What Can Hawk-Eye Data Reveal about Serve Performance in Tennis? François Rioult 1 Sami Mecheri 2 Bruno Mantel 2 François Kauffmann 3 Nicolas Benguigui 2 CNRS UMR6072 GREYC - Normandy University, F-14032 Caen, France EA4260 CesamS - Normandy University, F-14032 Caen, France CNRS UMR6139 LMNO - Normandy University, F-14032 Caen, France September 13th, 2015 What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 1/10
What is the Hawk-Eye system for tennis ? ◮ up to 10 cameras around the tennis court ◮ vision processing, triangulation ➥ 40 Hz 3D-trajectory of the ball What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 2/10
XML data <?xml␣version="1.0"␣encoding="UTF-8"␣standalone="no"␣?> <point␣valid="true"> ␣␣<hawkeye_header> ␣␣␣␣<xmldate␣d="Data"/> ␣␣␣␣<server␣p="CLIJSTERS"/> ␣␣␣␣<receiver␣p="HENINHARDENNE"/> ␣␣␣␣<positive␣p="HENINHARDENNE"/> ␣␣␣␣<serve_class␣c="1"/> ␣␣␣␣<scorer␣s="-1"/> ␣␣␣␣<PointDuration␣w="6.26786"/> ␣␣␣␣<score_raw␣s="1␣0"/> ␣␣</hawkeye_header> ␣␣<serve␣name="CLIJSTERS"␣player="1"␣speed="46.46"␣speedEnd="31.87"> ␣␣␣␣<coord␣t="0"␣x="1.47"␣y="-11.89"␣z="2.70"/> ␣␣␣␣<coord␣bounce="true"␣t="0.49025"␣x="-3.05"␣y="6.17"␣z="0.033"/> ␣␣</serve> ␣␣<shot␣speed="31.3742"␣speedEnd="19.8407"> ␣␣␣␣<coord␣t="0.8"␣x="-4.15587"␣y="11.766"␣z="1.06128"/> ␣␣␣␣<coord␣bounce="true"␣t="1.57191"␣x="0.24"␣y="-6.26"␣z="0.033"/> ␣␣</shot> ... </point> What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 3/10
Description Data : ◮ from 2003 to 2008 ◮ 87 grand slam tournaments ◮ 1729 matches ◮ 262,596 points Task : ◮ decode Magnus effect ◮ analyze performance for men and women ◮ focus on serve What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 4/10
About the surface surface 1st serve 2nd INDOORS 72.01 53.03 GRASS 71.19 54.85 HARD 68.34 52.68 CLAY 66.28 52.24 Table : Winning percentage ➥ INDOORS » GRASS » HARD » CLAY ➥ The serve provides a great advantage What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 5/10
About the gender gender 1st serve 2nd women 62.85 49.43 men 71.00 54.18 Table : Winning percentage ➥ still an advantage on the first serve What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 6/10
Serve speed What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 7/10
Distribution of impacts −6 −4 −2 0 2 4 diagonal : deuce court diagonal : deuce court serve : first serve serve : second serve 3 2 1 percentage of total number of points 0 diagonal : advantage court diagonal : advantage court serve : first serve serve : second serve 3 2 1 0 −6 −4 −2 0 2 4 y near impact ➥ Difference between deuce and advantage court What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 8/10
Distribution of the angle −100 0 100 (0.3,1] (0.3,1] first serve second serve 12 10 8 6 4 2 0 (0.2,0.3] (0.2,0.3] first serve second serve 12 10 8 6 4 2 0 (0.1,0.2] (0.1,0.2] first serve second serve percentage 12 10 8 6 4 2 0 (0.05,0.1] (0.05,0.1] first serve second serve 12 10 8 6 4 2 0 (0,0.05] (0,0.05] first serve second serve 12 10 8 6 4 2 0 −100 0 100 ➥ no spin on the first serve phi What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 9/10
Conclusion ◮ painfull data cleaning ◮ a first analysis of big data ◮ confirmation of known results Perspectives : ◮ impact of the number of strokes ◮ detection of routines What can reveal Hawk-Eye? MLSA 2015, PKDD’15, Porto September 13h, 2015 10/10
Recommend
More recommend