Evidence-Based Betting: The Lies Bookmakers Tell

Open almost any betting app and you'll see the same trick: a player's last five games highlighted in bold.

"Scored in 4 of the last 5." "25+ disposals in the last 5 matches." "Over this line in 5 straight games."

It looks like powerful evidence. It feels recent, relevant, and convincing.

But it's also deeply misleading.

Why bookmakers push "last 5 games"

Humans naturally overweight recent events. If something just happened, we assume it will keep happening.

Psychologists call this recency bias, where people give greater importance to recent information over long-term evidence (Kahneman, 2011).

Bookmakers know this.

Showing only recent form nudges bettors toward patterns that often don't last. A player might have hit a stat line five times in a row due to matchups, injuries, weather, or random variation, not because it's their true performance level.

Small samples create strong emotions, not reliable predictions. Research consistently shows people draw overly confident conclusions from small sample sizes, what Kahneman calls the law of small numbers (Tversky & Kahneman, 1971).

And emotional betting is profitable for bookmakers.

How bookmakers actually set prices

Here's the part they don't advertise.

Bookmakers don't price markets using just the last five games.

Professional odds-making relies on massive historical datasets and predictive modelling. Sports analytics research shows that pricing models draw on long-term performance patterns, matchup data, and contextual factors far beyond recent streaks (Stern, 1991; Hvattum & Arntzen, 2010).

In reality, pricing models consider factors such as:

• Full career performance
• Multi-season trends
• Home vs away splits
• Performance at specific venues
• Matchup history against opponents
• Team structure and role changes
• Injuries and lineup changes
• Weather and match conditions
• And thousands of historical comparisons

Recent form matters, but it's just one small piece of the puzzle.

The true odds are built on large evidence sets, not five-game snapshots.

When recent form really matters

There are exceptions.

If a superstar player is suddenly injured or suspended, teammates' roles can change dramatically. A midfielder might suddenly attend more centre bounces, or a forward might become the primary target.

In these situations, historical data without that player becomes important, and sometimes there isn't much of it.

Markets can shift quickly when team structure changes.

But even then, professional models still rely on as much evidence as possible, not just short-term streaks.

How sharp bettors approach betting

The best bettors don't follow promotional trends.

They build their own prices using evidence from multiple datasets, deciding which factors matter most for each bet.

They might weigh:

• Career averages
• Recent seasons
• Venue performance
• Opposition history
• Home or away splits
• Current role in the team
• Recent form

Different bets require different weightings.

The goal isn't to guess.

The goal is to measure.

Making evidence-based betting simple

At StatChecker.app, we built tools we wished existed when researching bets ourselves.

Instead of spending 20 minutes searching stats, you type in an AFL player bet and instantly see:

• Up to 15 years of career data
• Post-COVID performance, default from 2021 onward
• Home vs away splits
• Venue history
• Opposition performance
• Recent form
• And more game filters

Then you decide what matters.

With a simple weighting slider, you can choose how much importance to give each data set, instantly creating your own weighted prediction.

Because betting decisions should be based on evidence, not marketing tricks.

The takeaway

If a bookmaker shows you only the last five games, ask yourself:

What aren't they showing?

The difference between entertainment betting and evidence-based betting is simple.

One reacts to headlines. The other studies the full picture.

And over time, evidence wins.

References

Hvattum, L. M., & Arntzen, H. (2010). Using ELO ratings for match result prediction in association football. International Journal of Forecasting.

Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux.

Stern, H. S. (1991). On the probability of winning a football game. The American Statistician.

Tversky, A., & Kahneman, D. (1971). Belief in the law of small numbers. Psychological Bulletin.

OpenAI. (2026). ChatGPT 5.2 large language model. OpenAI.

Anthropic. (2026). Claude Opus 4.5 language model. Anthropic.