I could have sworn I've written this material down someplace, and perhaps I have I just can't find it.
Choosing content for, and arranging that content in, a portfolio is, well, it matters. I'm positive that this is something that is taught in depth in the right schools, but I have not myself run across any decent references about it. So, I'm making it up with my big fat brain, as far as I am able. A larger grain of salt than usual, therefore, is indicated.
First, some examples. Ming Thein, despite his many flaws, at any rate grasps that the idea of a portfolio, a coherent body of work, is an important object. He spends a lot of time making these things. And he is right. There, I don't say that enough. The boy's thinking hard all the time, and he hits on some stuff.
Take a look at this thing: studies in blue.
I don't much like these pictures, but that is completely irrelevant here. He's got a strong theme going on here, it's a coherent collection of very closely related photos. He's definitely got an idea, and he even tells us what it is: Magritte. Magritte did a bunch of these pictures with puffy white clouds in a blue sky, seen through some sort of cutout. The connection might not be obvious but it's there.
But here's the thing. You spend a ton of time constructing these pictures. You're carefully placing things in the frame relative to one another. You're creating patterns, breaks in the pattern, progressions from light to dark, all kinds of stuff. Within a single picture you're working hard. Outside the picture, the tendency is to simply pick the best ones and huck them into a bucket.
What Ming has done here is give us a bunch of variations on the same picture. What's he going for here? Is it skyscrapers? Hong Kong? Reflections? Blue sky? Clouds? What's he trying to do, here? The connection to Magritte is not made clear. The point is, he helpfully tells us with words, is the blue sky with white clouds, seen through frames. Had he changed the frame up, made it trees in one picture, skyscrapers in another, a window in another, the actual unifying point would be made clear. There are probably other methods he could have used, but that's what occurs to me in the moment.
There does not seem to me to be any particular progression here, not that there needs to be, but it's not a bad idea.
Evans and the people who are related to him did a thing in their books, where each picture is visually connected to the next. There's a repeated visual element, picture 1 to picture 2. In picture 2 there's a different element that connects it to picture 3, and so on, in (often) a nearly unbroken chain through the book.
Because I am a simple creature, and make small portfolios, I tend to go light-to-dark or something straightforward like that.
You want an organizing principle, it doesn't really matter what. You want an idea of some sort.
These are Magritte-like
The people in this valley are really poor
This car is an erotic hallucination
Flowers are basically about sex
You might find that to create the best portfolio, you have to discard some of the stronger individual pictures, and insert some weaker ones. Don't insist on sticking with your original top ten. You've probably got, at this point, a A list of the best and a B list of the almost-made-it's. When the flow of the portfolio feels off, think about what it needs, and consult that B list. There's likely to be a better fit for that one spot, somewhere in there.
Use it. At this point the pictures are secondary, and it's the portfolio that matters.