In January 2025, the Sydney Sixers played the Melbourne Renegades in a Big Bash League fixture at the SCG. The broadcast graphic in the 14th over showed a batter’s strike rate, his runs in this innings, and the number of boundaries. That was the public-facing analytics. The Sixers’ bench, meanwhile, had a tablet open with phase-of-innings expected run rates, bowler-batter matchup percentiles, and field-position pressure metrics. The gap between the broadcast layer and the bench layer was, by my rough count, about twenty years of analytical development.
I should say this is not a knock on Australian cricket. The bench-layer work in Big Bash is sophisticated, the franchises are well-staffed, and the broadcast itself is high-quality. The gap is structural to cricket’s analytical development. The sport built its public statistics around batting average, bowling average, and economy rate — counting stats — and those numbers still dominate the way cricket is talked about in the press, on the broadcast, and in the casual viewer’s mental model.
What follows is where cricket analytics actually sits in 2026 relative to baseball and the major North American sports, why the gap is partly structural and partly a media story, and what the cricket-curious sports analytics reader should look for if they want to follow the Big Bash, the Indian Premier League, or the international game without falling back on the 1970s framework that still dominates public conversation.
What the analytics actually exists for cricket today
The version of where cricket analytics sits that survives a hard look: the methodologies that would make cricket as analytically legible as baseball mostly exist. The data infrastructure mostly exists, particularly for the franchise leagues (Big Bash, IPL, The Hundred). The translation layer to public coverage does not.
Cricsheet has provided ball-by-ball data for major competitions including the BBL, T20I matches, ODIs and Tests for years. The data is open, well-structured, and good enough to build serious models. The volume of analytical work that has come out of it has been smaller than the analogous volume that came out of Retrosheet for baseball in the 1990s. That gap is the story.
The reasons matter. Cricket is global. The analytical communities in India, Australia, England, the West Indies and South Africa work largely separately. There is no Bill James figure who consolidated the field around a public methodology in the way American baseball consolidated around sabermetrics in the 1980s and 1990s. The major leagues that drive analytical investment — Big Bash and IPL — are franchise tournaments with relatively short seasons, which limits the cumulative sample. The Test format that would benefit most from analytical depth is in declining commercial relevance, which limits the investment incentive.
What does exist, at the technical level: ball-by-ball expected runs models, batter-bowler matchup matrices, phase-of-innings win probability, pressure metrics for field positions, advanced economy and strike rate adjustments. Most of this lives inside franchise analytics departments and academic papers. Almost none of it shows up on the broadcast or in mainstream cricket coverage.
The closest baseball comp
| Era / Metric layer | Baseball | Cricket (2026) |
|---|---|---|
| Counting-stats era | Pre-1970s (batting avg, ERA) | Public coverage today (batting avg, bowling avg) |
| First-wave sabermetrics | 1980s (OPS, OBP) | Just emerging (strike-rate-adjusted economy, batter aggression scores) |
| Advanced linear weights | 1990s (wOBA, FIP) | Inside franchise departments only |
| Statcast / tracking era | 2015 onward (exit velocity, spin rate) | Partial, IPL and BBL only, not public |
| Public-facing analytics media | FanGraphs (2005), Baseball Prospectus (1996) | Cricinfo Statsguru and a small set of blogs; no dominant analytics-first outlet |
Cricket’s analytical infrastructure in 2026 sits roughly where baseball was in the early 1990s, with one important asymmetry: cricket has access to better tracking technology and ball-by-ball data than baseball had in the same window, but the public translation layer is much thinner. The bench-level analytics is closer to baseball’s 2015-era Statcast world. The broadcast-layer analytics is closer to a 1985 box score.
What the Big Bash actually does well
The BBL specifically has been an analytical leader within cricket for about a decade. The format itself — T20, 8 franchises, condensed schedule — produces a relatively clean sample that the leagues’ analytics departments have used to build genuinely sophisticated batter, bowler and lineup-construction models.
