There is solid evidence on the cross-training phenomenon, but the training load required to achieve it has yet to be established. The aim of this meta-analysis was to deduce which unilateral strength training load (duration, frequency, intensity, rest and type) would enable the biggest strength increases to be obtained in the inactive contralateral limb. The examined studies were limited to those written in the English language within the Web of Science, PubMed and SPORTDiscus databases. Ten of the 43 eligible studies were included, covering a total of 409 participants. The studies included in the meta-analysis showed a low risk of bias and had an estimated pooled effect size of 0.56 (95% CI from 0.34 to 0.78). Greater effect sizes were observed in lengthy protocols involving fast eccentric exercises using designs of 3 sets of 10 repetitions and a 2-minute rest time. Effect size did not relate to absolute volume, relative intensity, absolute duration and speed of execution. In conclusion, to optimize contralateral strength improvements, cross-training sessions should involve fast eccentric sets with moderate volumes and rest intervals.

• Inter-limb transfer of the strength is more effective in high speed eccentric exercises.

• Muscular endurance training is not advisable to induce contralateral adaptations.

• Cross-education effect may depend more on volume of training than on load.