Friday 11 January 2013

The Premiership Title Race. What Chance Manchester City ?

Over much of its life the Premiership title race has been a contest lacking in runners. It is never less than absorbing for much of its course, even if you can merely admire the brilliance of a runaway winner, but the preseason list of potential winners would rarely qualify for each way betting were it actually a horse race.

Midseason is an ideal time to examine those contenders at the head of the table with an eye to quantifying the likely advantage that the current leaders have gained. More league matches have now been played than remain to be contested and only the January transfer window remains as a means to bridge the gap between Premiership triumph and mere Champions League qualification.

You have to go back five completed seasons before the team leading after 21 games failed to see through the task to May. On that occasion Arsenal led the table with 50 points, two points clear of eventual champions, Manchester United and six clear of Chelsea, who also toppled the Gunners for second place. Therefore, current table toppers, Manchester United enter 2013 in an extremely strong position, seven points clear of near neighbours, City and an impressive 13 in front of their next nearest contenders, Tottenham.

 Recent history would suggest that Sir Alex Ferguson should be celebrating yet another title come early May and that view is unsurprisingly reflected in the prices. However, forecasting events has to be couched in terms of uncertainty and probability. An event may be extremely likely to happen, but nothing is certain until all alternative outcomes have been completely extinguished.

Estimating United’s chances of wresting the title from City is dependent upon the size of their current points advantage, but smaller, more subtle influences are also at play. For example, a seven point lead is welcome at any time of the season, but the number of nearest challengers can make the gap seem more or less comfortable. A single pursuer turns the contest into a match bet, but four challengers, each within seven are potentially a much bigger threat.

A related factor is the overall quality of the division as a whole. The chasing pack will view a game against the current leaders as a vital opportunity to claw back their lead. But increasingly the challengers will hope that other teams, unconnected with the title race will be able to muster a performance capable of also toppling the quarry. A team’s overall quality will be made up of a small number of performances which eclipse their usual levels of performance and the chasers ideally want lesser teams to produce such performances against the leaders, but not the chasing pack. No one single game can be said to have won or lost a title, but there is no denying the huge influence Wigan’s surprise 1-0 win over Manchester United in April 2012 had on the increasingly closely fought title race. On that particular day, Wigan produced a performance well in excess of their overall season long ability and blew the 2011/12 title race wide open.

So to begin to make an informed decision regarding the likely destination of the Premiership trophy, we need to have views and information on the amount of games left to be played, the quality of the leaders and the chasing pack, the number of credible challengers and lastly, the quality of the bit part players, who may potentially take a hand, without being actively involved in the finish.

Many factors are easily defined. Average number of games played denotes the stage in the season and points per game accrued by teams is a good proxy for ability and requires little strength of schedule correction at around halfway. The spread of ability within the division and closeness of the challengers can be adequately described by converting each team’s points per game figures to standard scores. Such conversions incorporate the average points per game total of all teams in the league and also how dispersed each individual total is around the mean.

Runaway leaders who are also vastly superior to the overall talent pool are therefore, more easily identified. For example, Chelsea’s standard score for their points per game halfway average in 2004/05 was 2.43, with Liverpool their nearest challengers with a score of just 1.35. The most congested midterm title race of recent years was 2007/08 when United were locked together with Liverpool, followed closely by Chelsea.

To win the title, the champions, rather obviously must finish above every other team in the league. So by using our estimations of each side’s ability compared to those of their immediate challengers and also the spread of quality throughout the rest of the league, we can create a variety of regressions to attempt to predict the finishing order in 2012/13 using historical precedents from the previous Premiership tables at halfway.

No one method perfectly captures all the inputs we have available. We can for example estimate the chances of United finishing above City given the current state of the table, but this approach then has to account for the admittedly small possibility that Spurs may then finish above both of them.

Alternatively, we can use the halfway position of each contender and their four nearest challengers as the inputs an actual finishing position as the output. Overall each method tries to capture at least some of the quality of the title contender and the quality of the challengers and in the table below I've averaged the likelihood of the top seven going on to win the title using a variety of such methods. No one method gives the current leaders a greater than 78% of winning in May.

United's Title Chances Given the Strength of their Challengers.

Team. Title Chance.
Manchester United. 71%
Manchester City. 22%
Spurs. 3%
Chelsea 2%
Everton. 1%
Arsenal. 1%
WBA. 0.5%

United, as expected are strong favourites and failure to lift the title would be a major shock, although not as big as Newcastle's post January slump in 1995-96 when they ceded a six point lead, made more formidable by their rivals having mostly already played their 23rd games of the season compared to Newcastle's 21. Newcastle should have been considered 80% favourites at that stage.

None of these figures account for anticipated transfer dealings. Everton would be more likely to be net sellers in the window than Chelsea and many would consider WBA's odds in excess of 250/1 still to be optimistic in the extreme. In addition these odds will have potential for rapid fluctuation as games begin to slip away.

The predictions are based on the life of the Premiership and the make up of the league was different in the 90's compared to now when a handful of teams dominate the title race. Hopefully some of the variety has been accounted for in the analysis and City fans will be equally hopeful that the '95-96 season, where Newcastle were caught by a certain other team from Manchester, will be repeated 17 years later.

For a much more detailed look at the title race check out Simon Gleave here.

No comments:

Post a Comment