Background: The objective of this work was to define and determine the effectiveness of defensive actions applied in judo fighting.Material/Methods: The study was based on video and DVD recordings of fights during Polish Championship finals in the years 2005-2008. Actions of female and male contestants participating in the total of 56 fights were analyzed (to an accuracy of 10-second periods of time). All actions: attacks, counter-attacks and defense without counter-attack were recorded and their effectiveness, preparatory actions, breaks and referees' decisions were analyzed. Altogether 69 fights of female and male contestants were subject to analysis.According to the author's classification of defensive actions, 12 types of defense without counter-attack were defined as follows: (1) hand block, (2) hip block, (3) maneuvering around, (4) twist onto belly, (5) hand and hip block, (6) hand block and maneuvering around, (7) stepping aside, (8) separation from grasp, (9) leaving the mat, (10) leg entanglement, (11) bridge, (12) return to tachi-waza (escape from ne-waza).Results: The types of defense most often applied according to the study - hand block and twist onto belly - were the least effective (86% and 66% accordingly).Ashi-waza i sutemi-waza were throws most often applied as counter-attack. Sutemi-waza (44%) throws were most effectively applied.Leg and hand throws were the most frequent and most effective counter-attack throws. O-goshi throw appeared to be counter-effective most often - not less than every fourth one was successfully counter-attacked.Conclusions: High effectiveness of counter-attacks (29%) should be the tip for coaches and contestants to carry out professional trainings, which follows (and not the other way round as it used to be) the rules set forth by professor Jigoro Kano (among others: "give up in order to win"). Moreover, trainings of defensive skills should be modified since defense actions most often applied by Judoka players (hand block and twist onto belly) proved to be the least effective.