This blog has previously discussed the importance of using statistical analysis when making decisions and analyzing information. However, one point that was not made was that both the information used in the analysis and the methodology of the analysis need to focus on generating meaningful conclusions. Without meaningful conclusions the analysis itself is rather worthless and may even lead people to misunderstand the power of statistics.
For example a very simple example of the real descriptive power of statistics can be taken from analysis of possession percentage from soccer (football). Ball possession is often considered one of the more important statistics because it typically describes which team is controlling the flow of the game. However, the description of control is extremely broad. If possession was divided between the offensive and defensive half then statistical analysis is significantly more powerful in generating an understanding of the general behavior of the game. If the new possession statistic demonstrates a large amount of total possession, but most of it in the defensive half then without watching any tape one can reasonably anticipate that such a team uses a long-ball based offense and pushes extra bodies back on defense. Interestingly soccer has already demonstrated the power of deeper statistical analysis where stats are taken of not only of shots on goal, but also how many of those shots are on target illustrating the overall effectiveness of the shots attempted.
One may argue that using such a simplistic example does a disservice to the importance and power of precision statistics, but most people inherently shy away from statistics and would probably not appreciate and/or understand more eloquent and complex examples. Therefore, when dealing with statistical neophytes it is important to introduce precision statistics through a medium that these individuals will care about, thus motivating an attempt to understand driven to better their own knowledge as a means to better enjoy the medium. Basically killing one bird with two stones. Overall not only is it important to include statistics in any deterministic analysis, but the representation of those statistics needs to be appropriate both in accuracy and depth so they tell the important parts of the story.