A team on a 5-game winning streak is not necessarily in great form. A team that has lost three in a row might be the stronger side. Here is how to read form properly — using data instead of narrative.
The standard football form guide — W, D, L, W, W — is one of the most widely reported and least useful pieces of football data available. Wins and losses contain almost no information about how well a team actually played. They are the output of performance, not performance itself.
A team can string three wins in a row by beating relegated sides 1-0, scoring a late penalty in each. Their xG might be 0.6 per game — well below average. They are not in good form; they are getting lucky at the right moment against weak opposition, and the wins are not going to continue at the same rate.
Properly analysing form means looking at how teams are playing, not just whether they are winning. That requires digging into chance quality (xG), defensive solidity (xGA), opponent quality, and venue splits.
Set the right sample window
The last 5 games is too small for reliable statistics. 10–15 games is better — but you need to weight recent games more than older ones. A common approach: last 3 games weighted 3×, games 4–7 weighted 2×, games 8–12 weighted 1×. This captures genuine recent form while retaining enough sample size for meaningful patterns.
Split home and away form
Home and away performance are different datasets. A team averaging 2.1 xG at home and 0.9 xG away is not "a 1.5 xG team" — they are genuinely two different teams depending on venue. Always analyse home xG and xGA separately from away. For the upcoming match, use the relevant venue-specific figures.
Adjust for opponent quality
Generating 2.5 xG against the bottom 3 clubs means less than generating 1.8 xG against the top 6. Raw xG averages over a sample mix very different opponents. Ideally, use the opponent's defensive xGA rank to contextualise each team's attacking xG figure. A team averaging 1.6 xG against strong defences is more impressive than 1.6 xG against weak defences.
Check xG trends, not just averages
Is the xG increasing or decreasing over the window? A team whose xG has risen from 1.1 to 1.8 over the last 8 games is in improving form — their upcoming xG should be modelled closer to 1.8 than the average of 1.45. A team whose xG has dropped from 2.0 to 1.2 over the same period is declining. Trend matters as much as average.
Cross-check xG against results
Are the results matching the xG? If a team is winning games on 0.8 xG per match, they are over-performing. If they are losing on 2.0 xG, they are under-performing. The gap between results and xG is the form data's most actionable signal — it tells you which team's record is likely to regress and which is likely to recover.
Consider two teams heading into the same fixture. Team A has won 4 of their last 5. Team B has won 1 of their last 5. Based on results, Team A look like heavy favourites. Now look at the xG:
| Metric (last 5 games) | Team A (4W 1L) | Team B (1W 4L) |
|---|---|---|
| Avg xG per game | 0.82 | 1.95 |
| Avg xGA per game | 1.45 | 0.90 |
| xG difference per game | −0.63 | +1.05 |
| Goals scored avg | 1.80 | 0.80 |
| Goals conceded avg | 0.60 | 1.80 |
| Result-based form | WWWLW | WLLLL |
Team A are winning games on low xG and conceding heavily against their xGA (they are conceding more actual goals than their xGA suggests — their goalkeeper has been exceptional). Team B are generating elite xG per game but converting at a low rate.
The result-based narrative says Team A are in form. The xG narrative says Team B are the superior team and are likely to perform better going forward. In a match between these two, Team B are actually the stronger side despite their results — and may be underpriced by a market that has overweighted recent results.
Research across multiple leagues has established a hierarchy of how much various form metrics predict future performance:
Captures both attacking threat and defensive quality in a single number. Highly stable over 10+ game samples.
Leading indicator of attacking momentum — teams improving in these metrics tend to see xG rises shortly after.
More predictive than shot volume — filters speculative efforts and measures attempts that genuinely challenge the goalkeeper.
Good defences limit xGA consistently. More stable than actual goals conceded, which includes goalkeeper variance.
Useful for identifying psychological momentum but heavily contaminated by luck, opponent quality, and finishing variance.
Most commonly reported but least predictive — highest noise of any common form metric.
The betting market is driven heavily by narrative. Media coverage, fan sentiment, and bookmaker pricing all disproportionately weight recent results over underlying xG form. This creates specific, recurring value opportunities:
📈 Value play: Strong xG, poor results
Teams losing games on strong xG are regularly underpriced. The market sees the losses; the model sees the underlying quality. Back them against similarly-priced opponents whose form is based on lucky wins.
📉 Fade play: Weak xG, strong results
Teams winning on low xG are regularly overpriced. Back the other side or take Under bets — their goal output is likely to drop as luck normalises.
🏠 Home form divergence
Teams with strong home xG but poor away form are systematically underpriced when playing at home. The market combines home and away, missing the venue split.
📊 xG trend momentum
Teams whose xG is rising sharply over the last 3–4 games are undervalued if the market is using full-season averages. Bet toward the improving trend, not the average.
FPL managers who treat form as the primary selection criterion are making a systematic mistake. A player who has blanked three gameweeks in a row but is maintaining high xG per 90 is a buy, not a sell. A player on a hot streak with declining xG is a sell candidate.
The same framework applies: analyse xG and xA per 90 over the last 5–8 gameweeks, weighted toward the recent end. Check whether team form is improving (rising team xG benefits all their attackers) and whether fixture difficulty is changing (upcoming FDR). Players combining high xG + improving team form + easy upcoming fixtures are the clearest buys.
For defenders, team xGA trend is the primary form signal — a defence that has reduced xGA from 1.5 to 0.8 per game over 5 matches is improving, and clean sheet probability for their players is rising. The market reacts slowly to these trends, creating a window to buy before prices rise.
A minimum of 5 games for a quick sense-check, 10–15 for reliable statistical patterns. Weight recent games more heavily — games from 8+ weeks ago are less predictive of current performance due to potential tactical and personnel changes. A rolling weighted window of 8–12 games is the practical optimum for most analysis.
They should be treated as separate datasets. Home advantage is real — home teams score roughly 40% more goals on average across European top leagues. A team's home xG and xGA should be analysed separately from their away figures. For the upcoming match, use the relevant venue-specific form data rather than combined totals.
Yes — xG-based form analysis is one of the most consistent sources of betting value. Teams whose xG significantly diverges from their results (in either direction) are frequently mispriced by a market that overweights results. Identifying this divergence early — before the market adjusts — is the core of xG-based value betting.
xG vs Actual Goals
Why goals lie — and what xG reveals about true performance
What is xG?
How expected goals are calculated and why they matter
Value Betting Explained
How to convert form signals into actionable betting value
Poisson Distribution
Using xG form data to model match outcomes and odds
FPL Tips: Using Data
How to apply form analysis for Fantasy Premier League decisions
Expected Points (xPts)
Converting xG form into expected league table position
