The issues you raise are in fact two of the central tensions in the study of history, distinct, as you suggest, but in some ways related. At a discplinary level, the central concerns of history are to identify, describe, and explain change over time. Given the inherently fragmentary and subjective nature of our source materials, a series of problems present themselves; the problem of people writing less about things that aren't changing and the problem of "winners" writing history are different manifestations of the nature of our data. I'll treat them separately.

So, yes, it is true that people record less about things that do not change or which they take for granted. In this sense, things can be hidden in plain sight. My own area of research, for example, deals with bread. Bread is obviously the cultural and caloric foundation of the Western diet, and has been for a very long time. As such, it was the single biggest staple of diet, and it did and still does carry tremendous cultural significance. But, because of its ubiquity, there was little need to write about it, and when it was written about, it was often done in a metaphorical way. So, when you encounter a nineteenth century text noting that one might "earn his bread," it's meaning is ambiguous. "Bread" here means "food" or even "sustenance" generally (I suppose as a metonym), but in practice that sustenance would in fact include a great deal of bread.

This presents a problem for all sorts of things, including the very basic relationships that structure society: between genders, races, classes, lord and vassal, families, master and apprentice, and so on. For our subjects (and for us as well), these relationships constitute the very fabric of life, so deeply taken for granted that they don't need to be expounded on very much. Who today, for example, needs to sit down and explain exactly what their relationship to their siblings is? And, given the ubiquity of these relationships combined with the inherent diversity of human experience, if everyone sat down and wrote descriptions of their relationships to their siblings, they would not all be the same. There would be some common themes, most definitely, but I think we would see a tremendous variation. As a historian, this is a considerable difficulty in reconstructing the lives of past peoples; social and cultural history is hard. Those two terms actually present two approaches to reconstructing these relationships.

Social history, broadly, is more materialistically oriented. It looks at how things were in a material sense: how much money did people make? What were their working conditions like? What kinds of houses did they live in? At what age did most people get married, and how many children did they have? These kinds of investigations were very popular after World War II, and remain so; they tell us a great deal about the past, and form an invaluable base of material knowledge. They do have their limits. How much, for example, can we really know about people's lives by their 'real' income? A social history of diet might tell us what people were eating, who cooked it, what patterns to consumption existed, but it won't tell us what it meant to those people to eat things. In particular, social history is great at identifying economic indicators of the past, but it's not so great at explaining why people make the choices they do. People do not always behave in ways that maximize their material benefits, after all.

Cultural history emerged as a critique of social history very much on those grounds: at a fundamental level, cultural history is concerned with meaning (we might contrast this to social history's concern with experience). It asks how the categories through which people experience their lives are formed, and what they mean. So, a social history might identify and describe the work that people did, while a cultural history would ask what that work meant to them, typically in categories of identity; so, it might ask, "What did it mean to a man to be a 'breadwinner,' or a factory worker, or an apprentice? What did it mean to women who entered the workforce?"

Now, this can explain how historians approach history in certain situations, but it doesn't explain how we get around the fact that for things that are mundane, people don't always write everything down. To get at those very basic structures of everyday life, then, historians often focus on moments of crisis. Crises generate more documentation, and because they are moments at which the social order threatens to break down, the foundations of that order are sometimes laid bare. So, for example, if you're interested in the lives of peasants, one important source for study will be peasant revolts. In those revolts, we get a chance to see what they're angry about and therefore, by implication, what they expect the normal state of affairs to be.

This brings us to the other major problem you identified, which is that the 'winners' write history. Let me start by saying that that formulation is far too simple. Sure, it applies in some cases, but by and large that's not a particularly useful way for understanding the production and preservation of textual materials past and present. A more useful way to think of it to note that the production of textual materials--written accounts of any form, from diaries and journals to newspapers to government reports to advertisements--reflect the power relationships of the society that made them. The powerful groups in society will use texts as means of articulating and enforcing their power, as a government builds, maintains, and preserves tax records, for example. This doesn't mean that ALL texts ever written are part of a power structure that oppresses subordinate groups--people can and do resist in all manner of ways--but the documents produced in support of power are going to be numerous and well preserved. As such, they will frequently form the foundation of our historical investigation.

To get around the inherent power relationships built into the documentary record, there are a set of strategies for what we might term reading between the lines, or reading along or against the archival grain. A classic example of this is Ranajit Guha's Elementary Aspects of Peasant Insurgency, a founding text for subaltern studies. Guha looked at peasant revolts in British India, the reports of which were one of the largest collections of documents dealing with Indian peasants. These peasants were a subordinate group in Indian society, and produced comparatively little of their own documentation; so, if you wanted to study them, reading the reports of the people who were trying to suppress them was one of your only options. Before Guha, historians very friendly to British rule and relatively uncritical in their analyses read these reports at face value. They took for granted the explanations of British administrators that these peasants were foolish, engaging in irrational revolts for reasons that were beyond the ken of rational administrators. Guha read these reports in a more critical way, looking for the voice of the subaltern behind the British administrators, and finding that the world of the Indian peasant was far more complex than anyone had suspected. These peasants were revolting for reasons that made perfect sense in their world and according to their value systems; it's just that their world and value systems were invisible to the British.

Now, Guha's work is just one example. But, it's an important one, and it speaks directly to your questions.

Winners writing history is news to Athens in the Peloponnesian War, the Mongols in Europe, Vietnam, the US Civil War in the South, etc.

The writers write history, even if they lost.