As I reached the top of the stairs at Embankment tube station, heading around to the ticket barrier, a voice behind me asked me to stop.

Had I dropped something?

As I turned around, the man produced a police ID, and said he wanted to ask me some questions.

Odd. I asked to see the police badge a second time. It looked authentic enough.

I was told that I had been observed as exhibiting possibly suspicious behavior, so the officer would like to know what I was doing and where I was going.

It was about quarter past six on a Thursday evening. I was taking the tube to Charing Cross station, so that I could get a train home, I said. The sort of thing that a lot of people were up to around that time in the evening, I imagine.

What was the suspicious behavior, I asked?

It turned out that I had been spotted in the very act of changing my mind.

You were headed in one direction to the Jubilee Line escalator, but you spotted a uniformed officer and went the other way, I was told. That had formed the basis for a theory, he explained, that I might well be a pickpocket, looking for a victim to target. Apparently, while cunningly avoiding surveillance by the law too.

I had indeed changed direction. After taking several steps towards the Jubilee Line escalator – and my most direct route home, via London Bridge – I had walked across to the much emptier platform on my right, realising that I could spend less time on the tube, and avoid those long Jubilee line escalators too, if I headed to Charing Cross.  

I hadn’t seen a policeman, just a lot of wet commuters with coats and umbrellas.

Perhaps I was in two minds because Westminster wasn’t my usual route home. I explained. I had been at the BBC offices on Millbank after doing an interview for Radio 4. Coming out into heavy rain, I had ducked into Westminster tube rather than walking down Whitehall to Charing Cross. (Historian Anthony Beevor and I had been discussing the government’s plans to commemorate the centenary of the Great War. We weren’t in the same studio, so I can’t use Professor Beevor as any kind of control in terms of his chances of having been stopped at Westminster tube on his way home.)

Was that suspicious enough to be stopped, I wondered? The explanation for stopping me meant, I realised, meant that the two police officers had probably come from the ticket hall at Westminster, and followed me on and off the tube at Embankment. Perhaps the slightly lazy one stop journey added to their suspicion that I was targeting my next mark.

Some people could hang around on the tube, for no reason, perhaps for eight hours at a time, in order to carry out criminal activities, hence the activity of stopping some people.

I had, I was told, given a perfectly legitimate reason for being on the tube network that evening, and so could go on my way.

Before going our separate ways, it did seem worth a couple of minutes more dig into how to carry out that tricky job of protecting us all from these undesirable criminals without too much inconvenience to law-abiding citizens, and in a way which Londoners of all types could trust.

My wife would be amused, I said, at my being taken for a pickpocket, I said. Why? They asked. She would doubt that I would make much of a pickpocket, I suggested, though I couldn’t be totally sure, having never tried it.

It felt to me that, even if I were a mastermind pickpocket, that I pulled off quite a good impression of a commuter. I was even, for me, at least a tiny bit more spruced up than normal, because my colleagues had rightly insisted I put a tie on to go and talk about the solemn issue of Remembrance on the BBC news channel about half past three. But perhaps the criminal masterminds can pull off the commuter look too, making decisions about who to suspect rather trickier.

So how did they decide? Was there any profiling involved, I asked. Of what kind?

Were single Asian males in their thirties travelling alone more likely to be stopped, I asked.

No. We just try to base it on behavior, I was told.

I am never quite sure whether or not to believe this official "no profiling" policy. I explained that my uncertainty about that is not based on a great deal more than my (inconclusive) anecdotal experience of seemingly happening to be stopped more often than my colleagues in think-tanks and journalism – though, only, as it happens at transport hubs. I am only, however, talking about four times in the last ten years overall: once at Stanstead airport, once at Luton, once on the Eurostar platform when it used to be at Waterloo, and now by the ticket barriers at Embankment.  

Any think-tanker knows that one could never derive any proof from such as small anecdotal sample.

But it has led me to ask the same question, and always to receive the same reply.

When I asked what details I could have from the police officers, I was told it was possible to record the incident, so I thought that I would like to do that.

Looking over the officer’s shoulder as he used his rather smart e-device, I questioned his initially selecting the option “white”.

I don’t think “white” is the right option for me, I said.

“That’s what I would have said you looked like to me”, he said. 

I wondered if that was over-compensating a little from one potential inference of the earlier profiling question.

But I thought non-white was the right answer if the question was ethnicity, while British was fine for nationality. That threw up a much more extensive menu, for “self-defined ethnicity” where I could choose an “M3 Mixed White/Asian” to reflect my Indian-Irish parentage..

That was one small insight into the stuff that the official statistics are made of.

Recording the stop seems to involve quite a lot of information – including giving your name, date and place of birth, and your home address if you filled out the full thing.

I wasn’t entirely sure that the opportunity would necessarily have been spontaneously offered if I hadn’t specifically asked for some type of identification or record, and opened up a broader policing policy discussion.

And the initial instinct to flag me for the computer as "white" offered confirms the insight that any statistics can only be as accurate as the source material.

But the officers were professional and polite throughout, and not necessarily any less so before I had mentioned coming out of a BBC interview either.

The hand-held printer – rather in the style of the Apple Store – took two or three minutes to print out, so we were able to have a bit more of a chat about policing, trust and indeed the politics of the think-tank scene. I established that I could do whatever I wanted with the receipt of the encounter – if, for example, I wanted to put it on Facebook – and was even given the Twitter account of the British Transport police. (Another culture shift there, as it tweeted me back this morning). 

The receipt shows that it was printed eight minutes after the stop, but we seemed to have covered a fair amount of ground.

It showed that the “power” was “Rec 61 – Stop and Account”, and the outcome was “no further action”.

The policeman asked me how I had found this experience, personally.

I said that it had been odd – but they had shown a professional willingness to answer my questions about it.

Stop and account, if you will, but it needs to work the other way around too.

I had found a rather more surly response when asking similar questions when being interrogated about my business at an airport. The policeman had found the same when flying to the US, so we turned out to have that in common too.

I said that I understood that the police have a job to do. My instinct was to be sympathetic to them, I said, because I felt they had tried to come quite a long way over the last fifteen years, certainly from some of the controversies that I had followed closely when I had lived in Eltham, on the Well Hall Road, during the inquiry into the Stephen Lawrence case. So my instinct is that progress has been made – and that more progress is possible.

The question mark on which people want assurance have is that policing is effective, transparent and fair.

I think that may be why questions such as who gets stopped – and why – can still matter.