An economy’s inventory of produced inputs is both complex and simple. Maintaining and improving upon congeries of productive inputs is an indispensab… - Paul Samuelson

Econometrics may be defined as the quantitative analysis of actual economic phenomena based on the concurrent development of theory and observation, related by appropriate methods of inference.

Let us summarise the lessons learned.
1. Though achieving private-property perfect competition will achieve productive efficiency, moving toward it may have made matters worse. And, in the land-enclosure problem, the transition must make things worse before they get better when lands are virtually homogeneous.
2. In the realistic case where lands differ in quality, partial privatisation may also worsen efficiency - even though ultimately efficiency will be enhanced by the deregulation process.
3. However, in the realistic case, an optimal choice of lands to be first privatised could be first to improve the commons. The rule for a 'perfectly-discriminating deregulator' to follow is evidently this: privatise those lands first whose 'imputed labour share'- as measured by the MP/AP fraction- is the lowest. That way, the first bit of deregulation can do more good than harm, as labour is transferred away from low marginal-product locations; that way, if there is inevitable transition harm, it can be kept to a minimum. There would seem to be a presumption that things get worse before they get better in such a programme of moving toward laissez-faire.

I think the acceptance of "mathematical expectation of utility" or its "arithmetic mean" was an unthinking carryover from the mathematical theory of the law of large numbers as applied to asymptotic processes.