Tornado plot showing the change in the expected value (EV) of an opportunity from factors of division/multiplication, with a base case EV of +$120 million (MM) and base case values of cost of failure = $100 million, reward of success = $1000 million, and chance of success = 0.2. Tornado bars show the effects of multiplying (or dividing) the base case values by factors of 1.1, 1.5, 2, and 2.5.
(A) Standard decision analysis cycle (after Matheson and Howard, 1968; Bickel and Bratvold, 2008). (B) Pot odds approach, which may enable more efficient decision making. EV = expected commercial value; n = no; Pg = chance of geological success; RoR = rate of return; y = yes.
Comparison of the chance of geological success (Pg) estimate against the pot odds for three prospects (a, b, and c), which all have the same expected commercial value. The values shown are, in order, the mean reward of success, mean cost of failure, and current estimate of Pg. MM = million.
Estimation of the chance of regret based on an estimate of prior (current) chance of success and estimates of what that could change to in light of bad news or good news. The x-axis represents the estimated chance of geological success (Pg); the y-axis shows the passage of time and the acquisition of new information. A decision and commitment to proceed has been made on the basis that the current Pg estimate exceeds the pot odds threshold. If the new information brings good news, the Pg estimate increases, and vice versa. We will regret the decision only if the revised Pg estimate falls below the pot odds threshold (gray shading). (A) The general case: P1 represents our current (prior) estimate of the chance of success, Pg, prior to gaining new information. P2 and P3 represent the possible future values of the Pg estimate, given that new information is bad news (P2) or good news (P3); x is the chance of bad news, and (1 − x) is the chance of good news. (B) Values used in the worked example described in the text.
(A) Unrisked exceedance curve (black) and probability density function (gray) showing the range of volumes predicted if the success case model is correct. In the event of geological success, the model is correct. The minimum volume in the success case is the P100; there is an n% chance that the volume will exceed the Pn value. (B) Risked curves, showing the absolute chance of exceeding the same volumes. (C) Risked curves for a prospect for which some of the geological failure cases may contain hydrocarbons. Pg = chance of geological success.