In the second part of the New Horizons series, let's begin to get to grips with probability distributions.  A 'probability distribution' is simply a typical arrangement of data.  Over time, statistians have found that certain distributions of data occur often in the real world.  Our question today is: which is these distributions best represents number of home goals scored in a game?  Using a distribution leads to theoretical odds for particular outcomes; that, in turn, can help identify 'value' and promising bets.

In this video, we look at the normal distribution which you might know as a 'bell curve'.  It's a symmetrical distribution and is one of the most commonly used in statistical analysis.  But, is it useful for forecasting number of home goals?  We learn how to 'set up' the distribution in Excel using the mean and standard deviation parameters, then 'sample' the distribution by inputting a random number between zero and one.  After some adjustments, we can then create a chart and evaluate its suitability by comparing to the real-world phenomenon: number of home goals.  How good is the 'fit'?  It gets quite theoretical in places but hang in there!  It should all make sense in the end ...

Chris also explains why football trading is guerilla warfare!

Tools And Techniques

- NORM.DIST formula
- NORM.INV formula
- Mean Average
- Standard Deviation
- RAND formula
- Sampling A Distribution
- Goodness Of Fit
- COUNTIF
- Line Chart