The Longest Unbeaten Runs in Premier League History
The longest unbeaten runs in the Premier League: the 49-match record, winning streaks, home records, winless sequences, and how English records get verified.
By CricketTaken EditorialPublished Premier League19 min read
- Longest unbeaten league run
- 49 matches
- Previous English record
- 42 matches
- Longest home league unbeaten run
- 86 matches
- Longest run without conceding
- 1311 minutes
A record sequence is a strange thing to hold, because it is only a record after it stops. Nobody knew Arsenal's run was 49 matches long until the afternoon it became 49 and no longer. Until then it was just a number going up, and the reporting around it treated every fixture as a possible ending.
English football has kept these sequences for as long as it has kept anything. The Football League began in 1888, the Premier League replaced the old First Division in 1992, and the records that matter to most supporters straddle that boundary awkwardly — the deepest unbeaten runs in the English top flight were set on both sides of it, under different competition rules and against different numbers of clubs.
What follows sets out the runs that stand, the ones that were broken, and the reason the answer to "what is the record?" is sometimes genuinely contested.
Forty-nine, and how a run gets that long
The record is Arsenal's, and it is 49 league matches without defeat. It began in May 2003, ran through the whole of the 2003/04 campaign, carried into the following autumn, and stopped at Old Trafford on 24 October 2004 when Manchester United won 2-0.
Two features of that run are worth separating. The first is that a complete unbeaten season sits inside it, which is what makes the sequence famous rather than merely long. The second is that the run is not a season — it is nine fixtures from one campaign, 38 from another and nine from a third, stitched together by the calendar rather than by anything the competition itself recognises as a unit.
That distinction matters when you compare records. Nottingham Forest held the previous English mark at 42 matches, set across the late 1970s, and their sequence had the same shape: a championship season with an unbeaten stretch either side of it. Liverpool ended it in December 1978. Forest, promoted from the Second Division in 1977 and champions of England at the first attempt, were an unlikely holder of any endurance record, which is a useful corrective to the idea that these things belong exclusively to the largest clubs.
Burnley's 30-match run in 1920/21 stood as the English record for more than half a century and is the one most often forgotten. It came in a 42-game First Division season immediately after the First World War, in a competition that had barely restarted, and Burnley won the title with it.
The full account of the season inside Arsenal's run is in the guide to the Invincibles, which goes into the squad, the twelve draws and the trophy. This page is about where that sequence sits among everything else.
Consecutive wins, which is a harder record
Not losing and winning are different disciplines, and the second is far more demanding over a sequence.
The Premier League record for consecutive victories stands at 18. Manchester City reached it during 2017/18, the season they finished on 100 points, and Liverpool matched it across 2019/20. City's sequence was completed inside a single campaign, which is the distinction usually drawn when the two are compared, and it is a fair one — a run that crosses a summer includes a transfer window, a pre-season and a change of circumstances that a continuous in-season run does not.
Eighteen consecutive wins means eighteen away days, home fixtures against sides defending in numbers, and midweek matches after European trips, all resolved in your favour. A single well-organised opponent holding out for a goalless draw ends it, and over eighteen attempts several will try.
The old First Division record for consecutive wins is a different number under different conditions, and comparisons across the 1992 boundary tend to generate more heat than light. The competitions had different numbers of clubs, different fixture loads and, before 1981, two points for a win rather than three — a change that altered the incentive to protect a draw and therefore altered the shape of every sequence recorded after it. Anyone treating a pre-1981 and a post-1981 run as equivalent is comparing two different games.
The constraint on any winning streak is the away fixture. Roughly half the matches in a sequence are played somewhere else, and the accumulated disadvantage of travel, crowd and unfamiliarity is what turns a probable win into a probable draw. A side capable of winning most home matches comfortably will still find that its eighteen-game run has to survive nine trips, several of them to grounds where results have historically been poor regardless of the table. That is why the very long winning runs cluster in seasons where a club was also breaking points records: it takes a genuinely dominant team to hold a high win probability on the road for two months.
Winning streaks also correlate less with titles than people assume. A side can win fourteen in a row and finish second, because the fourteen were preceded by a September in which it took four points from five matches. Consistency across ten months, not a spike inside it, is what the list of Premier League champions actually records.
Home and away, counted separately
Home records are their own category, and the numbers are much larger because half the difficulty has been removed.
Chelsea hold the English top-flight record with 86 consecutive home league matches unbeaten at Stamford Bridge, a run that began in February 2004 and ended in October 2008 when Liverpool won there. Four and a half years, four managers, two title-winning seasons and a Champions League final in the middle of it, and no home league defeat.
