Copper's supply gap is widening. The case for probabilistic targeting.
More holes is not the answer
The obvious response is to drill more. It is also the wrong one. Exploration already costs more than ten times what it did a generation ago to find a deposit of the same size, and the easy, near-surface ore is gone. Drilling everywhere faster only spends the deficit's worth of capital on dry holes.
The constraint is not effort. It is aim. The question that decides a copper programme is not how much you drill, but where, and on what evidence you commit the rig.
Why targeting is a probability problem
The subsurface is read from sparse data: a scatter of holes, some geophysics, decades of survey work of uneven quality. From that, someone has to decide where copper most likely sits, at what grade, with what confidence. That is a question of probability, not certainty.
A model that returns a single confident answer hides the very thing that matters to the decision: how likely it is to be right, and where it might be wrong. A probabilistic model does the opposite. It ranks targets by likelihood, shows the evidence behind each, and makes the risk legible before the rig moves.
What this means for capital
A widening copper gap rewards the operators and investors who aim well. Every rig-day spent on a low-odds target is capital the deficit does not give back.
Probabilistic targeting will not conjure copper that is not there. It does something more useful. It tells you, honestly, where the odds are best and how much to trust them. In a market this tight, that honesty is the edge.
