Bowery Yard
Bowery Yard AnalysisJune 2026~7 min read

The Oval: England's Best Chasing Ground

At England's seven Test grounds since 2002, batting averages collapse from the 1st innings to the 4th by as much as 14 runs per wicket. At The Oval, they go up. The toss math at Kennington is not the same problem as at Lord's.

+0.75
Oval 4th-inn avg above its 1st
36.03
Oval 4th-innings batting avg, top of England
+14.09
Trent Bridge 1st-to-4th collapse
#1 of 6
Oval rank for 4th-inn batting

The folk wisdom of Test cricket says the pitch breaks down. The Cricsheet record says that depends entirely on which pitch you mean.

01The fracture that didn't happen

The Oval is the only major English Test ground that finishes higher than it starts.

Across every Test played at one of England's seven Test grounds since 2002 - 136 matches in the Cricsheet ball-by-ball archive - the batting average in the 4th innings is, on average, several runs per wicket lower than the batting average in the 1st innings of the same match. The further the match goes, the harder the pitch is to bat on. That is the folk wisdom, and at four of the six ranked grounds the folk wisdom is right.

At The Oval, it is not. The 1st-innings batting average across 23 Tests at Kennington is 35.28. The 4th-innings batting average across the 18 of those Tests that reached a 4th innings is 36.03. Batting outcomes do not deteriorate across the match, unlike at most other English grounds. Chase outcomes track the same story: Oval chasers win 38.9% of the time, well above Lord's at 23.5% and Trent Bridge at 28.6%. Section 05 walks through that record in detail.

Drop in batting average from the 1st innings to the 4th, by English Test ground. Trent Bridge collapses by 14 runs; The Oval moves -0.75 (the 4th innings is actually higher than the 1st).
Fig. 1Runs per wicket lost between the 1st and 4th innings, by ground. Trent Bridge is the steepest collapse in the country. The Oval and Headingley both end a Test batting higher than they began; The Oval does so on the largest 4th-innings sample of the two.
02What 'average' means here

Runs per wicket, by which innings of the match.

One number per ground per innings: runs scored divided by wickets lost in that innings number, summed across every Test at the ground. Innings are numbered by position in the match, not by calendar day.

Not every Test reaches a 4th innings, so the 4th-innings sample is smaller everywhere. Per-ground counts: 34 at Lord's, 18 at The Oval, 15 at Edgbaston, 14 at Trent Bridge, 13 at Headingley, 10 at Old Trafford. The Rose Bowl has hosted one Test in the window and is excluded.

Batting average by innings number of the Test, by ground. Five English grounds slope downward from the 1st to the 4th innings. The Oval line dips at the 3rd innings and recovers to its highest point at the 4th.
Fig. 2Batting average by innings number, all six ranked grounds. The Oval and Headingley are the two lines that end higher than they begin; The Oval starts at the pack mean and stays there, while Headingley starts below it and climbs.
03The first innings looks similar everywhere

On Day One, the major English grounds look broadly similar.

Begin a Test at any of these six grounds and you are batting on roughly the same pitch. The 1st-innings averages run from 29.66 at Headingley to 37.12 at Old Trafford - a band of 7 runs per wicket across the six ranked grounds, half of which is the gap between any two adjacent grounds. The Oval sits in the middle of the pack at 35.28. On Day One, nothing about Kennington is unusual.

1st-innings batting average by English Test ground. Old Trafford 37.12, Edgbaston 35.68, Trent Bridge 35.49, The Oval 35.28, Lord's 35.14, Headingley 29.66.
Fig. 31st-innings batting averages across the six ranked English Test grounds sit inside a 7-run band. The Oval is unremarkable on the opening day.
04By the 4th innings, the pack splits

Six pitches break. One holds.

Stay at the same six grounds through to the 4th innings and the band falls apart. Old Trafford's 1st-innings average of 37.12 drops to 27.43. Lord's drops from 35.14 to 26.24. Trent Bridge collapses from 35.49 to 21.40, the deepest fall in the country at 14.09 runs per wicket. Edgbaston drops a more modest 3.30. Headingley actually climbs, from a low 29.66 to a fourth-innings 33.49. Headingley's improvement is real but rests on 13 fourth-innings Tests, the smallest sample among the five grounds that cleared that threshold.

The Oval, alone among the country's heavy-use Test grounds, ends higher than it began. The 4th-innings average of 36.03 sits 0.75 runs per wicket above the 1st-innings number and 14.6 runs above the 4th-innings number at Trent Bridge - a bigger gap, ground to ground, than separates Trent Bridge from Edgbaston on Day One.

