Win-stay lose-shift strategy in formation changes in football

EPJ Data Science

Table 6 Results of the GLMM analysis with all independent variables for the J-League data when a year was regarded as a season

Variable	Coefficient	SE	p -value
\(\mathrm{Win}_{t-1}\)	−0.306	0.123	0.013
\(\mathrm{ Win}_{t-2}\)	−0.223	0.111	0.045
\(\mathrm{Loss}_{t-1}\)	0.406	0.119	<0.001
\(\mathrm{Loss}_{t-2}\)	0.145	0.126	0.249
\(\mathrm{Win}_{t-1} \times \mathrm{Win}_{t-2}\)	0.021	0.180	0.909
\(\mathrm{Loss}_{t-1} \times\mathrm{ Loss}_{t-2}\)	0.003	0.162	0.985
Home	−0.062	0.072	0.388
Strength	−0.327	0.219	0.137

\(\mathrm{ Win}_{t-i}\) (i = 1,2) is the binary variable representing whether or not the team has won the (t − i)th match (0: no win, 1: win). Likewise for \(\mathrm{Loss}_{t-i}\) (i = 1,2) (0: no loss, 1: loss). Home is equal to 0 for an away game and 1 for a home game. Strength is equal to the fraction of the matches that the team has won in a season. SE: standard error.