Cricket vs Baseball Statistics: A Full Comparison

Baseball taught America to read a sport through its numbers. That habit is the best equipment an American fan can bring to cricket, and also the thing most likely to mislead them, because cricket uses several of the same words for completely different quantities. A cricket batting average is not a baseball batting average. They are not even measured in the same units.

What follows is the honest map. Every cricket figure quoted here is computed by us from CricketTaken's own datasets: the full Indian Premier League record across every season, the Big Bash League, and the complete Ashes archive. Where we are estimating or where the comparison stops working, we say so.

The one difference that governs all the others

Baseball measures a batter against opportunities. Cricket measures a batter against dismissals.

In baseball, an at-bat ends. You get another one in two innings. Over a season a hitter takes six hundred cracks at it and the statistics describe the average quality of those cracks. Failure is cheap and constant; a hitter who fails seventy percent of the time is going to Cooperstown.

In cricket a batter is dismissed once and their innings is finished. There is no second at-bat. A batter can face two balls or three hundred, and the only thing that ends the innings is getting out. So cricket's central batting number does not ask how often you succeed. It asks how much you produce before you fail.

Everything downstream follows from that. If you hold onto one idea from this page, hold onto this one, and if the underlying mechanics are still fuzzy, cricket explained for baseball fans lays out the game itself before the arithmetic.

Batting average: same words, different math

Baseball's AVG is hits divided by at-bats. It lives between .200 and .350 and is a rate bounded by one.

Cricket's batting average is total runs divided by times dismissed. It is unbounded and it is measured in runs. A batting average of 40 means that on average, this player produces 40 runs before getting out.

That is closer in spirit to a baseball hitter's runs created per out than to AVG. It is a production-per-failure number, and the correct instinct when you see 45 next to a name is not "he gets a hit 45 percent of the time" but "he is worth about 45 runs every time he walks out there."

Not-outs complicate it. If a batter finishes an innings undefeated, those runs count in the numerator but the innings does not count in the denominator, which is why a specialist finisher who is frequently stranded unbeaten can carry an inflated average. Baseball has no equivalent distortion.

What a good average looks like

From our own IPL dataset, pooling the 233 batters with at least 20 innings across the competition's history, the collective batting average comes out at 27.93. In our Ashes dataset, pooling the 60 batters with 40 or more innings, the collective average is 40.18.

Same statistic. Same sport. The Test number is more than 40 percent higher, because in the long format a batter is not obliged to take risks and can occupy the crease for hours. A batting average is only meaningful against a format, and comparing a T20 average to a Test average is like comparing a reliever's ERA to a starter's without adjusting for anything.

Strike rate: cricket's slugging percentage

Baseball's slugging percentage asks how much damage a hitter does per opportunity. Cricket's strike rate asks the same question against a different denominator: runs scored per 100 balls faced.

A strike rate of 150 means 150 runs per 100 deliveries, or a run and a half every ball. In baseball terms it is a rate of production per pitch seen, which is not a stat baseball keeps but is conceptually the family slugging belongs to. High is good. Very high is dangerous, both for the fielding side and, often, for the batter.

The comparison is imperfect in one specific way. Slugging weights outcomes by base value on contact. Strike rate divides by every ball faced including the ones you leave alone. So a cricket strike rate blends two things baseball separates: how hard you hit it, and how often you choose to hit it at all.

The average-versus-strike-rate tension, in our numbers

This is where cricket's two headline batting stats pull against each other, and our IPL data shows it cleanly.

Batter	IPL innings	Average	Strike rate
Virat Kohli	275	40.42	134.80
Andre Russell	114	28.20	174.18
Heinrich Klaasen	60	42.08	166.72
AB de Villiers	170	39.71	151.69
Chris Gayle	141	39.72	148.96

Figures computed by CricketTaken from our IPL dataset covering every season of the competition.

Kohli and Russell are the argument in two rows. Kohli produces 40 runs per dismissal at a pace of roughly 1.35 runs a ball. Russell produces 28 per dismissal at nearly 1.75. Neither is better in the abstract. Kohli is the batter you want if the innings needs building; Russell is the batter you want with four overs left and 60 required. Baseball has exactly this argument about a high-OBP table-setter versus a low-average slugger, and cricket resolves it the same way, by context and role.

Klaasen's row is the one that stops you. An average above 42 and a strike rate above 166 is the combination that almost never coexists, and it is the T20 equivalent of hitting .320 with forty home runs. The full treatment of this tension is in batting average vs strike rate.

How T20 batting actually scores its runs

Our IPL pool produces one figure that reframes the whole format for a baseball fan.

Across those 233 batters, 61.1 percent of all runs scored came from boundaries. Fours and sixes. Meanwhile only 17.8 percent of balls faced produced a boundary at all.

