Rig counts don’t add up like they used to

Do you know what day of the week it is? Or what week it is? Month? No, not those, not the Thursdays and the third week of July, not that stuff. I mean the out-of-control catalogue of diseases and feelings and occupations that have declared their own “day”. Did you know that this past Monday was “World Chimpanzee Day“? Or that the week of July 14-18 is “Dust Safety Week“? Or that according to a hair-brained site named “Always the Holidays” the month of July has 43 “official” awareness designations such as  Scleroderma Awareness Month and Dog House Repair Month, or that lost between National Bison Month and National Hemochromatosis Screening and Awareness Month and National Horseradish Month (all official July markers, not making this up) are some actual semi-plausible candidates like Make a Difference to Children Month? Who organizes all this crap? Why? “When everyone is special, then no one is special,” said Syndrome in the Incredibles, and that guy made a lot more sense. If you’re a charity and fighting for attention ok that’s one thing, but I hate to tell you that you’re going to have to come up with something a little better than being a grain of sand on that crazy beach.

If you’re reading this site though, there’s one that you know for sure and it never changes and everyone loves it: Friday is Rig Count Day. Every Friday, Baker Hughes releases their rig count activity statistics. Watchers wait for the weekly arrival like a cargo cult staring at the horizon; like a primitive culture observing a total eclipse, a mix of reverence and impotence before a greater power they can never understand.

The rig count has that effect, or so it seems, because we know it is a powerful indicator of something, and could indicate great riches or great calamity, but we’re never quite sure which. What is so vexing is that it should be straightforward, and the fact that it isn’t is very hard on everyone’s head.

More active rigs should mean increasing production, and historically that has been more or less true. There is always a time lag, as in how long it will take from a well’s drilling to when it actually goes on production.

But then companies started getting somewhat strategic, drilling wells but not completing them, which was quite a breakthrough in the quality of thinking as opposed to robotically increasing production no matter the consequence on prices (we can thank poorly designed incentive schemes for that, a practice that has had a staggering impact on national petroleum reserve quality remaining that no one likes to talk about). These drilled but uncompleted wells became known as the DUC inventory (whoever comes up with these acronyms isn’t very good at it, but anyway). So rig counts became disjointed from production increases, because all those newly drilled holes aren’t going to add much to production if no one completes them.

And then companies started completing wells and just sitting on them, not putting them on production, which meant a new category was needed beyond DUC and TIL (turn-in-line or put on production). These wells don’t yet have a name, these completed-but-not-yet-producing, but someone surely will come up with one because that’s how things go, we need labels, just like we need to label Boomers and Gen X and Gen Y even though the last boomer and the first Gen X have far more in common than the youngest Gen X and the oldest. Buckets give us comfort no matter how stupid the classification.

And we need names for these well categories because the industry pundits are not doing well trying to provide “analysis” of where production might be going based on rig counts. They do try, in a poignantly useless manner, like a 4 year old going out with a plastic wrench to help Dad change a tire. They do try. Consider this data analysis from Reuters’ most recent gripping account of the weekly change in rig activity: “U.S. energy firms this week cut the number of oil and natural gas rigs operating for an 11th week in a row for the first time since July 2020…The oil and gas rig count, an early indicator of future output, fell by two to 537 in the week to July 11, the lowest since October 2021…total rig count down 47 rigs, or 8% below this time last year…oil rigs fell by one to 424 this week, their lowest since September 2021…In Texas the rig count fell by one to 255, the lowest since November 2021…The oil and gas rig count declined by about 5% in 2024 and 20% in 2023…” Hey AI, synthesize that! No answer. The poor analysts just tie themselves in knots trying to describe the current state of affairs relative to the rows further up the spreadsheet in a frantic effort to sound relevant, even though anyone that pays attention to rig levels knows that DUCs and delayed TILs and, most importantly, technological advances just swamp all this pointless numerical breakdancing.

Here’s what really matters, and what makes a shambles of the sound-bite analyses of weekly rig activity. Consider this chart that compares oil production and frac crew activity levels:

Source: bluegoldtrader.com

Now isn’t that wild. Oil production has been quite disjointed from frac levels, except in extreme circumstances like peak-pessimism in the Covid years. And it isn’t even a comparison of drilling activity to production levels; it is the relationship between fracking levels and production which is a far better measure of correlation than drilling rigs to production, and even then the relationship is still evidently meaningless, over this time frame.

Here’s why, and here’s why the reporting industry needs to remake themselves somehow. A story from West Virginia news about how Expand Energy, formerly Chesapeake and Southwestern – two of the larges Appalachia players now combined – drilled a 5.6 mile horizontal lateral well, in 5 days.

That is pretty nuts. Five or ten years ago, no one was drilling even three mile horizontals, and no one was drilling anywhere near that fast. But drilling has advanced so far that a single rig can accomplish what it used to take multiple rigs to do, and that compounds the accomplishments of horizontal wells themselves, which increased the productivity of a rig far more than ten-fold.

Which means that if you are interested in rig counts for the same reason that people collect stuffed animals or watch birds or watch people play poker on TV, i.e., if you just enjoy it and you find that bucket of Reuters stats fun, then by all means have it. Probably best to keep that off your Tinder profile, but on the other hand who knows. Just don’t think it is going to help you predict oil and gas prices anytime soon.

And even if those activity levels lead one to believe that they can help in understanding forward markets, there are an infinite number of more meaningful events happening in the world like clockwork that can have a far bigger impact. Russia invaded Ukraine and invited sanctions which is bad for Russian oil and good for oil prices. Trump got elected, which is oddly enough bad for oil prices even though he is wildly pro-oil because he knows how important affordable energy is. Then Israel and Iran go to war which is good for oil prices. Then Trump bombs Iran which is really good for oil prices. Then Trump posts: “EVERYONE, KEEP OIL PRICES DOWN. I’M WATCHING! YOU’RE PLAYING RIGHT INTO THE HANDS OF THE ENEMY. DON’T DO IT!” Which is fairly straightforward messaging as bad for oil prices. Then US producers say that they won’t be doing much drilling at $60 oil which is good for oil prices. Then Houthis blow up cargo ships which is good for oil prices. Then…you get the picture. It’s very murky.

Having said all that, a few counterpoints are of interest. One, at some point, a low amount of drilling and fracking activity will have an impact on production, even if there is not much linearity in the short run. Totally fine to say that, but less so to make high-precision guesses on near term production levels based on those stats (and even far less wise to guess at price fluctuations…).

Then there is the “big bang” that is coming with respect to ultimate production capability. Just how many 5-mile wells can be drilled in a given oil or gas field? Obviously that’s a rhetorical question but we can say with great certainty it will be a hell of a lot less than the number of 2 mile lateral wells that can be drilled. These long wells are dangerous in the sense that they mask the reality of sweet spot exhaustion – everyone assumes productivity is going up and will indefinitely, but one day it won’t, and that will be a bad day indeed for each respective field.

And I don’t want to hear about “the infinite oil and gas” available, a theory which pops up every now and then, that because new technology has unlocked new potential it always will. Tell you what: go spend your life savings on an exploration well and let us all know what you think of the theory then. I’ll only listen to those with battle scars.

But for now, party on. The price signals for both oil and gas are dancing in the middle, not too much production, and not too little.