Data Workers · Ligue 1
Ligue 1 penalties: — seen from the taker
In each zone (6 zones: thirds of the width × halves of the height; 3 zones: left, centre, right over the whole height): % of goals among the penalties taken into the zone; an off-target kick counts in the nearest zone. Zones with fewer than 15 kicks are dotted: too few kicks to conclude. Kicks hitting the post or the bar are placed on the frame they hit.
The takers
All penalties
| Player | Foot | Outcome | Team | Opponent | Season | MD | Side | Height |
|---|
Key takeaways
The conclusions update with the filters (seasons, team, player).
All the p-values at a glance
The further left the dot, the less likely the gap is due to chance. Left of the 0.05 line: significant.
The hypotheses tested
Each card starts from a common belief (H₁) and tests it against the hypothesis "there is no effect" (H₀). The p-value is the probability of seeing a gap at least this large if H₀ were true: below 0.05 we reject H₀ (significant), between 0.05 and 0.10 it is a trend, above that not significant. Everything is recomputed with the filters, except "Foot" and "Outcome", ignored on this page.
Right-footers against left-footers
In brackets: the range holding the true rate with 95 % confidence (Wilson interval). Gap: Fisher's exact test.
| Right foot | Left foot | Gap |
|---|
Left-footers − right-footers: the gaps
Each bar is the 95 % range of the gap (Newcombe method). If it touches the zero line, we cannot rule out that there is no difference.
Where do they shoot?
Kicks seen from the taker. "High" = above 1.22 m (half the goal height), kicks over the bar included.
Height changes everything
Outcome of the kicks by height. High up, the keeper almost never saves them, but the target is missed more often.
The law of large numbers
Cumulative success rate, penalty after penalty, in match order. At first the curve jumps around; the more kicks, the steadier it gets and the narrower the range (pale area). With fewer left-footers, their range stays wider.
The takers against chance
One dot = one taker (at least 3 penalties). The funnel shows where an "average" taker (rate of the whole selection) would land 95 times out of 100 by pure chance: it is wide when there are few kicks. A dot outside it truly stands out.
Robustness
Does the result hold if the data change a little?
Season by season
Success rate of each foot by season (click a season to isolate it). Small samples move a lot from one year to the next.