Palace too swift to blame boss De Boer

Stanley Baldwin, a British prime minister in the 1920s and 1930s, said some newspapers exercised “power without responsibility, the prerogative of the harlot throughout the ages.” Football club managers, it seems to me, accept responsibility without power, which must have been the prerogative of the idiot throughout the ages.

Crystal Palace sacked Frank de Boer after four Premier League games. Chairman Steve Parish had said football is a “results-based business”. Was it De Boer’s fault that Palace lost four times without scoring?

I have likened changing manager to changing left-back. It might make a difference, but usually not a lot. A left-back should be treated fairly, though. You should not sell him because the team are not scoring. And you should not sack your manager either because the team are not scoring, if they have been creating chances. And Palace had.

According to the BBC’s expected goal stats Palace should have outscored three of their four opponents. The exception was at Liverpool, where most teams realise they should usually lose. Expected goals figures tell us how many goals a typical team would have scored and conceded with the same quality and quantity of attempts for and against.

Palace could have had nine points instead of none. Was it De Boer’s fault that defenders made silly mistakes and forwards were less accurate than usual with their shots and headers?

A manager’s job is to prepare players so they have the best chance of producing a good performance. De Boer was not the only one to think that Palace were by far the best team at Burnley in what turned out to be his last game.

De Boer said: “I am disappointed about the result but not about how we played… Usually if you create that many chances you will get what you deserve, and we didn’t.”

Burnley manager Sean Dyche, honest as ever, said: “Palace were the better side. They played well throughout the game.” Match of the Day 2 pundit Chris Sutton said: “Anyone who saw the game will know they should have won it – they created so many big chances and they did everything but score.”

Parish must be confident that at the end of the season he will be able to say results improved after De Boer was sacked.

Palace will not lose every game without scoring, even though they were beaten to nil again on Saturday under new manager Roy Hodgson. But this is an optical illusion that deceives many chairmen. Results generally improve after a manager is sacked, but bad results generally get better anyway even when a manager is not sacked.

I noted every sacking during the previous 20 Premier League seasons. Most fired managers had lasted longer than De Boer. On average managers were sacked 18 games into a season.

Before their manager was sacked teams won 25 per cent of their games, drew 26 per cent and lost 49 per cent. Afterwards they were seven per cent more likely to win and seven per cent less likely to lose. They won 32 per cent, drew 26 per cent and lost 42 per cent.

But there is something you cannot see in those figures. Other teams with similar results after 18 games of a season who did not sack their manager achieved similarly improved results in the last 20 games. Hodgson should eventually oversee better results for Palace, but probably no better than De Boer would have done.

More by Kevin Pullein

Why Mourinho is typical of Champions League group stage

Expected-goals stats provide a crucial sense of proportion

Why the lion's share of possession is no guarantee of victory

Champions rarely end up on the route to glory in Europe

Interviews are just a lot of hot air

When Crystal Palace appointed Frank de Boer as manager they said he had excelled in interviews. Seventy-seven days later they sacked him. De Boer was perfectly capable of managing Palace well. Even so my heart always sinks when I hear a football club talk about how their new manager stood out in interviews.

Interviewers, in all industries, are good at spotting candidates who do well in interviews. They are hopeless at identifying who would be best in the job.

Nobel Prize winner Daniel Kahneman discussed interviews in his book Thinking, Fast and Slow.

As a young man Kahneham was asked to redesign interviews for the Israeli army. They knew that their conventional interviews were useless for predicting which recruits would do well.

Kahneman listed six characteristics that he thought might be predictive of success. For each characteristic he wrote a set of questions designed to elicit factual information.

Interviewers were told to ask the same questions at every interview. After each set of questions they would rate the candidate on that characteristic with a score between one and five. At the end of each interview they would add up the scores. The candidates with the highest totals would be recruited.

Interviewers were appalled. They felt they were being denied the freedom to use their judgement and weigh-up candidates. But the army’s research had told them the interviewers’ judgement was useless. And the interviewers were in an army so they did as they were told.

Kahnheman said: “The sum of our six ratings predicted soldiers’ performance much more accurately than the global evaluations of the previous interviewing method, although far from perfectly. We had progressed from ‘completely useless’ to ‘moderately useful’.”

Today's top sports betting stories

Follow us on Twitter@racingpostsport