The purpose of this research was to analyze and interpret the latent (factor) area of a tennis match. The entities in this research make 128 tennis matches played at the 2007 and 2008 Grand Slams hard court surfaces. The variables were created by use of the official statistics kept by the IBM Software - IBM DB2 Universal Database. The original variables were standardized to the number of sets in a match. A factor analysis under a component model was conducted. The number of factors retained, which was determined by the G-K criterion, explained 83.38 % of the total variance. Five significant factors substantiated the hypothesis established in this paper. The first factor named Match Successfulness is determined by the total number of break points; break points won and received points. The second factor named First Serve Significance is determined by the total number of first serves and winning points after the first serve. The third factor named Serve Speed is determined by the average speed of the serve and the fastest serve. The fourth factor named Net Play is determined by the total net approaches as well as the winning points after net approaches which are directly dependent on the total number of serves. The fifth factor named Play Errors is determined by unforced and double-fault errors. Winning matches differentiate from the lost matches by a smaller number of unforced and double-fault errors; considerably better results of the first serve, maximum serve speed and the number of aces scored, high score of total break points and break points won. The facts that do not differentiate winning matches from the lost ones are: first serves total, first serve throw-in, winning points after the first serve, number of net approaches and winning points after net approaches. The classification results show that with a system of 15 variables it is possible to recognize 96.0% of lost and 96.9% of winning matches. The achieved results indicate that the official match statistics with a modified system of 15 selected variables can properly interpret and predict match successfulness. This enables creators of a match observation system to valorize and enhance it with new indices.