4th-innings batting average by English Test ground. The Oval 36.03, Headingley 33.49, Edgbaston 32.38, Old Trafford 27.43, Lord's 26.24, Trent Bridge 21.40.
Fig. 4By the closing innings of a Test, the band has split. The Oval bats highest in the country. Trent Bridge bats lowest.
On Day One, the major English grounds start from roughly the same place. By the closing innings, only Kennington holds.
On the chase at Kennington
05The chase, scoreboard-tested

Chases at The Oval succeed where chases at Lord's don't.

Batting average is one read. Did the chasing side actually win is another. Of the 18 Tests at The Oval that reached a 4th innings, the side batting last won 7, lost 8, and drew 3 - a chase success rate of 38.9%. Lord's, with the largest sample in the country at 34 4th innings, has seen the chasing side win 8 of them: a 23.5% chase rate. Trent Bridge sits between, at 28.6%.

Headingley and Edgbaston, the two grounds where the 4th-innings average runs at or above the 1st, post the country's top chase rates - 53.8% and 53.3%. The Oval ranks third of six. The order of finish on chase rate is almost the same order of finish on 4th-innings batting average, with one small reshuffle in the middle. Two separate measures of the same record agree on which grounds reward the side batting last. With 18 fourth-innings Tests at The Oval against 34 at Lord's, those percentages carry real uncertainty - the direction is consistent across both measures, but the precise rates should be read as estimates.

Share of 4th-innings batting sides that successfully chased, by ground. Headingley 53.8%, Edgbaston 53.3%, The Oval 38.9%, Old Trafford 30.0%, Trent Bridge 28.6%, Lord's 23.5%.
Fig. 5Share of 4th innings in which the chasing side won, by ground. The Oval ranks third in the country and well above the two London-and-Midlands grounds at the bottom.
06Over by over, the runs come

The chase scores faster at The Oval than the England average, almost block by block.

Bucket every delivery of a 4th innings into ten-over blocks. At The Oval, the chase-side run rate sits above the rest of England's 4th-innings average across nearly every block of the chase. Across the other five ranked grounds, the same metric runs about half a run per over lower for almost the entire 4th innings.

Runs per over by ten-over bucket of the 4th innings. The Oval line sits above the England average through nearly every block.
Fig. 6Runs per over by over-bucket of the 4th innings. The Oval line stays above the rest of England across nearly every block of the chase.

What changes at The Oval is the rate at which the runs come, not the way the wickets fall when they do.

07What the toss math says

The Kennington toss is not the same problem.

A Test captain's toss decision is an expected-value problem. The inputs are the side's relative batting and bowling strengths, the weather, the green tinge or the lack of one, and a number every captain knows by feel: how much does this pitch punish the side batting last. At six English Test grounds, that number is large and positive. At Trent Bridge it is 14.09 runs per wicket. At Lord's it is 8.90. At Old Trafford it is 9.69.

At The Oval it is -0.75. A captain reaching for the same heuristic at Kennington is reaching for a number with the wrong sign. Bat first to put pressure on the Day-Five chase is a sentence that makes sense at four other English grounds. At the four grounds where outcomes deteriorate, the fall is sharp; at The Oval it never arrives. Kennington holds.

Share this analysis

Method

How this was built

Match data from the Cricsheet ball-by-ball Test archive (tests_json.zip), every Test played at one of the seven England grounds - The Oval, Lord's, Trent Bridge, Edgbaston, Headingley, Old Trafford, and the Rose Bowl in Southampton - between 2002 and 2025. 136 matches in total. Each ball is parsed for runs and dismissal type. Per (venue, innings number) we sum runs and wickets across every match at the ground and report runs per wicket as the batting average for that innings of the match. Wides and no-balls are excluded from the balls-faced denominator used for runs per over; extras still count for the batting side's run tally. The 4th-innings sample is naturally smaller than the 1st-innings sample because some Tests finish inside three innings; the smaller sample sizes are listed inline with every chart. The Rose Bowl has hosted one Test in the window and is shown only in the source numbers, not in any ranking. Some grounds have fewer than 15 Tests that reached a 4th innings in the window, so the cross-ground rankings should be read as directional rather than as precise estimates. A 90% bootstrap CI for The Oval's 1st-to-4th delta runs from -8.95 to +6.58 runs per wicket, straddling zero. The comparable interval for Trent Bridge runs from +5.80 to +21.57, comfortably positive. The right reading of The Oval is that batting outcomes hold rather than deteriorate; the right reading of Trent Bridge is that they fall. The analysis compares venue outcomes and does not adjust for the mix of team strength, opposition quality, or era at each ground. Charts produced with matplotlib and the Bowery Yard brand palette; the pull, aggregation, chart, and audit scripts live at analysis/oval/ in the source repository, with FACTS.md as the source ledger and audit.py as the pre-publish gate.

Have a dataset you'd like analyzed like this? Talk to us