What the BBL has done well: phase-of-innings strategy is the most analytically defensible in world cricket. Powerplay, middle overs and death overs are treated as distinct strategic problems with different optimal approaches. Batting orders are constructed to optimize phase-specific run rates rather than the traditional “best batters at the top” approach. Bowling matchups are exploited at the over-by-over level. Death-overs bowling specifically is an area where BBL franchises have produced strategies that have influenced T20 cricket globally.
What the BBL still does badly, by the standards of mature analytical sports: the gap between bench-level decisions and broadcast-level explanation is the largest in any major franchise sport. Viewers watching at home have no access to the analytical framework that is driving the decisions on the field. The analytics literacy of the typical broadcast commentator is low. The result is that the same casual fan who has absorbed the basics of OPS or QB pressure rate over the last decade has no equivalent vocabulary for what is happening in cricket.
Where this gets weird
The baseball comparison is useful and it is also incomplete. Three complications matter.
First, cricket has multiple formats — Test, ODI, T20, T10 — and the analytical needs of each are different. Baseball is, analytically, one game. Cricket is functionally four different games sharing equipment. A statistical framework that works for Test cricket (where strike rate matters less and patience matters more) does not transfer cleanly to T20 (where strike rate is the dominant input). The proliferation of formats has split the analytical investment in ways that baseball never had to manage.
Second, cricket’s geographic distribution complicates the analytical community. Baseball’s sabermetric community consolidated around a relatively small group of writers and analysts who could meet in person, share notebooks, and build a common vocabulary. Cricket’s analytical talent is distributed across Mumbai, London, Sydney, Cape Town and Bridgetown, working in different leagues with different priorities. The communication friction is real and structural.
Third, the commercial pressure to translate analytics to broadcast is weaker in cricket than in American sports. The Indian audience that dominates cricket commercially is not the audience that drove the analytics-to-broadcast translation in American baseball and football. The economic incentive for ESPN-style analytics graphics on Big Bash or IPL broadcasts is genuinely lower than the incentive for the same graphics on an NFL or NBA broadcast.
What to actually look for if you want to follow cricket as an analytics reader
- Read the franchise analytics departments’ public outputs. Sydney Sixers, Mumbai Indians and a handful of others have begun publishing post-season analytical notes that translate the bench-level work into public-readable form. These are the most useful single-source analytics reads available.
- Track strike rate phase-adjusted, not raw. A batter’s overall strike rate is the cricket equivalent of batting average — partially useful, frequently misleading. The phase-adjusted version (powerplay, middle, death) is closer to OPS in its predictive value.
- Bowler economy is opponent-adjusted or it is meaningless. The single most-cited bowling number on broadcast is the bowler’s economy rate. Without adjusting for the strength of the batting lineup faced, the number is almost noise. The matchup matrices that franchise departments use are not public, but the principle is recoverable.
- The over-by-over win probability is the cleanest summary metric. When available on broadcasts (rare), it is the closest cricket analog to win probability added in baseball or expected points added in football. It compresses the strategic situation into a single legible number.
The callback
That January night at the SCG, the broadcast graphic and the bench tablet were measuring the same match with twenty years of analytical distance between them. The gap is not going to close on its own. Multiple formats, distributed analytical community, weak commercial pressure on the broadcast layer — these are deep structural reasons, not editorial laziness. But the gap is also not unbridgeable. The bench-layer work exists. The data is open. The vocabulary that would translate it for a public audience has been written in adjacent sports and could be adapted for cricket inside a five-year window if the major leagues decided to fund the translation. Whether that happens depends less on the data and more on whether the audience paying for IPL and BBL broadcast rights eventually wants more than what they already have. The piece on what makes a metric useful covers the methodology that travels. Cricket has the data. It does not, yet, have the FanGraphs. The interesting question is who builds it first.
Cricket ball-by-ball data via Cricsheet; statistical reference via ESPN Cricinfo Statsguru; comparative baseball analytics framework grounded in early FanGraphs methodology.
