Cheers evajb, had a look at the source coding but is all gobblygook to me.

One thing I was wondering is re the average odds - I assume that would be for the winning selections, not the average odds for the entire sample.