How Big Should Your Betfair Trading Bank Be? Why Survival Matters More Than Profit

One of the first questions people ask when they start trading or betting on Betfair is surprisingly simple:

How big should my trading bank be?

£100? £500? £1,000? £10,000?

It sounds like a straightforward question. In reality, there is no sensible answer until you know something much more important: what is the bank expected to survive?

A strategy with a 70% strike rate and average odds of 1.50 has a completely different risk profile from one operating at average odds of 12.00. A method placing ten bets a month is different from one placing hundreds. And a strategy that has experienced a longest losing sequence of eight bets needs to be treated very differently from one capable of losing 50, 60 or 70 times in succession.

Your trading bank should therefore not begin with how much you want to make.

It should begin with risk.

A £1,000 Bank Doesn't Tell Me Very Much

Suppose somebody tells you they have a £1,000 betting bank.

Is that conservative?

Perhaps.

But if they're staking £100 per bet, they have only ten units. A relatively ordinary losing sequence could destroy the bank.

If they're staking £10 per bet, they have 100 units.

At £2 per bet, they have 500 units.

The cash amount hasn't changed. The risk has.

That's why thinking about a trading bank purely in pounds can be misleading. For research purposes, I find it much more useful to think in units.

If one normal stake represents one unit, the question becomes:

How many units does this strategy need to survive its normal variance?

That is a much better question.

Your Worst Losing Run Is Not Necessarily Your Worst Losing Run

This is where historical data becomes important.

Imagine you've tested a strategy across 2,000 bets and its longest losing sequence was 20.

It would be tempting to build a bank around that number.

But what exactly have you established?

Only that the longest losing sequence observed so far was 20.

Bet 2,001 doesn't know that.

Neither does bet 5,000.

As the sample grows, you create more opportunities to encounter unusual sequences. A strategy that has historically survived 20 consecutive losses is not somehow protected from experiencing 25 or 30 in the future.

This is one reason I am cautious about extremely aggressive staking plans built around historical maximum drawdowns. Historical data can help us understand risk. It cannot place a ceiling on future risk.

Your bank therefore needs room not only for what you've already seen, but for the possibility of something worse.

Dog Pound Taught Me This the Uncomfortable Way

My own Dog Pound greyhound research provides a useful example.

The strategy has produced long-term profit, but it has also experienced a sequence of 79 consecutive losing bets.

Think about what that means psychologically as well as financially.

Loss.

Loss.

Loss.

And eventually another 76 of them.

A trader using a tiny bank or aggressively increasing stakes could have been wiped out long before the eventual winner arrived.

The interesting thing about a large trading bank is therefore not that it maximises your returns.

Quite often it does the opposite.

It reduces the percentage of your capital being risked and makes the numbers look considerably less exciting.

But it buys you something valuable:

time.

Time for variance to play out. Time to discover whether the hypothesis actually works. Time to collect enough evidence to make sensible decisions.

The objective isn't simply to have enough money to place tomorrow's bet.

It is to survive long enough to learn.

Bank Management Cannot Rescue a Bad Strategy

There is another important distinction.

A bigger bank does not turn a losing strategy into a profitable one.

Neither does staking 0.5% instead of 1%.

Bank management determines how your capital interacts with variance. It doesn't manufacture an edge that isn't there.

This is why I become uncomfortable when staking systems are presented as the solution to poor results. Martingale-style recovery systems, increasing stakes after losses or manipulating stakes to hit a daily target can make the equity curve look clever for a while.

The underlying bets haven't improved.

You've simply changed the amount of money exposed to them.

If the underlying expectation is negative, clever staking can change the journey towards the loss. It cannot magically change the underlying mathematics.

Drawdown Matters More Than the Headline Profit

When analysing a strategy, profit naturally attracts attention.

A strategy made 200 units.

Great.

But I immediately want to know what happened on the journey.

• What was the maximum drawdown?

• What was the longest losing sequence?

• How volatile were the returns?

• What were the average odds?

• How many bets produced the result?

• Was the staking level throughout?

Those questions tell me far more about the bank required to trade the strategy than the headline profit figure.

Two strategies could both make 200 units and require completely different banks because the path to those 200 units was completely different.

That path matters.

Start With Evidence, Then Decide the Bank

This is also why I don't think there is a universal answer to questions such as “Should I have a 100-point bank?”

Maybe.

Maybe not.

First understand what you're trading.

Collect enough data to begin understanding its behaviour. Examine losing sequences and drawdowns. Look at the odds distribution. Understand how frequently it bets. Consider whether your historical sample is remotely large enough to capture the variance you could encounter.

Then add a margin of safety.

Only after that should you translate units into pounds.

If your research suggests you want 200 units available and you're comfortable with a £5 unit, that's a £1,000 bank.

If losing £1,000 would materially affect your finances, then the answer isn't necessarily to reduce the bank.

It may be to reduce the unit.

That distinction is important.

The Bank Is Part of the Experiment

At Trade Carefully, I increasingly think of a trading bank as research infrastructure.

It isn't a pot of money whose sole purpose is to become a bigger pot of money.

It is the capital that allows an idea to remain alive long enough to generate meaningful evidence.

That changes the mindset.

Instead of asking:

“How quickly can I grow this bank?”

I prefer:

“Is this bank robust enough to survive what this strategy might throw at it?”

The second question is less exciting.

It is also considerably more useful.

Because in sports trading, surviving variance is what gives you the opportunity to discover whether you had an edge in the first place.

Research first. Evidence before conviction. Trade carefully.