Sport News

By Martin · Published 14th September 2026 · Last updated 14th September 2026

Quick answer: Form tables lie because "WWLDL" only records results, not performance. A team can win five games while creating fewer chances than it concedes, or lose five while dominating. Better football form analysis looks at underlying metrics instead: xG difference (expected goals minus expected goals against), shot quality, and performance against opposition strength. A team with a strong xG difference but poor recent results is usually better than its form suggests, and vice versa. Reading form through underlying numbers rather than the WDL string is one of the biggest edges available to anyone assessing football matches.

Every football broadcast shows it. Five little letters next to each team: W-W-L-D-L. Recent form, at a glance.

It looks like information. It's actually one of the most misleading things on the screen.

A form string tells you whether a team won, drew, or lost. It tells you nothing about how. It can't distinguish a fortunate 1-0 win against the run of play from a dominant 4-0 demolition, or an unlucky loss where a team hit the woodwork three times from a deserved thrashing. Two teams can carry identical "WWWLL" strings while being genuinely miles apart in quality. This is the full breakdown of why form tables mislead, and the underlying metrics that actually tell you how a team is performing.

Why do football form tables mislead?

Football form tables mislead because they record outcomes, not performances. The WWLDL string treats every win as identical and every loss as identical, ignoring how the result was achieved. A team riding luck to narrow wins looks the same as a team dominating opponents, even though one is far more likely to keep winning than the other.

The core flaw is that results contain luck, and form tables can't strip it out.

A single deflected goal, a hot goalkeeper, or a soft penalty can flip a result without reflecting how the teams actually played. Over five games, these lucky and unlucky swings pile up, producing a form string that can point in the opposite direction to a team's true level.

So a form table answers "what happened?" when the useful question is "how did they play, and will it continue?" Those are different questions, and only the second one predicts what comes next. Reading form through results alone is how people repeatedly overrate lucky teams and write off unlucky ones, a mistake closely tied to regression to the mean in football form.

What is wrong with WWLDL as a form metric?

WWLDL is wrong as a form metric because it's binary and context-blind: it reduces every match to win, draw, or loss with no information about margin, chance quality, or opponent strength. A 1-0 win against the league leaders and a 1-0 win against the bottom side count identically, even though they reveal completely different things about a team.

Three specific blind spots make WWLDL crude.

It ignores performance quality. A team can win 1-0 having been outshot and outchanced, or lose 2-1 having created double the opposition's chances. WWLDL records only the result, hiding which team actually played better.

It ignores opposition strength. Three wins against relegation-threatened sides and three wins against the top six look identical in a form string, but they mean very different things about a team's real level.

It ignores margin and dominance. A scrappy 1-0 and a commanding 5-0 are both just "W," even though one signals far more about a team's underlying strength.

Strip a form table down and it's a highlight of outcomes with the entire story of how those outcomes happened removed. That missing story is exactly where the predictive signal lives.

What should you look at instead of form tables?

Instead of form tables, you should look at underlying performance metrics: xG difference, shot quality, and results weighted for opposition strength. These reveal how a team is actually performing rather than what its recent results happened to be, and they predict future outcomes far more reliably than a WWLDL string.

The best replacements for the form table are all forms of "did they deserve their results?"

The single most useful of these is xG difference. As FBref's own explanation puts it, a team's xG difference indicates how it should be performing, so a negative goal difference paired with a positive xG difference often signals poor luck or below-average finishing rather than a genuinely bad team. That gap between results and underlying numbers is the signal the form table can't show you.

What is xG difference and why does it beat form?

xG difference is a team's expected goals minus its expected goals against, measuring the quality of chances it creates versus the quality it concedes. It beats form because it captures performance directly rather than results, revealing whether a team's recent outcomes were deserved or driven by luck. A strong xG difference predicts future results better than a strong WWLDL string.

The logic is straightforward. Expected goals, introduced by Opta in 2012 and now built on millions of historical shots, measures the quality of each chance. Sum a team's chances and the chances it concedes, take the difference, and you have a measure of how well it's genuinely playing, independent of whether the finishing went in.

A team with a positive xG difference is creating better chances than it allows. If its results don't yet reflect that, it's been unlucky, and results usually catch up. A team with a negative xG difference winning games is riding luck that typically runs out.

This is why a form table and an xG-difference table can tell opposite stories about the same team, and why the xG version is the one worth trusting. We cover the full metric family in our comparison of xG, xT and npxG, but xG difference alone already outperforms any WWLDL string for reading true form.

Why does opposition strength matter in form analysis?

Opposition strength matters because a result only means something relative to who it came against. Three wins against weak teams and three wins against strong teams produce identical form strings but reveal completely different levels of quality. Adjusting recent results for the strength of the opponents faced turns a misleading form run into a meaningful one.

Form tables treat every opponent as equal. They never are.

A team on a five-game winning streak built entirely against the bottom of the table is far less impressive than the same streak against top-half sides. Yet the form string shows five identical Ws. Without opposition context, you'll systematically overrate teams on soft runs and underrate teams whose recent fixtures were brutal.

This is also why upcoming fixture difficulty matters as much as past form. A team's "great form" against weak opponents can evaporate the moment it faces stronger ones. Weighting results by opponent quality, alongside xG difference, is what separates a genuine form read from a surface-level one, and it's one of the 12 data points that move predictions most.

How does AMpredict read form differently?

AMpredict reads form through underlying performance rather than the results table, weighting xG difference, shot quality, and opposition-adjusted results above the raw WWLDL string. The mathematical layer measures true performance, the AI layer identifies where results and underlying numbers diverge, and human review confirms whether a divergence reflects luck or genuine change.

Form is treated as a question of true level, not recent outcomes, throughout the three-layer methodology.

The mathematical model never takes a winning or losing streak at face value. It checks whether xG difference, shot quality, and opposition strength support the results. A team overperforming its underlying numbers gets its form discounted toward its true level; a team underperforming gets credit its results don't yet show. The AI layer, trained on thousands of matches, recognises the pattern of results and underlying numbers diverging and then converging. Human review then separates the crucial cases: a divergence caused by variance, which corrects, versus one caused by a real change like a new signing or tactical shift, which doesn't.

That distinction is why form is a signal to interpret, not a table to read off. The result feeds the confidence tiers across the VIP packages, where teams flattered by results are rated more cautiously than their form suggests, and unlucky teams more generously.

The Bottom Line

Form tables lie because they record results, not performance. The WWLDL string treats every win and every loss as identical, ignoring chance quality, margin, and opposition strength, so a lucky team and a dominant one can carry the same form while being genuinely miles apart. It answers "what happened?" when the question that predicts the future is "how did they play?"

The fix is to read form through underlying metrics: xG difference above all, plus shot quality and opposition-adjusted results. A team with a strong xG difference but weak recent results is usually better than its form suggests, and a team winning ugly with poor underlying numbers is usually worse. Those gaps are exactly what the results table hides.

At AMpredict, form is read through performance, not outcomes, so predictions track a team's true level rather than its latest run of letters. It's one of the clearest edges available to anyone willing to look past the form table.

Want form read through real performance, not five misleading letters? Explore AMpredict membership plans and get every team assessed on underlying numbers, built on the full three-layer methodology before your next weekend kickoff.