Similar to the system I devised Jamie.... except yours was way too simple, mine probably way too complex.
Between the two extremes is a workable one that would minimise the risks and optimise profitability.
In the statistical analysis there must be a number of factors such as current form, going, distance, Jockey, number of runners, course preference, current fitness, handicap, competition, prize money, potential SP etc etc
In a traditional gambling scenario probability can easily be calculated. Whether it's tossing a coin or picking five cards out of a standard deck.
The probabilities of a specific outcome of a horse race can’t be calculated precisely. These probabilities however can be positively assessed when subjected to an number of statistical and subjective inputs based on previous and similar occurrences.
All those years ago I worked out my statistics by hand with pencil and paper.
My guess is that with todays computing power, access to the relevent input data and a good statistical model there's compelling evidence that a horse racing betting system involving a sound statistics together with optimised staking could potentially produce large profits with a modest 'starting-pot' fairly fast.
Hmmmm, I've almost convinced myself to work on the model
