Cricket Batting Average: How It's Calculated, What It Means, and the All-Time Highest Records
Batting average = total runs ÷ number of times dismissed (not innings). A not-out innings does not count as a dismissal, so very good finishers (who often end not out) have inflated averages. The highest Test batting average is Don Bradman's 99.94 — he needed just 4 runs in his final innings to average 100, but was bowled for 0 by Eric Hollies. Modern averages: Steve Smith (58+), Virat Kohli (57+), Kane Williamson (54+). Average benchmarks: 50+ is elite, 40+ is very good, 35+ is competitive.
Batting average is the primary statistic for measuring a batter's quality in cricket. It calculates how many runs a batter scores, on average, per dismissal — not per innings. This distinction is critical: if a batter scores 50 and is not out, they are not counted as having been dismissed, so their average rises. A batter who is regularly not out at the end of a innings (a 'finisher' or 'tail-ender') may have an artificially elevated average.
The Formula
Batting average = total runs scored ÷ number of dismissals. Example: 1,000 runs in 25 innings, 20 times dismissed, 5 not outs = average of 1,000 ÷ 20 = 50.00. If those 5 not outs were each large scores (e.g., 60*, 80*, 45*, 35*, 55*), the 'true' average might be considered lower because those contributions were never completed. This is why the 'not out' adjustment is both cricket's most nuanced stat and its most contested.
Bradman's 99.94
Don Bradman played 80 innings, scored 6,996 runs, and was dismissed 70 times — a career average of 99.94. In his final Test innings (The Oval, 1948), needing just 4 runs to finish with a career average of exactly 100, he was bowled second ball by Eric Hollies of England for 0. The average of 99.94 is the most famous number in cricket. The second-highest Test average in history is Graeme Pollock (South Africa) at 60.97. Steve Smith has averaged over 60 at various points in his career.
Frequently asked questions
What would Don Bradman's batting average be in modern cricket, and is the 99.94 a fair comparison across eras?
Bradman's 99.94 cross-era comparison and modern equivalence. Don Bradman's average of 99.94 is almost universally accepted as the greatest statistical achievement in all of sport — not just cricket. The statistical gap between Bradman and the next-best Test batter is approximately 39 runs per dismissal. In any other sport where statistics are comparable, the best performer tends to be 10–20% better than the second-best; Bradman was approximately 65% better than his contemporaries. Cross-era objections and answers: (1) Objection: 'Bradman played in easier conditions (no helmets, less pace bowling).' Answer: Bradman played against Larwood, Tate, and the English bowling attack of the 1930s on uncovered pitches (which dried unpredictably and became extremely dangerous — far more difficult than covered modern pitches). He also played during the 'Bodyline' series (1932–33) where the specific strategy was to bowl fast deliveries at the batter's body — Bradman averaged 56.57 in that series, still extraordinary. (2) Objection: 'Spin bowling was less developed.' Answer: Bradman's average on the turning, uncovered pitches of the 1930s suggests he was adapting to conditions harder than a modern flat Test pitch. (3) Objection: 'Fewer international teams.' Answer: Bradman played against England, Australia, South Africa (all major nations of the time). What modern analysis says: cricket statisticians have applied 'era adjustment' models (similar to baseball's WAR) to Bradman's numbers. Even with the most aggressive modern adjustments, Bradman's equivalent modern average is estimated at 65–75 — still higher than any other batter has ever achieved. The statistical consensus: Bradman's 99.94 is not a product of era inflation — it reflects genuine outlier performance across multiple contexts (Australia, England, different conditions, against the best bowlers of his time). His average in England alone (87.98 in 12 Tests) is higher than any other player's career average.