Liverpool's own Anfield sequence reached 68 home league matches unbeaten before Burnley won there in January 2021. That ending is instructive: it came during a season played in empty grounds, which removed the single largest component of home advantage and produced results across the division that nobody would have predicted in a normal year.
| Sequence | Club | Length | Ended by |
|---|---|---|---|
| Home league matches unbeaten | Chelsea | 86 | Liverpool, October 2008 |
| Home league matches unbeaten | Liverpool | 68 | Burnley, January 2021 |
| All league matches unbeaten | Arsenal | 49 | Manchester United, October 2004 |
| All league matches unbeaten | Nottingham Forest | 42 | Liverpool, December 1978 |
Away unbeaten runs are shorter, less often quoted and arguably more impressive. There is no equivalent headline record in the English game because away sequences are broken so much faster — the travel, the crowd, the referee's marginal decisions and the pitch all move against the visiting side, and the compounding effect of that over a dozen fixtures is severe.
What home advantage is actually made of
Four things, roughly, and they are not all crowd noise.
Familiarity with the surface is the first and the most underrated. IFAB's Law 1 permits a range of pitch dimensions rather than a single size, and although UEFA standardised its own competitions at 105 by 68 metres, English grounds have historically varied enough that a home side knew the width of its own pitch in a way a visiting winger did not. Travel is the second: a Saturday fixture 300 miles from home involves a Friday night in a hotel and a coach journey, and the physiological cost of that is measurable.
Officiating is the third, and it is uncomfortable. Studies of crowd influence on refereeing decisions across European leagues have consistently found a home bias in marginal calls, which is one of the reasons the 2020/21 season played behind closed doors produced results that broke long home sequences across several divisions. The fourth is simply routine — the same dressing room, the same warm-up, the same kick-off time.
Add those together over 86 consecutive fixtures and the achievement looks slightly less mythical and slightly more explicable, without becoming any less difficult.
One quiet consequence of the home records is what they say about scheduling. A side playing at home on a Saturday after a European away trip on Thursday is defending a home record under conditions the record was not built in, which is one reason how the fixture list is compiled is worth understanding before drawing conclusions from any sequence.
Unbeaten runs that ended in nothing
A long unbeaten run does not guarantee a title, and the exceptions are more common than the folklore allows.
Forest's 42-match sequence bridged the 1977/78 championship and the following season, in which they finished second. They also won the European Cup that year, so the season was hardly a failure, but on the domestic record the run outlived the title it started with.
The more general pattern is that unbeaten runs are frequently built on draws, and draws do not win championships. A side that goes 20 matches unbeaten with 12 draws has 24 points from a possible 60. A side that goes 20 matches with 14 wins and six defeats has 42. The second side has lost six times and is comfortably ahead. That arithmetic is the whole reason three points for a win was introduced in the English league in 1981: the Football League wanted to make the draw less attractive, and it worked.
There is a psychological version of the same point. A run that becomes newsworthy starts to affect team selection, and there is decent circumstantial evidence across English football of sides playing more conservatively as a sequence lengthens. Whether that is real or a story told afterwards is not something a record book can settle, and any page claiming to know is guessing.
Unbeaten starts, and why they get more coverage than they deserve
Every August, a club goes eight or nine matches without losing and the phrase "record start" begins appearing. Most of those sequences end by October and are forgotten by Christmas, which is a reasonable outcome for a statistic that was never worth much.
The reason a start looks impressive and usually is not comes down to who a side has played. A fixture list is not evenly distributed by difficulty. A club can be handed four home matches in its first six, three of them against promoted sides, and emerge unbeaten having achieved almost nothing. The same club can then play four of the previous season's top six in a fortnight in November and lose two. Judging a sequence before it has passed twenty matches is judging the fixture computer rather than the team.
There is a second problem with starts, which is that a club is fresher in August than at any other point in the season. Injuries accumulate. Suspensions accumulate. European fixtures have not yet begun for clubs entering at the league phase, and the international window has not yet returned a squad from three continents. A run built in the first month is built in conditions the rest of the campaign will not repeat.
Where a start genuinely counts is at the other end of the table. A newly promoted club that goes six or seven matches without defeat has usually banked points it will need in April, and the psychological effect on a dressing room that expected a difficult autumn is real even if it is hard to quantify. The value is the points rather than the record.
The honest way to read any in-progress sequence is to ask two questions: how many away matches has it included, and how many of them were against sides likely to finish in the top half? A twelve-match unbeaten run containing seven home fixtures against the bottom half is a different object from a twelve-match run containing seven away trips, and only one of them is evidence of anything.
