Add reference to paper explaining geometric mean for normalized results

Author: tobami

example/templates/home.html

@@ -199,7 +199,7 @@ <h3>How fast is PyPy?</h3>
         <h3>How has PyPy performance evolved over time?</h3>
 
         <div id="cpythonplot2" style="width:600px;height:400px;margin: 0 auto;"></div>
-        <p style="font-size:smaller;">Plot 2: Speedup compared to CPython, using the inverse of the geometric average of normalized times, out of <span id="num_of_benchs"></span> benchmarks.</p>
+        <p style="font-size:smaller;">Plot 2: Speedup compared to CPython, using the inverse of the geometric average of normalized times, out of <span id="num_of_benchs"></span> benchmarks (see <a href="http://dl.acm.org/citation.cfm?id=5673" title="How not to lie with statistics: the correct way to summarize benchmark results">paper</a> on why the geometric mean is better for normalized results).</p>
     </div>
 </div>
 {% endblock body %}