Fooled by Randomness: Improving Decision-Making with Limited Data (Distinguished Lecturer Series)

Date: Oct 12th, 2016 Time: 11:30 – 1:00 pm Location: Calgary Petroleum Club For more info refer to this link: https://goo.gl/0SQjLu

Professionals routinely face the challenge of making informed decisions with limited data sets. Our exploitation of Unconventional resource plays has exacerbated this issue. We commonly refer to these resource plays as “statistical plays”, as large programs have provided repeatable year over year results. Competitive pressures and the desire to get to the right answer as soon as possible has driven the observed decision-making based on limited data sets. In an environment where horizontal well costs can exceed $10 MM and programs hundreds of millions of dollars, decisions based on limited wells have become our industry’s “money pits”. Development decisions are often made without due consideration for the representativeness of the data. Similarly, we frequently test new technologies with limited samples with the expectation that a simple arithmetic comparison of the average results can validate or refute their further application.

This talk presents the theory and utilization of aggregation curves as a pragmatic graphical approach to determining the uncertainty in the sampled mean relative to the desired average program outcome. The presentation will conclude with a discussion on the use of sequential aggregation plots as a graphical approach to validating the representativeness of our forecasted results versus based on limited actual results.