The Baker Hughes rig count is the oldest and most-watched activity indicator in the oil patch. It has been published in some form since 1944, which makes it one of the longest continuous data series in the industry. That longevity is exactly why it's useful — and also why people read more into a one-week wiggle than the number can bear.
What the count actually measures
The rig count is a tally of active rotary rigs — rigs that are turning to the right, actually drilling, during the reporting period. That's a narrower thing than it sounds. It does not count rigs that are stacked, rigs being moved between locations, or wells that have been drilled but not yet completed. It's a snapshot of rigs working, not of total industry capacity or of production.
That distinction matters because production and rig count can move in opposite directions. Thanks to drilling efficiency and the backlog of drilled-but-uncompleted wells, output can keep climbing even as the count flattens or falls. The count tells you about drilling activity, not directly about barrels.
The one-line definition: the rig count is the number of rigs actively drilling in a given week. It's a measure of how much new drilling work is happening — nothing more, nothing less.
Why it's a leading indicator
The count is valued because it sits near the front of the activity chain. When operators decide to spend, they contract rigs and start drilling before any new production shows up months later. So a sustained change in the rig count tends to precede changes in output and in field employment. Rising counts signal operators committing capital; falling counts signal them pulling back.
That forward-looking quality is why it ripples into the labor market and into rig day rates. More rigs working means more demand for crews and consultants; fewer rigs eventually loosens that demand — though, as we've written elsewhere, the most experienced seats stay sticky even when the count dips.
Oil-directed vs. gas-directed
One of the most common misreads is treating the total as a single signal. The count is split by what the rig is targeting, and the two halves can tell completely different stories.
| Split | What it tracks | Why it can diverge |
|---|---|---|
| Oil-directed | Rigs drilling primarily for oil | Driven by crude prices and oil-basin economics (e.g. the Permian and other plays) |
| Gas-directed | Rigs drilling primarily for natural gas | Driven by gas prices, storage, and demand — a separate market |
Because oil and gas trade on different fundamentals, the oil-directed count can be climbing while the gas-directed count falls, or vice versa. A flat or modestly changing total can hide a big rotation underneath. If you only watch the headline number, you'll miss the actual story — always look at the split. The count is also broken out by basin and by drilling type (vertical, horizontal, directional), which adds further texture.
Read the trend, not the tick. A single week can move on weather, holidays, rig moves, or noise. What carries signal is the direction over several weeks against the year-ago level. Treat any one Friday's number as a data point, not a turning point.
How to use it without getting fooled
Put the pieces together and a sensible reading routine emerges. Look at the multi-week trend rather than the weekly delta. Separate oil-directed from gas-directed before you draw any conclusion. Compare against the same week a year ago to strip out seasonality. And remember what the number excludes — stacked rigs, the DUC backlog, and production itself.
Finally, because the count is published on a regular cadence and revised over time, always check the figures against the live source rather than a number you remember from last month. The rig count is a genuinely useful leading indicator when you read it as a trend with its parts broken out — and a misleading one when you treat a single week's headline as the whole truth.
Watching the count to plan crews?
When activity firms up, the binding constraint is people. The reference library shows reference guides by basin, ready for review.