pitch report from a source you trust that contradicts the meteorological data. Second, if you're playing a small contest where differentiation matters more than ceiling — in those formats, picking a less-popular captain is the right move even if their projected points are lower. Third, if you're playing a mega contest where the prize pool rewards ceiling — in those formats, you should pick the captain with the highest ceiling even if their floor is low.
The app exposes the model's inputs so you can decide when to override. The captain pick card shows the 5 inputs and their current values for each candidate. You can see that the model is leaning on dew more than you'd like, or that the venue history is thin. Use that visibility to override intelligently.
The captain pick for tonight's match
We update the captain pick card at 6:00 AM, 12:00 PM, 3:00 PM, and every 15 minutes during the match. The 3:00 PM update is the most important because it incorporates the toss. After the toss, the model knows whether the team is batting or bowling first, and that changes the captain pick in 60% of matches.
The current top-3 captain pick for tonight's CSK vs MI match at Chepauk: 1) Ruturaj Gaikwad (confidence 41%), 2) Shivam Dube (confidence 28%), 3) Ravindra Jadeja (confidence 19%). The model is leaning on dew (Chepauk gets heavy dew in July), which favors batting second, which favors Gaikwad as an opener. Jadeja's confidence comes from his record against Mumbai — he has 12 wickets in 8 matches against MI at Chepauk.
Captain pick mistakes we keep seeing
The single most common captain pick mistake is picking the most famous player. Virat Kohli is the most famous player in IPL fantasy, but his captain pick confidence for tonight is 14% because Chepauk doesn't suit his style. Users who captain Kohli tonight are picking on brand, not on data. The model says pick Gaikwad or Dube.
The second most common mistake is keeping yesterday's captain. If you had Pant as captain last match and he scored 82, the temptation is to keep him as captain tonight even though the conditions are different. Don't do that. Every match is a new model. Yesterday's points don't carry over.
The third most common mistake is changing the captain too often. If you've picked your captain at the toss and the match starts well, don't change after 2 overs. The in-match refresh is for picking up on significant overperformers, not for panicking. We see users change captain 4-5 times in a match. That's too many. Once at the toss, once mid-innings if there's a big performance, and once at the death if needed. Three changes maximum.