Read those two numbers together. Fewer than one ball in five goes to the rope, and those balls account for nearly two-thirds of the scoring. The other four-fifths of deliveries produce the remaining third, a single run at a time, by running.

That is a three-true-outcomes distribution in everything but name. T20 batting is a boundary-hunting exercise punctuated by low-value contact, and the rest of the innings is a holding pattern between attempts. Our Big Bash dataset, pooling 87 batters with 500 or more balls faced, puts boundary runs at 54.5 percent with a collective strike rate of 131.13, slightly lower on both counts than the IPL, which reflects larger Australian grounds and generally more bowler-friendly conditions.

Economy rate is the real ERA

Here is the cleanest one-to-one mapping in the two sports.

ERA is earned runs allowed per nine innings, a rate of damage per unit of work. Economy rate is runs conceded per over, and an over is six deliveries. Both are rate stats, both measure runs surrendered per fixed unit of bowling or pitching, and in both the lower number is the better number.

If you want a single sentence to carry into your first cricket broadcast: economy rate is ERA, scaled to a six-ball unit instead of a nine-inning one.

From our IPL dataset, pooling the 81 bowlers with 600 or more deliveries, the collective economy rate is 8.06 runs per over. From our Ashes dataset, pooling the 61 bowlers with 50 or more wickets, it is 2.59.

The gap is roughly threefold and it is entirely a function of format. Twenty overs a side forces bowlers to be attacked. Five days does not. Nobody bowls at 2.59 in a T20 and nobody survives a career conceding 8 an over in Tests.

What good looks like

Bowler	Competition	Economy	Note
Sunil Narine	IPL	6.80	Over 200 IPL wickets at that rate
Muttiah Muralitharan	IPL	6.52	Lowest in our qualifying IPL pool
Shane Warne	Ashes	2.52	Across 195 Ashes wickets
Glenn McGrath	Ashes	2.70	Across 157 Ashes wickets

Figures computed by CricketTaken from our IPL and Ashes datasets.

Narine's number is the one to sit with. Conceding 6.80 an over in a competition where the collective rate is 8.06 means giving away roughly a run and a quarter less every over than the field, across a full career. Over a four-over spell that is five runs saved, which in T20 is frequently the entire margin. It is the cricket equivalent of a reliever sitting a full run of ERA below league average for a decade, and it is exactly why he is used the way he is used.

Bowling average, and why low is good in both

Cricket's bowling average is runs conceded per wicket taken. Low is good, for the same reason low is good in ERA: you are dividing a bad thing by a good thing.

This is the stat that trips up newcomers most often, because the same word means opposite things depending on which side of the ball you are on. A batting average of 45 is excellent. A bowling average of 45 is career-ending. There is no getting around it; you simply have to know which column you are reading. The scorecard conventions that keep them straight are covered in how to read a cricket scorecard.

There is also a bowling strike rate, and confusingly it works the opposite way to a batting strike rate. A bowler's strike rate is balls delivered per wicket taken, so low is good. Our IPL bowling pool sits at exactly 20.0 balls per wicket, meaning the average qualifying IPL bowler takes a wicket roughly every three and a third overs.

The finding that surprised us

Compare the two bowling averages from our own data:

Dataset	Pooled economy	Pooled bowling average
IPL (81 bowlers, 600+ balls)	8.06	26.87
Big Bash (85 bowlers, 500+ balls)	7.91	25.31
Ashes (61 bowlers, 50+ wickets)	2.59	26.13

Computed by CricketTaken from our IPL, Big Bash and Ashes datasets.

The economy rates differ by a factor of three. The bowling averages are almost identical, all clustered in the mid-twenties.

That is not a coincidence and it explains why cricket keeps both statistics rather than collapsing them into one. Economy measures how expensive you are per unit of time. Average measures how much a wicket costs you. In Tests, bowlers are cheap by the over but need many overs per wicket. In T20, bowlers are expensive by the over but take wickets far more frequently, because batters are attacking every ball. The two effects cancel, and the price of a wicket turns out to be roughly stable across formats separated by five days of playing time. Baseball has no analogous pair, which is why ERA has to carry the whole load and why FIP and WHIP exist to fill the gaps. The full breakdown of the three bowling metrics is in bowling economy, average and strike rate.

The century, and what it is actually worth

A century is 100 or more runs by one batter in a single innings. It is the achievement cricket organizes careers around, and there is no baseball statistic shaped like it, because baseball has no individual performance that both accumulates continuously and can be ended at any moment by a single mistake.

