Source note: the JPG poster was recovered and read in full as an image. Every number below is transcribed directly from it. Note the title discrepancy: the filename and the cited title say “kinetic variables,” but the poster’s own printed title is “Anthropometric and kinematic variables and the relationship with ball release speed in female cricket pace bowlers.” No kinetics (forces) were measured. The poster title is the correct one.
Research theme
Same question as Felton’s 2015 abstract, asked by a different group on a different squad: do the kinematic and anthropometric variables that predict ball speed in male pace bowlers also predict it in female pace bowlers? Eleven high-performance female pace bowlers were measured with a low-cost, field-deployable two-dimensional setup — two Apple iPhone 11s at 240 fps, digitised in Kinovea 0.9.5 — plus tape-measure anthropometry. This is a measured, correlational study, and deliberately a cheap one: the method is reproducible by a coach with two phones and a tripod.
What they measured
- Ball release speed (BRS)
- Height, body mass, arm length and leg length
- How far back the bowling arm is at the moment of release (shoulder angle at ball release)
- How straight the front knee is when the front foot lands (front leg knee angle at front foot contact)
- Run-up speed
Findings
- Shoulder angle at ball release was the only kinematic variable with a strong significant correlation with ball release speed: r = 0.95, P = 0.001. A more delayed bowling arm at release went with a faster ball.
- That single variable was the best predictor of BRS, adjusted R² = 0.89 — 89% of the variance in ball speed among these eleven women, from one measurement.
- Arm length: r = 0.61, P = 0.01. Longer arm, faster ball.
- Height: r = 0.76, P = 0.01. Taller, faster ball.
- Run-up speed and knee angle at front foot contact had only a MODERATE, NON-significant (P > 0.05) correlation with female BRS. The poster is explicit that “this relationship was not as strong as what is seen in previous research conducted with male pace bowlers.”
- The authors’ stated conclusion: female pace bowlers “are more likely to use their upper body to generate high BRS’s,” and coaches “should refrain from extrapolating all findings from previous research conducted with male pace bowlers and applying it to female pace bowlers, as it may be restricting female bowling performance.”
- Recommended next steps stated on the poster: three-dimensional analysis, larger samples, and investigation of the relationship between delayed arm circumduction and trunk flexion in maximising shoulder angle at release.
What a coach should look for on video
Cue 1 — How far back the bowling arm still is at release (the one that matters here)
- Camera view + frame: Side-on, camera perpendicular to the bowler’s plane of motion (the poster used two phones at 240 fps, 6 m either side of the crease). Pause on the exact frame the ball leaves the hand.
- What “good” looks like: The bowling arm still trailing further behind the line of the upper trunk at release — the arm “arrives late.” This is the strongest relationship in the whole female-bowling literature: r = 0.95, adjusted R² = 0.89.
- What the fault looks like: The arm has already come through and is level with or ahead of the trunk at the instant of release; the bowler looks like she is pushing the ball out in front rather than whipping it over.
- Why it matters: On this evidence a single frame explains ~89% of the speed differences between these bowlers. It is by some distance the highest-yield thing to film.
Cue 2 — Height and arm length (selection, not coaching)
- Camera view + frame: Not a video cue — a tape measure.
- What “good” looks like: Taller bowlers with longer arms bowled faster (r = 0.76 and r = 0.61).
- Why it matters: The poster frames this as talent identification, not technique. It is a reason to look at tall, long-armed athletes, not a reason to change anyone’s action.
Cue 3 — Run-up speed: handle with care in women
- Camera view + frame: Side-on, last few strides.
- What this poster found: Only a moderate, non-significant relationship with ball speed in these eleven women — notably weaker than in men.
- Why it matters: This is the poster’s central warning. Driving a female bowler to run in harder, on the strength of male data, may not buy speed and may cost control. But see the CONTRADICTION below — the journal version of this same study reports the opposite.
Cue 4 — Front knee at front foot contact
- What this poster found: Moderate, non-significant. No cue supported here. See the 2023 journal paper, which recovers a significant relationship once height is controlled for.
Caveats and limits
- n = 11. Very small. An r of 0.95 on n = 11 is impressive but fragile, and adjusted R² = 0.89 from a single predictor selected out of a handful in a sample this size will be optimistic out of sample.
- Two-dimensional analysis from two iPhones and Kinovea. Cheap and ecologically valid, but a 2D shoulder angle is sensitive to how squarely the camera is aligned with the bowler’s plane of motion, and cannot resolve out-of-plane rotation — which, per Felton et al. (2019), is precisely where female bowlers differ most.
- Conference abstract/poster. No table of means, no confidence intervals, no regression coefficients on the poster itself.
- No kinetics were measured despite the cited title.
- Correlational. Nothing here establishes that lengthening arm delay would make a given bowler faster.
Relationship to other Felton work
- This is the preliminary version of Lyons 2023 in South African Journal of Sports Medicine, same eleven bowlers, same setup. The journal paper adds a full correlation table, partial correlations controlling for height and arm length, and formal stepwise regression. Prefer the 2023 paper for numbers.
- Independently reproduces the “delayed arm at release” finding that Worthington et al. (2013) reported in men.
- Directly builds on Felton et al. (2019) (cited on the poster as the basis for expecting male/female divergence).
- CONTRADICTION: run-up speed. This poster reports run-up speed as having only a moderate, non-significant (P > 0.05) correlation with female ball release speed. The 2023 journal paper on the same eleven bowlers reports run-up velocity r = 0.75, p = 0.01 — significant. The two accounts of the same dataset disagree on whether run-up speed matters for women. The 2023 paper is the peer-reviewed version and its correlation table is explicit, so it should be preferred; but the reversal is large enough that neither should be quoted as settled. (In the 2023 paper the relationship does wash out to r = 0.41, p = 0.28 once height is controlled for — which may be what the poster was reflecting.)
- CONTRADICTION: front knee at front foot contact. Non-significant here; r = 0.68, p = 0.04 (significant) in the 2023 journal version once height is a covariate. Same eleven bowlers.
- TENSION: the poster’s conclusion (“female bowlers are more likely to use their upper body,” don’t extrapolate from men) is drawn from finding that run-up and front knee don’t matter. But the one variable that did matter — a delayed bowling arm — is exactly a male finding, from Worthington et al. (2013). The evidence for “coach women differently” here is weaker than the poster’s rhetoric.