# Cluster Theory & Betting Sequences

A common source of frustration for casual gamblers is when it seems that lady luck has sided against them. Every time they bet red on the roulette table, the ball lands on black. They place 3 bets at evens on a Saturday afternoon of football and all 3 of them lose. They understand the probability. They know that all of those bets had a 50% chance of winning so inevitably they feel annoyed when they’ve lost them all.

This is a common occurrence in betting and the assumption that even money chances will win one out of every two times, but it does not always happen in a neat and tidy sequence.

E.g. Win, Lose, Win, Lose

Sometimes you could end up with 4 losses in a row. Sometimes it might work in your favour and you might get 4 win in a row.

Take the example of a coin flip. If you’re truly bored, you can even try this yourself. If you flip a coin 10 times, 100 times and 1000 times you should notice that the number of heads and tails will start to even out.

## Sample Size

The issue that many gamblers have is they lack the patience to cope with losing runs and the variance that inevitably happens. A sample size of a few weeks of bets is actually very small and not an accurate measure of whether you have an edge on a market or not.

Take the example of casino slot machines. You may be aware that they have an RTP (return to player) ratio that ranges somewhere between 92-97%. This doesn’t sound too bad does it! For every 100 pounds, you will end up losing somewhere between 5-8 pounds.

However, there’s still a good chance that you’ll spin 100 times and lose your entire bankroll. Why? Well, the RTP calculation is taken from simulations that run millions of times. 100, 500 or even 1000 spins is quite a small sample within that. So it can explain why the slot machine isn’t actually out to get you but you’re just a victim of variance.

## Odds / Strike Rate

Of course the odds that you bet at will have a direct link to the clusters or streaks. If you are betting on very short odd selections, you are far more likely to encounter an incredible winning run and avoid long losing runs. This is simple probability. However, each loss you do have will inevitably take longer to recover than those at higher odds.

This is often why you see high odds horse racing tipster services often encountering long losing runs and why their winning streaks are so spectacular.

Even with a 60% strike rate, it is very much a possibility that you will see a losing run of up to 10, perhaps even more.

Let’s take the example of the coinflip again.

The coin has just landed on heads for the 2nd time in a row. What is the probability of heads landing for a 3rd time in a row? Technically, the odds of heads landing is always 50%. The 3rd time is not linked to the first two times. However, you can try to exploit probabilities by betting on the most likely outcome and adjusting your stake management. After a loss, you can adjust your staking to cover your previous loss and end up in profit. Just be prepared that this is also a risky strategy and you must have a sufficient bankroll to do this.

In casino games, the edge is in the house’s favour and any profits you find will only be in the short term. However, in sports betting, it is possible to find a genuine edge by examining these clusters more closely.

## Hot Hands

The hot hand phenomenon is a term originated from basketball where a player is on a streak of shooting a high percentage of baskets, usually far more than their statistical average suggests. There is a lot of debate about this. Early research debunked this as a fallacy and without real evidence but recent studies have found that some of the effects cannot be explained purely by chance.

This example can be applied quite easily to other sports and in particular to individual sports or players. A striker in football who has scored in consecutive games, a tennis player who has won a tournament and is in form etc…

With a good knowledge of the participants, you will be able to judge how lucky these streaks have been. If you feel a player/team has been quite lucky in achieving these results to date, then they may be worth opposing if you believe the run is unsustainable and will come to an end in the near future. You may be surprised how often the market and other bettors may not be taking this fully into consideration and may be able to find a value selection using this logic.

I work off a number of racing models for my day's betting, which can include private models and the strategies that we share here via FA Racing and Safe Bets.

Sometimes, this can leave me with 5 to 10 selections across a whole day's racing. Putting all of these to an accumulator would  be both prohibitive based on the number of lines, and the actual chance of it happening.

But what I like to do is then break these up into  smaller groups. Let's say I had 9 selections, i would make 3 each way trebles and doubles (at reduced stakes to my main win stake), back selections 1,2,3 then 4,5,6, and finally 7,8,9 by race time.

Here is one I did the other day, which gave me a 50% profit, but could very easily have been much more.

My other coupon lost, but left me broken even for the day (just on the accas – profit on the singles)

Yet on other days, I've had much bigger collects; like:

I'm also a fan of looking at shorter odds in the right circumstances and we will talk more about this in a future post.

We can see from the chart above that even with a strike rate of 85% (betting odds around 1.12-1.16, we can expect, eventually, to see a losing run up to five in a row.

Conversely, however, we can also expect to see long winning runs and by using a short parlay (rolling up your bets) you can extract more profits.

A couple of good articles here and here go into this in more detail

You can see from the above that I have a simple, but effect strategy set up in my Bot, which carefully selects strong favourites and I roll over my stake and profits for three bets at a time. A pretty decent profit from just 30 bets, with no chasing loses and a steady staking plan.

There is no one solution or metric to making money from betting, it is a number of variables or mixing around like a blender – Odds, Strike Rates, Variance, Clusters and more…

But when you understand the role (and limitation) each plays, you can start to a little more creative in your approach to finding vale

Enjoying this topic? Why not check out:

> How To Scalp The Betfair Markets

> What Causes Betting Odds To Move?