In absence of further information [i.e. zero knowledge about any of the participants], the only reasonable probability estimate to come up with is 1/N where N is the total number of horses in the race. I have no idea why I went with the specific range [.08, .12] though as it doesn't accompany the edge cases, hahah.

[.06, .13] would be a better range, since 1/8 < 0.13 and 1/15 > 0.06.

Great post and thanks for this Substack.

Can you help explain how in a race with a range of 8-15 horses we should expect probabilities in the range of [0.08, .12]?

Great question!

In absence of further information [i.e. zero knowledge about any of the participants], the only reasonable probability estimate to come up with is 1/N where N is the total number of horses in the race. I have no idea why I went with the specific range [.08, .12] though as it doesn't accompany the edge cases, hahah.

[.06, .13] would be a better range, since 1/8 < 0.13 and 1/15 > 0.06.

Thank you for the explanation Bettor. Understand that now.