Clean sheets, and the goalkeeping records
The defensive sequences are cleaner to verify than the team ones, because a goalkeeper either conceded or did not.
- 1311Van der Sar, minutes without conceding
- 14Consecutive clean sheets in that run
- 24Cech, clean sheets in 2004/05
- 15Chelsea, goals conceded in 2004/05
Individual and club records from specific documented Premier League seasons. These are not season averages.
Edwin van der Sar went 1,311 minutes of Premier League football without conceding for Manchester United in 2008/09, across 14 consecutive clean sheets, before the run ended in March 2009. It is the competition's benchmark and it has stood a long time.
Petr Cech's 24 clean sheets in the 2004/05 season is the other landmark, set in the campaign Chelsea conceded 15 league goals in 38 matches — the fewest in a Premier League season and a figure that looks more absurd the longer you sit with it. Fifteen goals across nine months. Two clubs have gone a whole season unbeaten; none has come close to that defensive total since.
The runs are a goalkeeper's record and a defence's record simultaneously, which is why the individual name attached to them is slightly misleading. A keeper cannot manufacture a clean sheet against a side that creates ten chances, and the sequences above were built by back fours that did not permit ten chances.
The record nobody logs properly
The reverse statistic — the longest run without a clean sheet — is one of the more commonly searched and least reliably recorded numbers in English football. It is tracked by statistics providers rather than by the competition itself, and the figures that circulate depend on whether cup matches are included, whether the run is treated as belonging to a club or a goalkeeper, and whether a sequence that crosses a relegation counts as continuous.
The honest answer is that no single authoritative English figure exists in the way it does for unbeaten runs, and anyone quoting one should be asked where they got it. This is a genuine gap in the record-keeping rather than a fact this page is withholding.
Winless runs, and the seasons that ended in relegation
Derby County's 2007/08 season is the reference point for the other direction. They won a single Premier League match, against Newcastle in September, and did not win again in the competition that season. The final record was one win, eight draws, 29 defeats, and 11 points from 38 matches — the lowest total in a 38-game Premier League campaign.
- Won1
- Drawn8
- Lost29
The final record from that campaign. It is the lowest points total recorded in a 38-match Premier League season.
Show the numbers
| Item | Value |
|---|---|
| Won | 1 |
| Drawn | 8 |
| Lost | 29 |
Relegation was mathematically confirmed in March, earlier than any club before or since, which is its own record and a bleak one. The club had been promoted through the play-offs the previous May and had assembled a squad on second-tier resources, which is the mechanism behind most seasons of this kind rather than a failure of effort. The financial and squad-building reasons a promoted club can be overwhelmed are set out in the guide to Premier League relegation.
Winless sequences are where the verification problem becomes obvious. Sunderland's sequence of consecutive Premier League defeats is frequently quoted as spanning two separate spells in the division, on the basis that the seasons in between were played in the second tier and therefore do not interrupt the club's Premier League record. That is one defensible convention. Treating a relegation as ending a run is another, equally defensible. The two produce very different numbers, and neither is wrong — which is exactly why the figure you see depends on who compiled it.
Scoring streaks, at both ends
Arsenal hold the English record for consecutive league matches in which a club has scored, at 55, set across the 2001 to 2002 period. The previous holder was Chesterfield, who managed 46 in the Third Division North in 1929/30 — a reminder that these records are not the property of the top flight and that the lower divisions of English football have kept their own books for just as long.
The individual version belongs to Jamie Vardy, who scored in 11 consecutive Premier League matches for Leicester City during 2015/16, breaking Ruud van Nistelrooy's mark of 10. Vardy's run came in the season Leicester won the title, and it is one of the few streak records that coincided exactly with the achievement it helped produce.
Scoring streaks are the most fragile of all these sequences, because they depend on a single player being fit, selected and in the right kind of match. A hamstring in the eighth game ends it. The equivalent goalscoring records over a season rather than a sequence are covered in the guide to the Premier League Golden Boot, which is a different and more forgiving measure.
At club level, a long scoring run tells you about attacking depth rather than about a striker. A side that scores in 55 consecutive matches has scored through injuries, suspensions, away fixtures at defensive teams and dead rubbers in May.
Why long runs are harder to build now
The calendar has changed, and it has changed against the sequences.
European competition expanded in 2024 to a 36-team league phase, which added matches to the autumn for every English qualifier. The Club World Cup expanded to 32 teams from 2025, taking a summer from the clubs involved. International windows return players from three continents 48 hours before a Saturday fixture. The cumulative effect is that a first-choice eleven cannot play every league match, and a rotated eleven is a beatable eleven.
