I'd really like some cold hard stats (life expectancy, infant mortality, (and more importantly, mortality-corrected life expectancy), perhaps some PPP) to back up this assertion that there has been no improvement in the standard of living from 1000 to 1800.
I'd really like that too. Alas, there are no "cold hard stats" of anything until the late 19th C. There are guesses and conjectures, and extrapolations from examples. And that's the best we can do. Some, like Maddison, are almost purely arbitrary. Others are more cautious.
Private wills are the best source of knowledge about average living standards. And they reveal that people didn't have much of anything - some cutlery, some tools, a couple of pieces of homemade furniture, a couple of sets of homemade clothes. And that didn't perceptively change.
You can also draw inferences from things like contemporary paintings and other descriptive accounts of daily life.
Estimating productivity depends on knowing how much a peasant had to work with - there's a technical upper limit on how much you can get with x tools on y acres of land with z type of crops. Deduct the rates of rent, taxation, tithes and other extractions that you might be able to find in manorial or magistrate's accounts, and that's a rough guesstimate of what he had left over for himself.
Maddison isn't nearly as careful. He comes up with his "growth" statistics by assuming a unduly low subsistence baseline to begin with, and just imposing that baseline everywhere he doesn't have data for. So it "looks like" growth or divergence, because of that.
- 3