All horses are assumed to be backed at flat stakes of 10 units for a win, and 10 units each way (10 units total) for a place.

The result profit / loss shows the amount you would have won / lost backing these horses above.

Our each way calculations are done as shown in the table below, with number of runners down the left, and position across the top. The % corresponds to the % of the SP paid for gaining an each way place.

All horses are assumed to be either backed at flat stakes of 10 units for a win or layed for a 10 unit profit if the horse does not win the race.

The result profit / loss shows the amount you would have won / lost selecting these horses with 2% commission factored into any winnings.

The figures do not account for a dead heat and in these scenarios a horse finishing 1st would return a full profit (if backing) or return a full liability loss (if laying).

The calculated p-value will range from 50% to 0%. The smaller the p-value, the less likely the record will have arisen by chance, in other words the more likely something else like skill may be the primary cause. Anything more than p-value=1% is not considered by statisticians to be evidence of something other than chance going on.

A p-value of 1% means that your record will arise by chance 1 in 100 times. This estimation is fairly robust for variable staking provided the stake sizes do not vary too much. This test is less reliable for records shorter than 100 bets. This test also assumes that you are betting to market efficient (or 100% value) prices, i.e. random pick selection will break even over the long term.

It also assumes there is no back fitting taking place and you are generating outputs for selections in a pre defined system, rather than testing random historical outputs. For typical fixed odds match bet markets, market leading (best) prices will closely approximate 100% value prices.

A p-score below 5% gives moderate evidence against the thesis that the yield is a result of chance alone. A p-score below 1% could be considered a strong evidence against this thesis, and under 0.1% a very strong one.