some image


Assessing Risk at the Horse Races

Oscar Foulkes January 12, 2014 Horse Racing, Risk No comments
The writing of the likes of Nate Silver, Nassim Taleb and Daniel Kahneman (read my post on this here) encourages us to have a better awareness and understanding of the risks we need to weigh up in making choices (choice could be a proxy for forecast/prediction). We also need to be aware of the bias that our own psychology brings to the decision-making process.

Under many conditions, I am not risk-averse. I’m not sure that this therefore makes me a risk-taker, but I’m certainly willing to have a bet.

I went to the races for the first time at the age of 15. I must have got very lucky, because I won a Place Accumulator that paid out about R1500, which was a lot of money 30 years ago. By the time I got to my final year at university I could win – or lose – a similar amount in an afternoon. Overall, I was a winner, but I didn’t feel comfortable in this space. Over the course of the next few years, I had the occasional punt (generally very successfully), but when I got into wine I stopped completely. It would frustrate Andrea, on the odd occasion when we went racing, that I resolutely refused to do anything other than watch.

Since the middle of 2012, I’ve been doing a lot of work within the racing industry, which has drawn me back into the fold. In parallel, I’ve been peripherally involved in my mother’s breeding operation. And then, since November, Dish Food & Social has been operating the hospitality area at Kenilworth racecourse for the Racing Association. It being part of the family business, I have been very involved, which has given me yet another reason to go racing.

Since then, I’ve taken about eight PAs. Three have been outright losses, another lost money because the payout was less than my stake, and the rest have been profitable. The nett position is a nice surplus in relation to my initial stake (but not of a scale that suggests I should consider giving up my day job!).

So, with the work of Silver, Kahneman and Taleb in mind, what have I observed about the field generally, and myself specifically?

In keeping with a contest that incorporates not just known unknowns, but also unknown unknowns, forecasting is tricky. Anyone who tells you they can tip more than a couple of winners on a card is delusional. The best anyone can do is to hope that the wins outweigh the losses.

I know of several people for whom betting on horses is their sole income, but for the rest of us it’s entertainment. The sport offers the opportunity of exercising skill and strategy, which makes it a different proposition to casino games. Maybe I’m just a nerd, but I enjoy the intellectual process. And, even when that runs out, occasionally there are great athletes engaging in titanic contests.

The first rule of trading, whether it’s stocks, futures, options or forex, is to practice money management. What this means is that you need to preserve capital (i.e. don’t run out of money), by limiting the amount that can be lost on any one position.

Accept that no-one knows what is going to happen. We can guess – and some people may make better assessments more often – but we don’t know. Even the insiders don’t know for certain. Top jockeys who freelance can pretty much pick their rides, and even they are unplaced about half the time.

If you’re losing, don’t chase it. You started the day wanting to have fun while staking a small amount of money; just stick with the programme. Making desperate bets to recover losses just leads to bigger – less manageable – losses.

My choice of bet remains the PA, which requires bettors to select placed runners in seven designated races (also known as legs). It’s a permutation, which means that you don’t have to be 100% right in every leg. A good approximation is often all that’s necessary. It also works as a permutation on the payout side, so if you pick more than one placed horse in several legs the payout multiplies. The outsize returns are necessary to cover the times that you don’t win.

Of course, if the favourites are placed all the way through the day, the payout will be small, which is what happened the time that I won the PA, but still ended up losing money.

I like to put my PA on before I even leave for the races. At that point I have a maximum known loss (R100-ish is plenty, or one can put it on at less than 100%), which leaves me free to enjoy the day. If it’s all going according to plan I have seven races worth of entertainment, with some drinks money at the end of it.

After winning (or is that a case of getting lucky?) I need to try to get back to a state of humility. It doesn’t take the genius of Nate Silver, or research of Daniel Kahneman, to tell you that cockiness is not a great state of mind for making good forecasts.

Everyone knows of a gambler who has lost everything. I have hesitated to use the word here, because it implies someone who bets without considering all the risks. Even if we are not betting, or professionally engaged as forecasters, we make probability (i.e. risk) assessments every day.

I think my betting hobby has made me realise that I have at times been guilty of making over-confident predictions, not just on the racecourse.

[Deep breath as I realise what a big admission this is … and wife with predisposition to irrational anxiety chortling somewhere in background]