How often does the ECJ follow Advocates General? Or should that be CJEU?

I say 'fact', but on closer inspection it seems that the status of the 80% figure is rather more lowly than 'fact'. It turns out that it languishes somewhere between rumour and conjecture.

So here is OUT-LAW.COM's confession: we have been sloppy. We have been using – for a long time – a figure that we took as received wisdom and never actually checked. We won't be using it again.

So is it accurate? And where did it come from in the first place? We did some digging, and we are none the wiser. Close observers of the workings of the ECJ have filled us in, though, and this is what we found.

First of all, the figure has been hanging around for years, but it doesn't appear to have any official original source.

It is most likely to have been an informal estimate that suddenly caught on. The keen-ness of case observers to adopt a figure is understandable – any case referred to the ECJ will be a high-stakes issue, and months can elapse between the publication of an AG opinion and an actual ruling.

That those involved in, and reporting on, a major and controversial issue want to know the likelihood of a published opinion turning into judicial fact is no surprise. And in the absence of official statistics on this, it is understandable that any figure will be jumped on.

So is it accurate? Well, first off we have to look at whether it's possible to even conceive of an accurate figure at all.

A ruling might agree in part with an AG opinion, and disagree in other parts. It might come to the same conclusion about which party should be favoured by its ruling, but for very different reasons, fundamentally disagreeing with the AG's analysis along the way.

How should these nuanced, complicated inter-relationships between AG opinions and rulings be interpreted? Do they 'agree'? In many cases it is impossible to say.

There is also a problem in how the statistic might be used. It is inevitable that the cases that draw the most interest will be the ones that are most controversial, or most finely balanced.

Because of this controversy and fine legal balance, these are the cases in which any two experts will be most likely to disagree, so are the cases in which the AG opinion and the ECJ ruling are most likely to diverge.

The result of this is that the cases attracting the most interest and the most reliance on the 80% speculative rule are actually the ones least likely to hold to it, the ones in which there is more likely to be disagreement between the two parts of the Court system.

The situation was further complicated by the Treaty of Nice, which came into effect in 2003. Since then AGs have stopped giving opinions in all cases, and now only publish views on cases involving new questions of law.

This, naturally, once again skews the figures. The safe cases on which judges will most likely agree are taken out of the statistical equation altogether, and this is likely to make the 80% figure even less reliable, since the AGs focus on cases where disagreement is more likely.

So what is to be done? Well, OUT-LAW.COM will stop using the 80% figure since we cannot find evidence to back it up. Our ECJ moles tell us that it is certainly the case that the AG and ECJ agree in a majority of cases, so 'a majority' it is.

That is one ECJ question sorted out. Or should that be CJEU? This is another European Court issue we have wrestled with recently.

Earlier this year we had a crisis of terms in relation to Europe's top court. It changed its name to the Court of Justice of the European Union. We dutifully stopped referring to the ECJ and started referring to the CJEU.

It turns out we were wrong there as well. Actually the Court had just changed its name from the Court of Justice of the European Communities to the Court of Justice of the European Union. So it had never, actually, been called the European Court of Justice (ECJ) at all.

This means that the ECJ is as valid an acronym now as it was before the change. Which is why we switched back to referring to the Court in that way. It might not be exactly right, but at least it is wrong in line with history, and everyone knows what we are talking about.

By Struan Robertson, editor of OUT-LAW.COM. You can follow Struan at Twitter.com/struan99.