Our IPL data gives it a price. Across the 233 qualifying batters in our pool there are 120 centuries and 1,901 half-centuries. That works out to roughly one century every 131 innings, and one score of fifty or more every 7.8 innings.

One in 131 is the number to hold. In a competition of that scale, a T20 century is a genuinely rare event, and a batter who has collected several is doing something unusual. Kohli's nine hundreds and 68 fifties across 275 innings in our dataset is the outlier career, not the standard.

The closest baseball analogue in feel, not in mechanism, is the complete game shutout or the four-hit night: rare enough that it makes the highlight reel, common enough that great players accumulate them. It is not a no-hitter and it is not a perfect game, because it does not require the opposition to cooperate.

Test centuries are a different currency again. In our Ashes dataset, Don Bradman scored 19 hundreds in 63 innings against England alone, at an Ashes average of 89.78 and a strike rate of 58.06. Those numbers do not belong to the same distribution as anyone else's and never have. He is the reason cricket fans get impatient with arguments about the greatest athlete in any sport.

The full conversion table

Baseball statistic	Cricket equivalent	Direction	Fidelity of the comparison
Batting average (AVG)	Nothing exact	Higher better	Cricket's "average" is a different quantity entirely
Runs created per out	Batting average	Higher better	The closest true match
Slugging percentage	Strike rate	Higher better	Good, though denominators differ
On-base percentage	Nothing	—	No walks, so no concept
ERA	Economy rate	Lower better	Very close, different unit of work
WHIP	Economy rate	Lower better	Overlapping, not identical
Innings pitched	Overs bowled	Higher usually better	Close
Strikeouts	Wickets	Higher better	Loose; a wicket is worth far more
K/9	Bowling strike rate	Inverted	Cricket counts balls per wicket, not wickets per unit
Home run	Six	Higher better	Close in feel, worth six runs not a variable number
Ground-rule double	Four	—	Loose
Complete game	Batting through an innings	—	Loose but the right idea
No-hitter	Nothing	—	No true equivalent
Save	Nothing	—	No closer role, no substitutions

Where the comparison genuinely breaks

Four places, and they are not small.

There is no walk, so there is no on-base percentage. Cricket has no ball-four and no forced advance. A batter can leave fifty deliveries alone and be rewarded with precisely nothing except survival. The single most important development in baseball analytics over thirty years, the elevation of on-base skill, has no cricket counterpart at all.

Wickets are not strikeouts. A strikeout removes one baserunner-in-waiting. A wicket removes a player from the match permanently and moves the batting side one step closer to being finished entirely. The correct baseball analogy for taking a wicket is not a strikeout; it is closer to a double play in leverage and closer to an injury in permanence.

There is no substitution, so there is no bullpen math. Every bowling statistic in cricket is generated by one of eleven players who also has to field and possibly bat. There is no specialist deployed for one batter. Bowlers in limited-overs cricket are capped at a fixed share of the innings, which means the best bowler cannot simply bowl more.

Park factors are wilder than anything in baseball. Baseball fields vary within known bounds. Cricket grounds vary in boundary distance, and the playing surface itself is a prepared strip of soil that changes character over the course of a match and can be built to favor spin or seam. A pitch is a variable in a way no ballpark is. Adjusting cricket statistics for conditions is a harder problem than park-adjusting baseball ones, and the sport does it far less rigorously.

Where to take this next

The vocabulary is the last barrier. The cricket statistics glossary defines everything you will meet on a broadcast, and the rules explained for Americans covers the situations those numbers describe.

Then apply it. The Indian Premier League is where the T20 numbers in this article come from and the highest standard in the format. The Big Bash League supplies the second T20 sample used above and is worth comparing against directly. For international cricket, the T20 World Cup and the 50-over World Cup hubs collect the tournament records, and the full players database carries the career figures behind every name mentioned here. When the numbers start making sense on their own, the quiz will tell you, and the rest of the cricket in the USA hub covers everything else.

Sources and method

Every cricket figure in this article is computed by CricketTaken from our own datasets and is identified as such at the point of use. The IPL pool covers all 233 batters with at least 20 innings and all 81 bowlers with at least 600 deliveries across every season in our IPL record. The Big Bash pool covers 87 batters with 500 or more balls faced and 85 bowlers with 500 or more deliveries. The Ashes pool covers 60 batters with 40 or more innings and 61 bowlers with 50 or more wickets across the full series archive. Pooled rates are calculated from summed raw totals rather than by averaging individual players' rates, so they are weighted by volume. Baseball statistical definitions follow standard published conventions; where a baseball league-wide figure is referenced it is described qualitatively rather than quoted, because those values move year to year. Historical ball-by-ball cricket records used in building our datasets derive in part from Cricsheet (cricsheet.org), licensed ODC-BY 1.0.