There are a few things, only a few, in this life that can be reasonably counted on. I know that's not a new perspective, and the ones we usually hear about are death and taxes.

But racing also has a couple, and one of these is to do with "trends." or "runs".

I don't know how many times I've said that if only I knew when a good run was about to commence, I might be sitting back sunning myself on some northern beach, and simply making the appropriate phone calls at the appropriate times.


Unfortunately that isn't how it goes. Nevertheless, there are hot and cold streaks. Ask any punter or investor (and there is a difference). If they know what they are talking about, they will acknowledge that some days are diamonds and some days are stone. In fact, some days are utterly brilliant whilst others are black as jet.

I have also written, and found a lot of support for the idea, that most of the year is spent somewhere in between these two extremes. You win most of what you are likely to win (in any given year) on about four or five days of that year. Sometimes if you're lucky this will extend to another three or four. That is generally it. I have had experiences where my "good luck" has extended right through a carnival, and I have had the reverse.

A close friend of mine won a lot of money just last year in September and sought my advice. As near as I can recall it, I told him to allocate a small but significant percentage (perhaps 20 per cent) to his betting practice over the remainder of the Sydney and Melbourne spring carnivals.

Further, I advised him to place the remainder in a fixed deposit for at least the three-month period which would get through the entire carnival era. He didn't have another success of any significant nature throughout the carnivals, and he's still telling me that without this advice, he would probably have given his winnings back to the machines.

It wasn't that I was so clever about this, it was that I've been there, I've done it, and I am crucially aware that no matter how careful and clever your successful analyses are, they still need an element of luck. Luck exists in racing. Bad and good luck. It"s why you lose races on protests and it's how your horse gains a miracle run on the rails to come from eighth to first in the last 50 metres.

Naturally I wouldn't argue that the jockey in that previous example was anything short of a genius, but, if the opening hadn't come, that same jockey would have been setting there on your horse as it passed the post in eighth place.

Genius needs a bit of luck. You read about the origins of DNA and life, and you read about Darwin and company, and there is an element of luck there amongst all the amazing processes. Mathematically, it comes back to probability. The “likelihood" principle. And there again, if you could get a handle on the kind of probability attached to your selecting, you'd probably be invincible.

The trouble is, as one individual, you'd probably need to live about 5,000 years and to make approximately 3,650,000 individual investments to get anything like a reasonable sample from which to work. How did I arrive at that figure? Well, I reasoned on two bets per day over the 5,000 years. It may not be enough. You'd then need to live another I,000 years to make use of the information. Certainly 50 or 60 years of your adult life is not going to be enough, as it is only one sample, and anyway, let's be frank, you are going to be heading towards 80 by then, and I leave the rest to your imagination.

So, if there is one thing I'm absolutely certain about in my mind, it's that things come in runs. Good and bad runs. If only we could tap into them! But going back to something I said before, the important thing is to come to terms with one absolute fact: when you do your tallying at the end of your betting year, you will find that on something like 35 to 40 of the 52 weeks, not much happened financially. You cut even. You made a little. You lost a little.

It's those other 10 or 12 weeks or so, that either make the profit or do the damage. They are when your Hot Streaks are likely to occur. When I was a lot younger than I am now, I recall reading some very optimistic systems to do with recognising a streak.

Invariably, the recommendation was to raise your stake, usually after a winning bet, and to reduce your stake after a losing bet. The inadvisability of this became obvious to me, as I became far more familiar with the racing scene and with its ups and downs. There is a simple truth to do with this kind of thinking.

The truth is that your previous bet has absolutely nothing to do with your next bet.

It's true to say that you may have an idea of trends, because you have devised a system based on a very reasonable number of sample selections. You might for example be able to say with some confidence that the longest losing streak you can experience with your system is eight.

As a matter of fact, that happens to be the longest losing sequence for The Whisper. Based on what we know of a good racing system, the strong likelihood is that the figure 8 will probably be pretty close to the mark for this very clever plan over an extended period. Remember what I said above, about the very few certain things on earth, and this is why it is always wise to operate with a battery of methods, and not ever rely on a single one. But 8 is a very encouraging number indeed.

I put it to you, however, that the staking plan for The Whisper does not say "increase your bet if you ever have eight losers in a row". Arguably, you could put up a case for an increased stake after that point, as your expectations would be predicting an imminent winner or several. But that particular plan is designed for long-term security. It recognises, as all good plans do, that the long term is the only term that matters.

That's a tricky one, because when you know that you are dealing with expectations around 35 per cent or so win strike rate, the temptation is there to raise the stakes if the strike rate has slipped somewhat.

But what we always have to take into account with something like this is that, OK, in an anomalous year the strike rate might drop to (say) 30. And it might sit there for a year or two. The probability prediction is that, based on what we already have, a long term 35 or so is about right. What will not happen is that, if the win strike rate is running around 30, then suddenly you will get an unbelievable string of winners to bring things back to "normal".

Again, I point out to you, long-term is the only term.

In closing, I offer a positive suggestion as to how you might identify "hot streaks'”. I'm going to suggest one that will cause some readers to nod their heads in agreement, whilst others will probably be incredulous.

This isn't new. It also is not able to be analysed, and yet those who have "been there" know about it.

It’s a gut feeling. You just know. It happens.

And what's the biggest question? The biggest question is invariably "when will this stop?" It's what we call an imponderable. Oh, the number of times that I have been on a roll, and where nothing that I did could go wrong! Pick them to win and they win. Box a trifecta and up it pops. Identify two horses for the daily double and they both oblige. And then you start asking "Hey, how long has this been going on?!"

What you should be asking is ""How long is this likely to keep going on?"

And the advice? As soon as the feeling leaves you, jump off the bus! Don't wait for it to be spelt out.

By The Optimist

PRACTICAL PUNTING – SEPTEMBER 2005