Tolerability of Risk

So I need to post this. To get it all out of my head and move on with my life. Been thinking/obsessing over this too long. (Sorry its about the Grand National or more generally horse fatalities).

The only TWO things that matter are 1) What is the risk? 2) What is acceptable? The first part should be straight forward to obtain from stats. I expect a wide difference of opinion on the second but have some ideas on how to put it in some context.

1) What is the risk? So here it is from the great work done by horsepwr. https://horsepwr.co.uk I will talk about Jump Racing because its what every one is talking about re. GN.

I want to re-calculate the rate to a rate per horse per annum so that it is comparable to numbers I have found later in the thread. So it is 112 fatalities from 7,722 individual horses in one year. That is 1.45% or 145 in 10,000 per annum.

2) What is acceptable? I am going to pivot to humans for a minute. The HSE has this framework for "Tolerability of Risk" they use for creating policy around safety at work etc.

https://www.hse.gov.uk/enforce/assets/docs/r2p2.pdf

They even put some numbers on the border between the Tolerable region and Unacceptable region

So for humans who have risked imposed on them 'in the wider interest of society', we tolerate a rate of 1 in 10 000 per annum. Since horses have the risk imposed on them (they can't sign a disclaimer) this is the number I am using in the following comparisons.

So this is the comparison between 1 in 10 000 versus 145 in 10 000. Is it acceptable? We know the 1 in 10 000 is acceptable. Would you pull the lever to save the human, or let roughly 145 horses perish in this trolley problem? This is where I accept everyone will be different.

By the way, this is just in order to compare. I obviously know that if we cancel Jump Racing a guy doesn't die! Its just a thought experiment to put the numbers discussed in context! Do you value one side over the other or more or less equal?

Also as a guide. In policy making there is a concept of "Value of Life" and it is about £1.7m in the UK. And according to Tattesalls latest Sales Results the average horse went for 8,277 guineas. Which I think is £8,690.

Is it reasonable perhaps one human is worth 195 horses? There will be some that say a human life is infinitely more valuable and there will be some that will love their pet horse more than their best (human) friend. (or is one horse worth 30 humans!!)

I think the conclusion (for me) is if we accept that: * 1 in 10000 is the border of acceptable risk for humans and... * human > horse by orders of magnitude, and... * Racing/Entertainment is in the wider interest of society. Then the 145 in 10 000 probably is acceptable.

You could compare to other animals and the agricultural business. But horses have names, are loved, are someones pet, some are celebrities. If you go down the agricultural route then they are basically expendable like any other animal. ie any level of risk is acceptable.

Also probably would make a difference if that 1 in 10 000 guy was dying live on TV - I don't think even the HSE would find that tolerable.

Perhaps we can argue about the terms "in the wider interest of society" and "in the interest of entertainment". But in my opinion entertainment is in the wider interest of society. We need entertainment. And beyond that the Racing Industry has other benefits to society.

Ultimately I guess the real economics will come down to balancing one equation so that: The value of the customers racing loses because of fatalities = the value of customers racing loses by watering down the challenge in the events.

I guess there are no answers here but maybe just another perspective for someone. We have to talk about risk and accept there is no zero risk. If the HSE can build a framework for Human Life then we can be mature enough to do the same?