| Geldorn said: You do realise that using a ten-game sample is way too low to draw any meaningful statistical conclusions from, don't you? |
He's right. Do this again, but increase the sample size to 30+ - I'd really like to see the results of that (no, seriously!). In fact, if you have the raw data available, we can test for statistical significance in the results with a larger sample. Of course, it's critical that you make sure the samples are randomly selected.
Super World Cup Fighter II: Championship 2010 Edition









