That 106 Under the Rain at St George's Park: What the Bavuma-Jansen Session Actually Tells Us in the Data

On the morning of Day 3, the scoreboard read 97/5. By lunch it read 216/6. One hundred and nineteen runs for one wicket, across roughly twenty-eight to thirty overs, at a shade above four runs per over. To anyone watching only the scoreboard, this is a fine comeback story. To anyone who keeps the first two days' sheet beside it, this is a strange anomaly. Because across Days 1 and 2, South Africa scored 97 in 43 overs, an over-rate of 2.26. Same batting line-up, roughly the same pitch, same opposition—yet the tempo jumped about one point seven five times. The question is not who scored how many. The question is where the tempo shift came from, and how durable it is.

I have done this work for many years—opening a second sheet beside the scorecard. I did it at Brentford in 2026-17, logging set-piece recoveries across forty-six Championship matches. I kept a sixty-four match baseline at the Russia data desk so that a single match's tactical story could not pass itself off as a tournament truth. The habit is simple: before the narrative arrives, I check the baseline and the control group. The same habit applies here.

Method & Sample box: Format—Test, five-day first-class; match—second Test, bilateral series; venue—St George's Park, Gqeberha; analysis window—the Day 3 morning session; core metrics—run rate, ball age, partnership decomposition, lost overs. Sample—one session, one innings in progress. A warning up front: one session is not a trend.

Context: the match that weather governed

The biggest variable in this Test is not a batter or a bowler—it is the weather. Rain cut into the first two days so heavily that only 43 overs were bowled against a normal 180-over budget. Roughly 137 overs of scheduled play were lost. Without this number, the whole match is misread. Test results are decided by time, and here time was the most spent resource.

St George's Park's character matters too. This coastal venue is generally the slowest and lowest-bouncing of South Africa's main Test grounds, historically favourable to seam, reverse swing and second-innings spin. [Confidence: Medium]—because this venue character is general cricket knowledge, not stated in the report.

A pitch under covers for two days usually retains moisture, so Day 3 morning offers seam movement. But a crucial fact hides here: a pitch on which 119 runs fall for one wicket is not a 'minefield.' Despite two days under covers, batters scored this freely, which means bounce and carry were good enough. If anyone later builds an 'unplayable pitch' narrative, remember this.

The ball-age arithmetic is interesting, because the report's internal numbers reconcile. 43 overs across Days 1-2, plus roughly 28-30 on Day 3 morning, puts ball age at lunch around 71-73 overs. The report says the new ball was due eight overs into the afternoon, i.e. around over 81—matching the standard eighty-over threshold. This internal consistency modestly raises confidence, even without a named source. [Confidence: High]

Core analysis: who actually drove the partnership

Now the real work—decomposing the 106-run stand. 106 runs off 133 balls. Marco Jansen's share: 72 off 71, strike rate about 101.4. The remainder puts Bavuma at roughly 34 off 62, strike rate about 55 (assuming negligible extras). That is, the all-rounder supplied roughly two-thirds of the runs at nearly double the anchor's rate, while the anchor held. [Confidence: Medium—extras or rotation could shift this by a few runs]

Without this decomposition, a false impression forms. Media writes 'the Bavuma-Jansen fightback.' But the data says a Jansen-driven counter-attack with Bavuma as the steady side. These are not the same thing. One is risk-taking acceleration; the other is risk-averse occupation.

Why could Jansen score so fast? This is where ball age enters. The morning session was bowled with a ball roughly 43 to 73 overs old—the classic reverse-swing window. Counter-attacking against a scuffed old ball usually means a boundary-or-bust approach. Jansen's strike rate above 101 fits that profile. But there is a trap, and I will state it plainly: a strike rate of 101.4 must be read against Test benchmarks, not T20. In Tests this is extraordinary, but it is one session of one innings.

Something else happened this session that matters more than the run-scoring: only one wicket fell in thirty overs, and it was a tail-ender. Australia could not break the stand with an old ball for thirty overs. That is the real story of the session. South Africa's run-scoring is the pretty part; why the opposition failed to break through is the important question.

One reasonable answer: Josh Hazlewood's absence. Losing the most metronomic top-of-off bowler forces the remaining quicks into longer spells, especially with a reversing ball. [Confidence: Medium] Caution is due—this is correlation, not causation. Whether the stand would have broken with Hazlewood present, this data cannot say. But the mechanism is real: losing a control bowler changes the workload and roles of the rest.

Contrarian: look at the word 'record'

Now the place where I am most uncomfortable. Headlines and the report call it a 'record-breaking' partnership. Which record? It emerges that the 106-run stand broke South Africa's seventh-wicket record at St George's Park—the previous being 89 against England in 1896. [Confidence: High] That is a 129-year-old venue record, not a national record, and certainly not a world record.

This framing is media overstatement. The word 'record' makes readers think of a major feat, when it is a limited record of one ground, one wicket, one era. I am not diminishing it—breaking a 129-year-old record is no small thing. But journalism's duty is to state the record's scope clearly. Here that was not done.

At the Russia data desk I learned something I still apply to every report: vibes do not survive a second pass. A session's euphoria, an innings' tempo, a record's headline—all are first-pass dazzle. The second pass asks: what is the sample, what is the baseline, where is the control.

The biggest control variable here is rain. 137 overs lost means any conclusion about the result is time-constrained. Day 3 is half gone, on top of two days of rain. In this state the modal outcome is a draw. A definite result needs a rapid end to the SA innings, no further rain, and one side collapsing. My subjective estimate for a result: only 25 to 35 percent. [Confidence: Low—this is a subjective estimate, not data]

Empty stadiums did not erase home advantage; they revealed where it lived. Same lesson here—rain did not erase the match's 'result,' it showed where the result lives: inside time. When time shrinks, skill matters less and controllable variables—weather, declaration, follow-on—matter more.

One hidden fact, inferable though not stated: Mulder's dismissal came on review, meaning the umpire initially gave it not out. [Confidence: Medium] No controversy is reported over this DRS reliance, but it is worth noting that umpiring decisions quietly sit inside this session's story.

Takeaway: what to watch in the next session

The question now is what signal this session gives for the future. First, before betting on the result, do the arithmetic of time, not wickets. 137 overs lost means a draw-heavy equation. Second, if South Africa's innings ends quickly, declaration and follow-on questions arrive. On a slow St George's Park surface, fourth-innings batting is hard, but that depends on how much the pitch breaks—which this data does not say.

Third, Jansen's innings leaves a question: can he regularly score this fast at seven or eight, or was this one-session exploitation of an old ball? Without a sample-size warning, that question cannot be answered. Russia 2026 taught me that every group-stage miracle needs a sample-size warning. The same here.

Finally, a word. Journalism's job is not only to report the score, but the mechanism behind it. In this match the mechanism is rain, ball age, and one bowler's absence—three things combining to make one session look abnormal. Next time someone says 'a great fightback,' ask: in how many overs, off how many balls, with whom absent. The answer is always truer than the story.