How to Keep Your Betting Winnings

Strategy 1: Stop Betting on Overpriced Home Favourites 

For decades, home advantage in English football was a reliable edge, with home sides winning around 60-65% of matches. However, that era is over. In 2024-25, just 40% of Premier League games were won by the home side, the second-lowest rate in the entire Premier League era (only the COVID season, played behind closed doors, was lower).

Away teams are now averaging more possession (49.6%), more touches in the opposition box (25.5 per game), and their highest-ever expected goals figure of 1.4 xG per game, all records. Bookmakers have been slow to fully adjust. 

When a mid-table home side is priced at 4/6 or shorter to beat a top-six visitor, say Everton hosting Liverpool, I now check whether the market is pricing in outdated home-advantage assumptions. 

Liverpool won 2-1 at Everton in April 2026, which was unsurprising seeing as their away form that season was among the best in the league. Backing the draw or away win at 11/4 was statistically better supported than the short-priced home win the market was offering.

If a home team is odds-on and their xGA (expected goals against) at home is above 1.3 per game, I don't back them. I either pass or look at the away side.

Strategy 2: Use xG Regression to Bet on Underperformers About to Correct

Expected goals (xG) is the single most useful metric available to recreational bettors right now. Yet most punters still price teams by league position alone.

Crystal Palace this season have scored just 36 goals from an xG of 51.3, underperforming by over 16 goals. Statistically, that is an enormous gap. 

A team that good at creating chances doesn't sustain that level of poor finishing indefinitely. Regression to the mean is one of the most reliable forces in football analytics. When a team is statistically due a scoring run, their odds to score first, win, or land BTTS are still priced as if they're the team that's been misfiring.

I've been using this specifically with Crystal Palace's matches over the past six weeks. When they were priced at 19/10 to beat Wolves at Selhurst Park in February 2026, their xG in recent games was around 1.8, while Wolves were hitting 0.7. The match odds didn't reflect that gap, and Palace ended up winning 1-0. 

Before any match, pull up both teams' xG on Understat or FBref, compare it to their actual goals, and find the biggest gap, these sides are most likely to cause an upset.

Strategy 3: Exploit BTTS in High-Variance Fixtures

Both Teams to Score (BTTS) has landed in 55.2% of Premier League matches this season. Across the whole league, that's roughly an even-money proposition, which means BTTS "Yes" at 8/11 or shorter is usually poor value. But team-specific data tells a different story.

Exploit BTTS in High-Variance Fixtures

Manchester United has a BTTS rate of 74% in their matches this season, 23 out of 31 games. At the other end, Nottingham Forest have the lowest BTTS rate in the division at just 41%. The gap between those two figures is enormous, and online betting platforms price United's BTTS games only modestly higher than average.

When Manchester United hosted Brentford in April 2026, a side themselves averaging 1.4 goals per game away, BTTS "Yes" was available at 4/7. United's rate was 74%, Brentford had scored in 5 out of 8 of their last away games. The probability was substantially above 57%. United won 2-1 and BTTS “Yes” paid out.

That's a market inefficiency you can find every week by checking FootyStats' BTTS tables before placing a bet. Remember to only back BTTS "Yes" when both teams individually have a BTTS rate of 60%+. When you find that overlap, the odds rarely reflect the true combined probability.