You hear about a tipping service on the grapevine. You check it out.

The results look good.

You continue to monitor the results. They continue to look good.

You decide to go to the website and join the service. Then, just before you hit the ’Payment’ button, you change your mind.

You decide to give it one more week before you join.

You continue to monitor the results. They continue to look good.

You start to count the money that you would have made had you joined earlier.

You go to the website and, finally, that’s it, you hit the ‘Payment’ button.

There’s no going back now. You have paid for a month’s subscription.

The next day, you receive your first three tips and you place your bets.

You begin wondering how you will spend the profits.

Maybe, if you can build up your betting bank, you can afford that new car that you have been promising yourself for the past year. A silver one would look good parked in your driveway. It would sure take the smile off that sanctimonious bugger who lives next door.

Remember that wry smile that he gave you a couple of months ago when he parked his new car in his driveway for the first time?

You watch the first race. Your horse is right up there.

It looks as if you are on your way to getting the new car.

Then, suddenly, what’s this?

An outsider storms through from the pack and wins by two lengths.

Your horse comes third.

The next race is a total disaster.

Right from the start, your horse doesn’t even look as though it is going to win.

In fact, it trails in a distant last.

The third race looks better.

Your horse is clear by two lengths when suddenly it pulls up lame and trails in a distant last.

That’s the end of the first day.

All of the three tips lost.

On day two, you receive two tips. One wins at odds of 2/1 and the other manages to scrape home by a short head at evens.

Although you made a small profit today, you have lost more than 15% of your betting bank over the two days. Not good.

The rest of the week isn’t too good either. Most of the tips lose.

At the end of the week, you have lost more than 35% of your betting bank.

‘Oh well’, you say, ‘anyone can have a bad week’.

Week two turns out to be not much better than week one.

You have a couple of winners - but at short odds. You have now lost 45% of your betting bank.

Week three turns out to be an unmitigated disaster - not one winner!

You have now lost over 70% of your betting bank.

By the end of week four, most of your betting bank has gone up in smoke, together with the subscription fee.

At the end of the month, you receive an email from the tipping service informing you that your subscription fee is due.

You send the tipping service an email informing them of exactly what you think about their service.

You receive an apology.

The service blames the past month’s poor results on the unusual amount of rain which fell.

What will next month’s excuse be, leaves on the bloody track?

Does this sound vaguely familiar?

A similar scenario could also have been written for any newly implemented system.

I wish that I had £1 for every occasion that I heard the above. I would now be driving around in that new car - in silver!

Have you ever wondered why this happens and why so frequently?

Surely every tipster can’t be a rip-off merchant?

Surely, every set of past proofed results can’t have been ‘manufactured’?

I decided to apply some logic to this issue in order to see what I came up with.

I went right back to basics and asked myself some questions:

If I decided to join a tipping service, when would I join?

If I had created a new selection system, when would I implement it?

During a losing period?

No!

I would implement a new system or join a tipping service during one of their winning periods.

Why?

Well, who in their right minds would implement a new system or start to purchase tips from a tipping service during a losing period? It is contrary to human nature. It just wouldn’t make sense.

Or would it?

Again, let’s go right back to basics.

There is no such thing as the perfect system. All systems, therefore, have a strike rate of less than 100%.

If past records for a system exist, its strike rate can be calculated.

If its strike rate can be calculated, the odds of encountering a losing run of a given length can be calculated.

Let’s take this example:

Suppose that the system’s long-term past strike rate is 70%.

There is a 75% chance that the system will encounter 4 losing bets in a row.

There is a 50% chance that the system will encounter 6 losing bets in a row.

There is a 25% chance that the system will encounter 7 losing bets in a row.

There is less than a 1% chance that the system will encounter 9 losing bets in a row.

This is exactly why systems go through a bad patch.

If something can happen, eventually, it will.

Therefore, even though there is < 1% chance of encountering 9 losing bets in a row, given enough time, it will happen.

OK. That’s losing runs explained.

Now let’s go back to our initial questions regarding the point at which a new system should be implemented or a subscription to a tipping service purchased:

Generally, new systems are implemented and subscriptions to tipping services are purchased during the winning period.

Why? Because we don’t think that it’s a sensible idea to do either during a losing period.

Yet, we know that the system is going to lose, eventually.

At this point, I am going to upset one or two statisticians and mathematicians. For those of you who are of this ilk, please give me the benefit of the doubt for a few moments.

Let me ask you this: When is a losing run most likely to occur?

A. Immediately after a long losing run

B. Immediately after a long winning run

The statisticians and mathematicians amongst you will probably say both or neither.

Why is this?

Because, if horse racing is a random event, then past results have no bearing whatsoever on future results.

The fact that the last bet lost has no effect on the result of the next race.

However, consider this:

Horse racing isn’t random.

Some horses are much better than others. As such, they win far more often.

If horse racing was a random event, then this wouldn’t happen.

True, if a race consists of horses that are of equal ability, then, to some extent, the race does have some features which could be described as random.

However, the vast majority of races are not like this and certain horses do have more chance of winning than do others.

As such, provided that a system has exhibited a given strike rate over a sufficiently long period of time, it is reasonable to assume that it will maintain its past strike rate in the future, allowing for a reasonable margin of error.

Now, armed with these assumptions, let us ask the question again.

When is a losing run most likely to occur?

A. After a long losing run.

B. After a long winning run.

In my opinion, the answer is B (after a long winning run), and here’s my reasoning:

Let us consider a system which, over the past few years, has had a strike rate of 80%.

The system has had a very good run over the past month and the strike rate, during the period, has increased steadily and now stands at 90%.

If this short-term trend continues, the strike rate will begin to approach 100%.

However, given the past performance of the system, this is unlikely.

What is more likely to happen is that the future short-term strike rate of the system will fall well below 80% in order that its long-term strike rate of 80% will be maintained.

In other words, in the immediate future, it is likely that there will be more losing bets than normal.

The longer that the winning trend continues, the more severe the losing run is likely to be.

We may not be in a position to determine precisely when the winning run will end, but the longer the good luck continues, the more likely it is that the run of good luck will cease.

Just when a system has established a winning run is the time when most new systems are implemented and tipping service subscriptions purchased.

This is precisely the point at which a system is at its most vulnerable to a run of bad luck.

This explains why, when a new system is implemented or a subscription to a tipping serviced is purchased, people become quickly disillusioned since there is a tendency for the system to encounter a losing period.

I made this very mistake when I began proofing one of my new systems.

I waited until a winning run was well established before I made my first submission to the proofing service.

Guess what? On the very first day, I had 3 losing bets!

What a start, and, what a great advert for a system new to the market!

Well, that explains why it is, in my opinion, that newly implemented systems and tipping services start off by losing.