this is for holding javascript data
Lucas Fidon edited subsubsection_Position_based_metric_When__.tex
almost 8 years ago
Commit id: 24e8655d671320089903217f0567d7be734d3a8d
deletions | additions
diff --git a/subsubsection_Position_based_metric_When__.tex b/subsubsection_Position_based_metric_When__.tex
index ce74022..cf0d9d6 100644
--- a/subsubsection_Position_based_metric_When__.tex
+++ b/subsubsection_Position_based_metric_When__.tex
...
\subsubsection{Position-based metric}
When $C_{penalty}$ is low $(C_{penalty}\leq 1.3)$ almost 75\% of the clusters contain one single player and the other contain 2 or 3 players. On the contrary, when $C_{penalty}$ is high $(C_{penalty}\geq 1.5)$ all the players tend to be gather in a single cluster. While increasing $C_{penalty}$ from 1.3 to 1.5 the players first form largely 2-player clusters.
We can find several interpretations of those 2-player clusters.
\begin{itemize}
\item One player marking a player of the opposite team.
\item Two players of the same team following each other.
\end{itemize}
Goal keepers appear almost always isolated in two different clusters excepting during an action embedding or close to a shot. They are the last players to join the last cluster when $C_{penalty}$ is increased.