T20 Stats Explained: Strike Rate, Average & Impact

Quick answer

Strike rate is runs per hundred balls. Average is runs per dismissal. In a format lasting 120 balls, the scarce resource is balls, not wickets — so strike rate carries far more information than average, and both are close to meaningless without knowing the phase of the innings, the match situation and the sample size. Read balls faced first, then runs, then everything else.

Strike rate, properly

The formula is trivial: runs ÷ balls × 100. The interpretation is not. A strike rate is an average speed over a distance you were not told, achieved on a road whose condition you cannot see. Forty-five from thirty balls reads as 150 whether those thirty balls came in a powerplay against a new ball swinging round corners or in the fifteenth over against an off-spinner with a leg-side field.

Three questions rescue the number every time. When did the innings happen? Scoring at 150 in overs one to six is far harder than in overs sixteen to twenty. What was the required rate? A 150 strike rate chasing 11 an over is losing ground, not winning. How many balls? A strike rate over eight balls is barely a number at all — two boundaries produce a figure that looks like a career-defining statistic.

Why average misleads in T20

Batting average — runs divided by dismissals — was built for cricket where surviving is the primary skill. It maps beautifully onto Test cricket and badly onto T20, because it treats a wicket as a pure cost and gives no credit at all for the runs that could have been scored by taking a risk.

The clearest illustration is the not-out. A batter who finishes 18 not out from 22 balls has his average boosted by an innings that may well have lost his side the game. Meanwhile a finisher who makes 24 from 9 and gets out attempting the last big shot has his average dragged down by exactly the performance he was picked for. Averages therefore systematically flatter anchors and punish finishers — which is why the finishers in India's young T20 batting core are so routinely underrated by casual analysis.

Phases change everything

A T20 innings is three separate games. Any statistic that averages across all three is blending things that are not comparable:

Most good statistical databases let you filter by phase, and doing so is the single biggest upgrade available to a casual reader. A batter with an unremarkable overall strike rate and an outstanding death-overs strike rate is not an unremarkable batter; he is a specialist whose overall number is diluted by a job he does not do.

Economy, average and the bowler's version

Bowlers have the mirror-image problem. Economy rate is runs conceded per over, bowling average is runs per wicket, and strike rate for a bowler is balls per wicket. In T20, economy usually matters more than average, because containing is often the assignment — but only relative to the phase.

A death bowler with an economy of 8.5 may be elite. A middle-overs spinner with the same figure is expensive. Comparing them directly, as league tables invite you to do, produces nonsense. The same applies to wicket counts: a bowler used exclusively in the powerplay will take more top-order wickets simply because that is when top-order batters are at the crease.

What "impact" ratings are trying to do

Impact or match-contribution models exist because everything above is context-blind. They attempt to score a performance by how much it moved the probability of winning: entry point, required rate, quality of bowling faced, phase, and the eventual result. A 20-ball 35 in a collapse can score higher on such a model than a 40-ball 50 on a flat pitch, which matches what anyone watching would say.

Two honest caveats. Different providers build different models and produce different answers for the same innings, so no single rating is authoritative. And every model is fitted to past data, which means it describes rather than forecasts. Use impact ratings to settle "was that innings as good as it felt", not to predict anything.

Sample size and noise

This is the point most confidently ignored on social media. Twenty overs is a short contest, and short contests are dominated by variance. A batter can face six good balls in a row and be dismissed for 2 without playing badly; another can edge three boundaries and finish with 40. Over three matches, these effects swamp any real difference in ability.

The practical rule: treat anything under a full season as provisional, split by phase before drawing conclusions, and be especially wary of stats presented over conveniently chosen windows. "In his last five innings" is almost always a window chosen because it supports the argument being made.

Reading a scorecard line by line

1. Balls faced, before runsIt tells you how much of the innings resource the batter used.
2. Entry pointThe fall-of-wicket column tells you what over and what score he walked in at.
3. Boundary countFours and sixes against balls faced shows whether runs came in bursts or in ones.
4. The bowling figures oppositeAn expensive innings against elite bowling is worth more than a cheap one against filler overs.
5. The match result and required rateThe same score can be decisive or irrelevant depending on the chase.

For the full career picture on any player, ESPNcricinfo carries per-format records and phase filters, and official squads and fixtures come from the BCCI. Applying this method to a specific case, our Vaibhav Suryavanshi records and stats guide deliberately avoids quoting aggregate figures for exactly the reasons set out here.

Statistics are not predictions

Worth stating plainly, because the entire tipster economy depends on blurring it. A statistic is a record of something that already happened. Even a well-constructed model, applied to a T20 international, is producing an expectation that will be wrong a large share of the time — that is what high variance means.

Jai Game publishes no cricket odds, no match predictions and no betting tips of any kind. Our prediction games, responsibly guide explains why a colour or number game is entertainment rather than forecasting, and why the two should never be confused.

Between-overs entertainment

If you like a quick game during the innings break, Jai Game formats are built for short gaps — Wingo resolves every 60 seconds, and the free Demo Wingo table lets you learn it on play money. The match-day instant games guide covers keeping it to the breaks. These rounds are unrelated to the cricket and carry no match markets of any kind.

18+ and responsible play

Real-money play is 18+ and is entertainment, not income. Understanding variance in cricket statistics is genuinely useful preparation for understanding it in games too: a run of results proves nothing, and treating one as a signal is the thinking behind problem gambling. Set a limit in advance; the responsible gaming page has tools and helplines.

Frequently asked questions