Matt talks stats / Matthew Walton / 22 September 2008 / Leave a comment

"Magical" Matthew Walton wonders whether the team from the Big Four that does best out of playing each other automatically goes on to win the title. Let's see what the stats say.

And so it was honours even at Stamford Bridge. Neither were Chelsea nor Manchester United able to secure the victory which would have given them all three points.

As a result we've seen little change in the outright market for the Premier League Winner. Chelsea [2.25] stay as favourites, Manchester United [3.75] remain their nearest challengers - as far as the market is concerned - whilst Arsenal [5.8] and Liverpool [11.00] complete the top four of the list.

Thereafter, in betting terms, it's all over. Punters see virtually no hope of any other side getting into the shake-up (or 'sheikh-up' in the case of Manchester City) as these Big Four clubs appear to have it between them once again. And, as formbook students, it's hard to disagree with that view as 15 of the 16 Premier League titles (94%) have been shared amongst these sides with only Blackburn Rovers in 1994-95 gatecrashing this exclusive party.

Hence this discussion. If these top clubs are so dominant in the modern era, surely finding the Premier League winner is merely a question of determining which of these sides is the best. Why waste our form study on 20 clubs, when a simple assessment of the respective strengths and weaknesses of the 4 major protagonists would suffice.

After all, punters clearly think this to be true. Chelsea beat Manchester City the week before, something which made very little difference to the market. However, had the Blues beaten Manchester United the following week, that would have caused a seismic shift in the betting market (and, of course, vice-versa).

And we saw it the week before as Liverpool beat United at Anfield. A win for Benitez's side all of a sudden gave their title chances a credence with punters, and so the market, which any number of wins over the likes of Sunderland or Middlesbrough would never have achieved. Why? Because it was over one of the other Big Four sides and they are all that matters.

To take this a step further we've studied the last decade of Premier League fixtures to see if this theory stands up to scrutiny. Is the best of the Big Four the best of the top flight? Does a winning of their 'mini-league' mean a winning of the Premier League?

What we find is this four team, 24 match, seasonal inter-club competition - comprised of the four sides playing each other home and away - has been won by Manchester United on three occasions (2007-08, 2002-03 & 1999-2000), by Chelsea twice (2004-05 & 2005-06), Liverpool once (2000-01) and Arsenal no fewer than four times (1998-99, 2001-02, 2003-04 & 2006-07). There has not been a tie in any of the last 10 years.

In the seasons they triumphed, Manchester United went on to win the league every time. Three times they topped this mini-league, the same three times they went on to win the Premier League. The same goes for Chelsea. Their two wins in this mini-League this century correspond to their two main titles.

However, this is where it starts to go a little pear-shaped. Arsenal have only won the 'big one' on two occasions (2001-02 & 2003-04) when they came out best of this quartet of clubs. The other times, in 1998-99 they came 2nd to Manchester United by a single point and in 2006-07 they finished only 3rd behind United and Chelsea.

Same goes for Liverpool. Winners of this section in 2000-01, the Reds could only trail in 3rd as Manchester United beat Arsenal by 10 points!

How frustrating that must have been for followers of the trophy-less Merseysiders. They outperform their main rivals for the title (in 2000-01 they beat Manchester United home and away as well as beating Arsenal and drawing with Chelsea) only to blow the league by their performances against the supposed 'also-rans'.

And that highlights one of the key side-issues to this debate. In following this system blindly we're assuming that each of the Big Four will perform to a similar level against the rest of the division, as such making the winner of their in-fighting the only team to consider for the Premier League title.

But it wasn't the case for Liverpool that year, nor for Arsenal in 1998-99 and 2006-07. Still, despite these blips, this strategy returns a healthy 70% strike rate.

And a study of the whole period, taken collectively, does highlight the relative merits of these four clubs. Over the 10 year sample, these sides have played six matches a season against their rivals, home and away, making a total of 60 matches versus the other three teams.

The overall standings have Manchester United on 90 points (W24 D18 L18), Arsenal on 86 points (W21 D23 L16), Chelsea with 84 points (W21 D21 L18) and Liverpool trailing some way behind on 62 points (W16 D14 L30). In point of fact, Liverpool have collected the wooden spoon in each of the last six seasons in this study - no wonder they can't win the title!

After reviewing the Premier League, out of interest we then compared our findings to two other monopolised divisions, the Scottish Premier League and Spain's Primera Liga.

In the case of the SPL we have just two sides to consider, such is the dominance of Celtic and Rangers, but they do at least play each other four times a season. During the last decade, they have shared the seasonal points on three occasions but Celtic have gained more points five times and Rangers twice. When there has been a winner, Celtic won the SPL four times out of five (2005-06, 2003-04, 2001-02 & 2000-01) and Rangers once (1999-2000).

That's just a 50% strike rate over the last 10 years, lower than the Premier League.

As for the Spanish example, only Real Madrid and Barcelona are worth considering and here 2006-07 (Real Madrid) and 2005-06 & 1998-99 (Barcelona) proved the only accurate guide for overall league success. A very poor 30%.

So what should we conclude? Well, in the Premier League, the winner of this mini-League is a pretty accurate guide to the overall title - six of the last seven seasons, seven of the last 10.

What we shouldn't do, however, is pile on Chelsea just because they beat Manchester United or Arsenal, or vice-versa. We should be a little more patient and see this as just one match in a series of 24 such contests and try to see the bigger picture.

You might be very well advised to lay a bit of this premature money as opposed to jumping on the bandwagon.