The peak of my performance as head of geography came in August 2009. The department’s GCSE target grades had been comfortably exceeded, and I felt an overwhelming sense of pride. My colleagues and I basked in the glory. Contrast this with the despair of August 2016: we fell comfortably short of targets and, in the inspectorate Ofsted’s terms, were deemed to be “requiring improvement”.

So what does this mean? Am I the outstanding subject leader of 2009 or the abject failure of 2016?

As it turns out, I am neither.

In 2009, my department had the perfect statistical recipe for success: a large group of motivated Nepalese students had arrived with little grasp of English and low target grades, reflecting their EAL (English as an additional language) status rather than their academic potential.

In 2016, on the other hand, we had a disproportionately high number of non-attenders. With a transient classroom population and a number of challenging individuals whose behaviour resulted in exclusion, the validity of the targets was undermined.

Herein lies the problem with forecasting individual performance based on national prior attainment. So does that mean heads of department should stop paying attention to target grades? Mick Hill, the recently retired executive principal of the Northallerton and Catterick Federation, is sceptical about their value.

“It is absolutely the case that using numerically averaged targets removes the individuality of students and schools,” he says.

“There are potentially thousands of intricacies affecting a young person’s life that affect the accuracy of that individual’s targets.”

The reliability of targets has also been questioned by Sean Harford, Ofsted’s national director for education, during this transitional period of GCSE reform in particular.

“It is actually quite difficult to statistically target outcomes in ‘steady state’,” he says. “But in the situation where you have a change in GCSEs and comparable outcomes, it has a level of inaccuracy not appropriate to make it a worthwhile practice.”

And, Harford continues, accuracy is not the only concern. Target grades can also have potentially damaging implications for pupils.

“I am aware of schools where students are given their key stage 4 expected grades when they arrive in Year 7,” he says. “It must be pretty demoralising to be told as a child that if you work hard, you are on for a [grade] 2.”

In March 2017, the inspectorate duly released an inspection update with a directive to stop asking schools for target grades. “We are taking the heat out of that process by not asking schools to make predictions that can’t possibly be accurate anyway,” says Harford.

With Ofsted backing a move away from target grades, many schools have replaced them with “minimum expected grades” (MEGs) instead. MEGs are often based on data from the analytics group the Fischer Family Trust (FFT), which uses KS2 data to help schools create estimated benchmarks for achievement. But in practice, it seems that there is often little difference between MEGs and target grades. MEGs have become synonymous with an entrenched culture of continually “raising the bar”, which escalates pressure in classrooms.

While all teachers strive to maximise attainment, the targets by which they are judged need to be realistic, says maths teacher and former union representative, Matthew Marlow. “A minimum expected grade that is based on FFT 20 [the category of schools that make greater-than-average progress] is not a minimum expectation because it is based on the top 20 per cent performance,” he explains. “If you are having a minimum which is already aspirational, then you are using the words, but they don’t correlate to the numbers.”

So should we do away with MEGs, too? With schools judged on their Progress 8 scores, there is arguably little need for individual target grades. But if we did scrap targets completely, what would heads of department then use to measure the performance at subject level? After all, there is still a need for professional accountability.

“I certainly feel more accountable because of statistics and I think that is only right,” says Colin Scott, headteacher at Risedale Sports and Community College in North Yorkshire.

FFT founder Mike Treadway suggests that the problem lies not in the provision of statistics for achievement, but in how schools are using that data. “The key point is that we don’t provide a target grade - it is an estimate,” he points out.

“The estimated grade should always be used as a starting point.”

From this starting point, Treadway recommends actively involving students in the process of setting target grades and explaining what the data means.

“One method is to simply give a target of grade 5 [recognised as a ‘strong pass’ in the reformed grading system],” he explains. “Another way is to say the majority of people from your starting point got a grade 5, but 10 per cent got a grade 9; which do you want to aim for?”

Alternatively, schools can take another approach. At Huntington School in York, target grades are not shared with pupils, freeing them from the associated pressures and restraints. “We have been liberated by the decision not to inform parents and students about their target grades,” says headteacher John Tomsett. “With dedication, individuals can make progress beyond what anyone, including themselves, could have imagined.”

Whichever route schools take, it seems that there’s a balance to be struck between ensuring accountability and relying too heavily on a number that is really nothing more than an educated guess.

Neil Fatkin is a freelance writer and a former head of geography