I think you're probably correct that fare dodging is likely to be a small proportion of total journeys and total revenue. I think though I'd dispute that that makes it a small problem: I would strongly suspect that fare dodgers are much more likely to be people who would behave anti-socially on trains, making other passengers feel less safe. And allowing even a minority to fare-dodge feeds into a wider sub-culture that encourages people to feel they can get away with crime generally. Also, remember that even if it's only something like 2% or 3% of revenue - that can easily make the difference between a company being profitable or not. Or pay for some small-scale improvements to stations etc.
A couple of people have mentioned Woolwich Arsenal. I think it's true that there are barriers there, but the barriers are usually open in the evenings. There are sometimes RPI checks on the bridge between national rail and the DLR. I would also mention that I often change from DLR to national rail there, and there always seems to be a few people touching in or out at the Oyster Reader as they change trains - and I can't see that there would be many legitimate reasons to do that (Having a paper ticket from Woolwich Arsenal to a destination outside London and using Oyster on the DLR is the only legitimate reason I can think of, but that must be a tiny number of people).
I think you make a very good point about the connection between fare evasion and about anti-social behavior.
Any investment in staff -- RPI, guards, manned ticket lines -- is the kind of expense that SE might very well want to avoid because it appears as a cost, but the benefits of having these people is not measured on a balance sheet.
I have made the point on here that on a commuter railway, which SE largely is in terms of passenger totals and revenue, fare dodging is a tiny issue if the significant streams of revenue and passengers are checked for revenue.
But it occurs to me that on SE which has 183.2m passengers a year, then most things will appear insignificant against such a huge figure.
Earlier on the string, the user commutor referenced TOC target of 4% baseline figure for fare evasion to be targeted down to 2.5%.
If you make assumptions that this has already been done, and that the 2.5% is directly relatable to passenger figures, then you get more than 3 million people dodging fares each year. Suddenly, what I described before as a tiny problem looks more serious.
But that is perspective!
Pretty much most days I see examples of fare dodging -- four teenagers waiting for an early evening train at Chiselhurst, for example, with two bikes, they haven't bought tickets they are on a lark! But they make no dint whatsoever in the passenger figures.
So it seems to me this is a matter of perspective.
From the SE perspective it has maximised revenue as efficiently as it thinks it can, and the rest is not worth the trouble and expense.
But from a passenger perspective, every time we see fare dodging it is a qualitative example of petty crime paying off and of unfairness.