Set against that, five substitutions have been permitted since the 2022/23 season, and squads are larger and better resourced than they were in 1978 or 1921. The compression is real but it is not the whole story, and anyone arguing that modern records are simply harder should account for the fact that the two longest winning streaks in Premier League history were both set well after the calendar began expanding.
There is a second factor that cuts the other way and gets less attention. The bottom half of the Premier League is better coached than it has ever been. Low-block defending, set-piece specialists and structured pressing are now standard at clubs with no realistic ambition beyond survival, and the well-organised 0-0 that keeps an unbeaten run alive is the same match that stops a winning streak dead. Runs of both kinds are ended by the same phenomenon.
Records held outside the traditional elite
It would be neat if every record on this page belonged to the same four or five clubs. It does not.
Nottingham Forest held the English unbeaten record for a quarter of a century. Burnley held it before that. Chesterfield held the scoring-streak record from the Third Division North. Preston North End remain one of only two clubs to have gone an English league season unbeaten. Derby County's 2007/08 record is unwanted, but it is theirs, and it has stood a long time.
Below the top flight the picture is richer still, because a dominant second-tier or League One side plays a longer season against a wider spread of quality. Reading's unbeaten start to a Championship campaign in the mid-2000s and the enormous points totals recorded in the third tier are part of the same story of English record-keeping, kept by the EFL rather than by the Premier League.
The point is not sentimental. It is that endurance records reward stability rather than spending, and stability turns up in unexpected places — a settled back four, a fit goalkeeper and a manager who has been in the job four years will produce sequences that money alone does not.
It also explains why these records tend to be set by clubs in an unusual moment rather than by the wealthiest clubs in an ordinary one. Forest were newly promoted and playing without expectation. Burnley were operating in a league that had barely reassembled after a war. Preston were in the first season of organised league football in England and were simply better organised than anyone else. In each case the circumstances were not repeatable, which is a large part of why the records have lasted.
How a record is actually verified
There is no single office in England that certifies a football record, which surprises people.
- Match recordThe competition publishes an official result for every fixture, which is the primary evidence for any sequence.
- CompilerStatistics providers and long-established football statistical organisations maintain the continuous sequences from those results.
- ConventionThe compiler decides the rules: league matches only or all competitions, whether relegations break a run, whether abandoned matches count.
- PublicationThe figure appears in the competition's own material, in club records and in the football almanacs and reference books that have kept English figures since the nineteenth century.
- ChallengeClub historians and researchers query figures against contemporary newspaper reports, and totals for pre-war seasons are still occasionally revised.
- SettledA figure becomes the accepted record when the major compilers agree on it, not when any governing body rules on it.
The practical route by which a sequence becomes an accepted record. No single body has final authority over historic figures.
Most disagreements come from the convention step. Does a run include cup matches? Does it survive a relegation? Does a match awarded 3-0 after an administrative breach count as a defeat for the sequence? The Premier League's own records generally cover its own competition from 1992 and no further back, which means any claim about the English top flight as a whole has been assembled by someone rather than certified by anyone.
Pre-war records are the least stable, because the primary sources are newspaper reports and club minute books, and researchers still find errors. A figure from 1921 should be read as the best current reconstruction rather than as a certified fact. Contemporary Premier League records are far more secure, because every match has been filmed, timed and logged.
What a long run tells you, and what it hides
A record sequence is a description of an outcome, not an explanation of quality. That is the awkward thing about all of the numbers above.
Forty-nine unbeaten includes twelve draws in the middle of it, several of them recovered from behind in the last twenty minutes. Eighteen consecutive wins might have been nineteen but for one deflection. A club with 86 home matches unbeaten was, on several of those afternoons, second best and did not lose. The sequence records what happened, and what happened is always partly a matter of which way the ball fell.
What a run does tell you, reliably, is that a side was hard to beat over a long period and that its availability held up. Those two things are not glamorous and they are the closest football has to a durable measure of a good team. A side that avoids defeat for 30 matches has a functioning defence, a fit spine and a manager who has solved the problem of what to do in matches where the plan is not working.
What it hides is the margin. It hides how many of those matches were 1-0, how many were rescued, and how close the record came to ending a dozen times. It also hides the seasons of clubs who did everything right and lost in the ninetieth minute in October, whose runs nobody counted because they finished at seven.
There is more on the English game across the football section, and the rest of these guides are indexed on the England hub. Anyone wanting to see how a record sequence reads while it is still running rather than after it has stopped will find the archive on the blog a better place to look than any record book, because the record book only ever gets written at the end.