History provides no guidance on Houllier’s future

Villa fans voice their views on Gerard Houllier, but the board can be forgiven for not knowing what to do (Paul Childs/Action Images)

Aston Villa’s 1-0 defeat to Wolves on the weekend heaped more pressure on Gerard Houllier, who now appears to have alienated fans along with his players following a string of disappointing results.

The Holte end made its feelings clear during the game, and despite backing from the board one feels Houllier is very close to leaving Villa Park.

Would there be much sense in sacking Houllier at this stage of the season, with 8 games left? The club are in an extraordinarily precarious position; in fourteenth place but only one point above the relegation zone. I wondered if there was a precedent for sacking managers in similar positions, and looked at all relegation-related managerial departures from the past 8 seasons.

This of course is not an exact science – what is a relegation-related departure? I’ve tried to be as vague as possible, looking at any sacking, resignation or departure by mutual consent for managers with clubs at risk of relegation. I’ve ignored, however, managers leaving clubs in a transfer-like situation, such as Owen Coyle to Bolton from Burnley, since they were not forced out in some way against their will, and in all probability wanted to leave the club more than their clubs wanted rid of them.

Since the start of the 2002/03 season, 27 managers have left relegation-threatened clubs:

• 19 were sacked (e.g. Paul Ince/Blackburn)
• 4 left by mutual consent (e.g. Billy Davies/Derby)
• 2 were put on gardening leave (e.g. Phil Brown/Hull)
• 2 resigned before they were likely pushed (e.g. Roy Keane/Sunderland)

The success rates – i.e. rate of avoiding relegation – for relegation-threatened clubs who part ways with their managers is remarkably high:

• 20 of the 27 clubs avoided relegation in the season they changed managers (74%)
• Of the 15 clubs whose manager left when they were in the relegation zone, 9 avoided relegation in the same season (60%)
• Managers left relegation-threatened clubs at an average of 17^th place in the league. By the end of the season, these clubs averaged 16^th in the league.

So if Houllier was sacked, Villa would probably stay up, right? Especially when only two clubs around them have changed managers this season (West Brom/Hodgson and Blackburn/Kean) in an eight-way relegation scrap?

Well, now is an incredibly risky point in time to change managers. In the 8 seasons studied, only four managers have seen sacked (including Phil Brown’s gardening leave last season) with 8 games or less left in the season. These were:

• Phil Brown (Hull), 2009/10, 8 games left, replaced by Ian Dowie
• Chris Coleman (Fulham), 2006/07, 5 games left, replaced by Lawrie Sanchez
• Terry Venables (Leeds), 2002/03, 8 games left, replaced by Peter Reid
• Jean Tigana (Fulham), 2002/03, 5 games left, replaced by Chris Coleman

Indeed, sacking relegation-threatened managers in the second half of the season has been a rarity, with only a further three managers sacked with less than 19 games left, with one of those occasions being a surely-doomed Mick McCarthy at Sunderland. On average, managers have left their relegation-threatened club with 22 games left in the season.

Going back to the four managers sacked with 8 games or less left in the league, three of these managers were 15^th and saw their former clubs avoid relegation, whilst Phil Brown left an eventually-relegated Hull in 19^th place. Whilst 75% is a decent success rate, there certainly haven’t been enough cases to definitively say that sacking a manager so late in the season has been a success. Fulham and Leeds in 2002/03 were saved with 2 weeks left in the season, whilst Fulham only avoided relegation by a point in 2006/07.

These clubs also enjoyed much larger point cushions from the relegation zone than Villa do now, despite their precarious league positions; Leeds (7 points clear), Fulham 2003 (6) and Fulham 2007 (4) could perhaps afford one or two bad results.

All this means that sacking Houllier would be an unparalleled move in recent Premier League years.

The predicament for the Villa board is obvious: if they keep Houllier, they preserve continuity but risk failure to recover form. If they sack Houllier, they may suffer an awkward transition period but gain the supposed lift generated by the new manager. (Though Kuper and Szymanski argue that this ‘new manager syndrome’ is deceptive and not attributable to the arrival of the new boss. They claim that clubs sack managers in a trough of form, and that the new manager benefits from the natural and inevitable rise in form. Hence the previous manager would have eventually benefitted from this rise in form eventually anyway. An interesting argument, but not one that I’m entirely convinced about.)

The club are an impossible situation, and perhaps this is why so many chairmen in the past have decided not to change status quo at this stage of the season, unless forced. Randy Lerner’s situation is made all the more difficult by club expectations, and I don’t think many envy him at this moment in time.

Note: I received a question on why Alan Shearer’s appointment at Newcastle with 8 games left in 2008/09 was left out. I’ve excluded this managerial change since it was  enforced upon Newcastle; Joe Kinnear stepped aside on health grounds. Similar cases have also been ignored in this analysis.