This week, I sat with Reception during lunch. This is what I observed:

One child spent the whole time speaking to me in a language they had made up which they called ‘Moo’ (which involved a lot of head shaking, until they got very dizzy and I had to steady them). 

Another kept making shapes with his cutlery. Whenever I asked him what he was doing, he looked at me and laughed manically. 

One little girl told me that she loved unicorns and that they "actually poo rainbows in real life."

And finally, another child kept counting "evil peas" – crushing some and eating others. I could see no rhyme nor reason behind his methodology.

So, who actually thinks that testing five-year-olds is a good idea? I have not spoken to a single person working in a school who does. Is there some amazing research paper out there that makes the argument or another country with huge success in testing that does it? Am I missing something fundamental?

I'm not talking about finding out what five-year-old children know and don’t know. From experience, this takes time, fluctuates and is complex. The current system is pretty good at observing and planning and reviewing practice to enable early years children to learn and make progress. It could be streamlined, in my opinion, but either way,  it’s a very strong system. Early years practitioners are some of the best in teaching.

Personally, I have no problem with Oftsted's Bold Beginnings report, and I do see the balance between good practices in early years as one of the big steps being made in education right now.

Baseline testing is bonkers

But testing in a formal way in which we draw some sort of national consensus or baseline for children entering Reception and compare this when they get to Year 6 (nationally!) is truly bonkers.

Sometimes it feels like the people who have the power to make things happen in education have never seen an actual child, never mind teaching 30 of them each and every day.

The idea that we can find some sort of rational understanding of the abilities of five-year-olds based on a test is bordering on the ludicrous.

Ask any early years practitioner and they will be the first to show you the vast variety of skills and knowledge in their classrooms. They will also show you massive gaps between summer- and winter-born children, vulnerable children, developmental stages, family circumstances, early trauma and the varieties on being a human being on the evolutionary track. They don’t need a test to tell them what the children can or cannot do. Every day they find out through well-planned and well-thought-out experiences, activities and teaching – that is at the heart of their job.

What is so sad is the low-level of – or indeed, lack of – trust.  Rather than allow the profession to lead on assessment, we have to have national tests to tell us what teachers cannot be trusted to tell us.

There may have been an era in education when teachers didn’t want the best from their students. A time where high-stakes accountability would have been the kick up the backside the profession needed. Those days are long gone.

Every teacher I meet, every school leader I talk to, wants the very best for the children in their schools. Why wouldn’t they? They care, they have far more skill, knowledge, and desire than any policymaker or politician I have come across (which to be fair is pretty limited) and yet their voice in the power of change is lost.

Ignoring the advice of teachers

It seems that a few people (with no real teaching experience) have all the power to make massive change and ignore the advice of the profession. That is wrong. That must not go unchallenged. There is no trust in our system because it is seen in the business mould and the principles are detached from the teaching realities.

The introduction of the baseline assessment and using this as the focal point for future school accountability is policy politics at its very worst. The way headteacher accountability has grown over the past five years is proof of the damage it will do. Our system is led by accountability that can be measured, rather than professional knowledge and skills and doing what is right. I cannot say this enough.

The basic strategy for a headteacher to survive this baseline accountability is a simple one though. Swallow your integrity (and try not to look at yourself in the mirror too often) and make sure the children doing the baseline do REALLY badly – unless you run a nursery, in which case you are up a creek without a paddle).

If the score is low, the progress will be all the better in Year 6. And no one can blame you for the abilities of Reception children who you have not taught. Well, not yet anyway. The fallout from this will be an influx of "outstanding" nurseries going into special measures and the government will lament the lack of academic rigour in our nurseries. This will lead to a pre-birth screening test based on DNA that will set a baseline for GCSE results.

(I know, I’m just being silly now.)

One of the side effects of crack cocaine (popularly referred to as 'base' in the 1990s) is "repeated and compulsive use". A little like our education system's obsession with testing.

We're addicted to using numbers to measure the quality of our education system; the effectiveness of our teachers. The problem with numbers is they present all sorts of very complicated narratives.

We could go into school funding numbers, exclusion numbers, teacher retention numbers – I am quite sure the Department for Education has many press releases explaining to us why there is more to the numbers and we need to look beyond them. It seems at times that it is playing the gameshow "Numberwang" – featured in the comedy sketch show That Mitchell and Webb Look – in which contestants call out seemingly random numbers, which are occasionally met with “That’s Numberwang!”. 

We're familiar with the game show format but Numberwang makes no sense at all. This is how I view the idea of testing five-year-olds. For someone who has worked successfully in education for over 20 years and been a headteacher for a long time in many schools, testing five-year-old children makes NO sense whatsoever. The results will be absolutely flawed from the start.

When the stats from the baseline come out, I will be calling out after each statistic,  “That’s Numberwang!” because trying to quantify children at this age is completely illogical.

Brian Walton is headteacher of Brookside Academy in Somerset. He tweets as @Oldprimaryhead1