Ceiling math
What is the player top-10 percent ceiling in this venue against this opposition? Use for differential mega contests.
Read more →Captain calibration lab
The captain calibration lab on Win Come Sports is where you compare captain options across match-up, form, venue, and opposition. The lab surfaces ceiling math, consistency math, and differential math - three different lenses that turn the 2x captain pick from a coin flip into a data-backed decision.
4 Lenses
Ceiling, Consistency, Form, Match-up
14 Days
Recent Form Window
38,400+
Historical Matches
12s
Live Update Cadence
Calibration lenses
Consistency math
What is the player 75th-percentile fantasy score over the last 14 days? Use for bankroll-preserving qualifiers.
Read more →Recent form
Rolling 14-day fantasy score average, weighted by match difficulty. Updated after every contest.
Read more →Match-up data
Historical performance against the opposition and at the venue. Updated nightly from the official scoring feed.
Read more →How to use the lab
1
Pick the match
Select the upcoming match on the captain calibration page. The lab loads all available players.
2
Compare captain options
Tap 2-4 captain candidates to compare them side-by-side across all 4 lenses.
3
Read the recommendation
The lab surfaces a primary recommendation and a differential alternative. Read both before locking.
4
Lock your captain
Lock the captain in your squad. The lab can save drafts for in-play swap windows on supported contests.
Captain pick anti-patterns
Captain by reputation
Picking the most famous player as captain is the single most common mistake. The lab surfaces better options 64 percent of the time.
Captain by recent single match
One big match distorts the 14-day form. The lab weights form by match difficulty to avoid this trap.
Ignoring venue
Some players play dramatically differently at certain venues. The lab surfaces venue-specific historical data.
Captain armband
The captain armband multiplies your player score by 2x. Choose wisely with the calibration lab.
Trophy gold
The captain pick that wins your bracket round unlocks the gold trophy stage payout.
Live score
Track live captain score during the match. In-play swap windows allow captain changes on supported contests.
Calibrate your next captain
Open the captain calibration lab before your next contest. The 4 lenses turn the 2x pick into a data-backed decision.
Win Bracket Context
Tournament context for Captain Picks
Captain picks on WinCome Sports are the single highest-use decision in any fantasy 11 bracket. The captain scores 2x, the vice-captain scores 1.5x, and the right captain is worth 30-40% of your expected value. The captain-pick calibration model is built on 7 years of ball-by-ball data, 142 features, and a daily recalibration cycle.
30-40%
Captain Pick Edge
Of expected value on a fantasy 11 bracket
142
Calibration Features
Form, venue, opposition, weather, captain pick history
Daily
Recalibration Cycle
Model retrained on the most recent 30 days of data
Frequently asked questions
What makes a good WinCome Sports captain pick?
Three factors in order of weight: (1) the player's expected points in the match, (2) the player's ownership percentage (lower is better for differential captain picks), and (3) the venue-specific historical performance. The captain-pick calibration model ranks all available players on these three factors daily.
Should I captain a top-order batsman or a strike bowler on WinCome Sports?
It depends on the match. In high-scoring venues, top-order batsmen tend to score more fantasy points. In low-scoring venues, strike bowlers tend to score more. The captain-pick calibration model handles this for you automatically — open the captain-pick page before the toss and the model will recommend the right pick.
Captain picks are the most important decision in any WinCome Sports fantasy 11 bracket. The 30-40% expected value impact, the 142-feature calibration model, and the daily recalibration cycle mean you can trust the recommendations. The next captain-pick recommendation publishes 4 hours before toss — your calibration is already running.
Win Bracket Deep Dive
The captain-pick calibration model on WinCome Sports
The WinCome Sports captain-pick calibration model is built on 7 years of ball-by-ball data. The model uses 142 features: form (recent performance), venue (historical scoring at the venue), opposition (matchup-specific performance), weather (pitch and weather conditions), captain pick history (how the player has performed as captain in the past), and 137 other features that the model has learned are predictive of fantasy points. The model is a gradient-boosted decision tree ensemble. The ensemble has 512 trees, each trained on a different random subset of the data. The trees vote on the captain pick recommendation, and the recommendation with the most votes wins. The model is recalibrated daily on the most recent 30 days of ball-by-ball data, so the recommendations improve as the tournament progresses. The model's top-1 accuracy (the percentage of matches where the top-1 recommended captain pick was the highest-scoring player on the field) is 71.4% across all tournament formats in 2026 Q1. The top-3 accuracy is 92.1%. The model also produces a confidence score for each captain pick recommendation: high confidence (top-1), medium confidence (top-3), low confidence (top-10). The user can use the confidence score to calibrate the size of their entry: a high-confidence captain pick supports a larger entry. The captain-pick recommendations are published 4 hours before the toss. The recommendations include a captain pick, a differential pick (a low-ownership alternative for players who want a contrarian captain), and a vice-captain combination. The recommendations are accessible from the live matches page and from the captain-pick page inside the WinCome Sports app.
How to use the captain-pick recommendations on WinCome Sports
The captain-pick recommendations are designed to be used as a starting point, not as a final answer. The user should review the recommendation, check the reasoning (which is published alongside the recommendation), and decide whether to follow the recommendation or pick a different player. The user's domain knowledge (e.g., a fan who watches every match) is a valuable complement to the model's recommendations. For most users, the best strategy is to follow the top-1 recommendation on high-confidence matches. The top-1 recommendation has a 71.4% accuracy rate, which is higher than the average user's manual pick rate. The top-1 recommendation also has a higher expected value than the average user's manual pick. For users who want a contrarian captain pick, the differential pick is the right choice. The differential pick is a player who is low-ownership and high-upside. The differential pick has a lower top-1 accuracy but a higher expected value if the player performs well (because low-ownership players have a bigger impact on the leaderboard). The differential pick is a higher-variance, lower-floor option. For users who want to combine the model with their own judgment, the user can pick their own captain and use the model as a sanity check. The model publishes the top-10 captain pick recommendations for every match. The user can check where their pick ranks in the top-10: a top-3 ranking is a strong signal that the user's pick is consistent with the model; a bottom-7 ranking is a signal that the user's pick is contrarian.
