The Oil Data That Made No Sense


Something Doesn’t Add Up In North Dakota’s November Production Report

If you want to find buried treasure, you have to dig.

And even then you may find the treasure chest is empty.

That’s what happened to me this last two weeks as I realized one bit of oil data made NO sense to me at all—and spent several hours chasing down what the real numbers were.

The Market bids up and sells down oil and oil stocks each week based on a lot of data points—much of it from international (IEA), national (EIA) and state (North Dakota) governments.

And when I read this month’s North Dakota Director’s Cut (published by North Dakota Department of Mineral Resources) something didn’t seem quite right.

In fact it seemed very wrong.

So I started digging, and found a mistake.

North Dakota’s Director’s Cuts can be found here

Common Sense Would Suggest Otherwise

Like all of the oil market data that investors rely on, the North Dakota’s Director’s Cuts is released long after the actual production took place.  The November edition of the report was released last week on January 15.

It did not contain good news for oil bulls.

The report showed daily North Dakota production of:

  • 1,171,119 barrels/day in October
  • 1,176,314 barrels/day in November

That’s only a 5200 bopd increase, but seeing oil production going up anywhere at this point is bad news for the price of oil.  Seeing oil production going up in one of the major shale plays is even more dismaying.

It is more than a little bit surprising that oil production in North Dakota (the Bakken) is holding up, given high decline rates and plummeting rigs counts.

The news in the comments section of the November Director’s report gets even worse.


What’s disturbing is the number of completions.   A “completion” refers to the horizontal well being fracked and actually put on production.

To predict where production is going you must follow the number of completions, not the number of wells being drilled.  

The November report is telling us is it only took 26 well completions to grow North Dakota production by 5,200 barrels per day.

Now I know that the industry has been maximizing efficiencies…but given the rate that we know shale production declines…this seems hard to believe.

What A Little Digging Reveals

It is amazing what you can accomplish squirreled away in your basement office thanks to the internet.  There is a lot of good data out there for anyone willing to spend the time to find it.

The North Dakota Department of Mineral Resources, Oil and Gas Division provides a Daily Activity Report database.  Here you can find all of the oil and gas activity that happens in the state of North Dakota on a daily basis.

Included in the Daily Activity Reports are the number of completions.

I went through each of those daily reports for the month of November, and this is what I found:

Date Number of Wells Completed
2-Nov-15 4
3-Nov-15 2
4-Nov-15 1
5-Nov-15 3
6-Nov-15 2
9-Nov-15 5
10-Nov-15 3
12-Nov-15 1
17-Nov-15 1
18-Nov-15 4
19-Nov-15 1
23-Nov-15 7
24-Nov-15 3
25-Nov-15 2
30-Nov-15 2
Total 41

The daily reports show 41 wells being completed in November.  The Director’s Cut for November showed only 26.

That’s a 50% difference.  Why?  There is more to the story.

To determine the true number of well completions you also must factor in wells that are on the North Dakota Confidential Well List.

What is a confidential well?  Here is the North Dakota Department of Mineral Resources explaining:
When an operator requests and is granted confidential (tight hole) status for a well, it restricts our ability to release information about the well. Section 43-02-03-31 of the North Dakota Administrative Code states in part:

All information furnished to the director on new permits, except the operator name, well name, location, spacing or drilling unit description, spud date, rig contractor, and any production runs, shall be kept confidential for not more than six months if requested by the operator in writing. The six-month period shall commence on the date the well is completed or the date the written request is received, whichever is earlier.
Any confidential well that was completed during the month of November would not be included in the number of completions (26) quoted by the Director’s Cut.

Given that the confidential list that currently includes 1,836 wells that could mean a significant number of wells are excluded.

For more perspective on the size of the confidential well list—consider  that there were 13,077 producing wells in total in North Dakota in November.

It wasn’t easy, but I found a way to find out how many confidential wells were completed.  Most of the information about these wells is kept private for 6 months.  But not all of the information.

Any production runs (oil sales) from these wells are reported.  If a well is selling oil, clearly it has been completed.

There is no single spreadsheet or database of those oil sales.

However if you are willing to look up each individual well and record when oil sales started you can estimate the completion date.

It is a painful process, but I decided to do it.  Then I stopped doing it when I found another piece of info in the daily reports that summarizes confidential wells that have commenced producing or have been plugged.  I guessed that would be the number (assumed that no wells would need to be plugged) and NDIC Director Lynn Helms confirmed it.

When I counted confidential wells that started selling oil in November I found 26 of them:

Date Confidential Selling Oil
2-Nov-15 1
3-Nov-15 0
4-Nov-15 0
5-Nov-15 2
6-Nov-15 1
9-Nov-15 3
10-Nov-15 0
12-Nov-15 0
13-Nov-15 1
16-Nov-15 3
17-Nov-15 3
18-Nov-15 4
19-Nov-15 0
20-Nov-15 4
23-Nov-15 1
24-Nov-15 0
25-Nov-15 2
30-Nov-15 1
Total 26

That would suggest that the true number of completions in the month of November was 67 (41 in the daily activity reports and 26 confidential wells) and not 26 as reported in the Director’s Cut.

Oil Market Data Quality Issues Isn’t Unique To North Dakota

The oil production data that comes out of North Dakota is arguably important for investors, and mistakes like this really shouldn’t happen.

A quick common sense review of the numbers suggests that 26 completions has to be wrong.

But don’t be too hard on these guys.  The oil reporting agencies with much, much bigger budgets make much bigger mistakes.

All the time.

John Kemp from Reuters last week had a great article on the questionable quality of oil market data.  In particular Kemp detailed the

International Energy Agency’s historical difficulties in forecasting global oil demand.

This is an issue I also touched on earlier this year when I referred to the following findings from Raymond James:

US brokerage firm Raymond James took the IEA’s initial reporting of global oil supply and demand over a 15 year period and compared it to what the IEA’s final figures were for the same period several years later.
They found that on average the IEA had to revise its demand estimates higher by 700,000 barrels per day.

It is one thing to make a mistake.  But the IEA underestimates demand over and over and over again!

Considering these data challenges you can see why predicting where oil prices are headed is so difficult.

Remember my analogy of digging to find buried treasure—and sometimes coming up empty? That’s what this exercise was, in the sense that it did NOT change how I would invest in the energy sector right now.

But it does prove that nobody really knows what is happening in the oil market unless they’re looking at history, and not trying to foretell the future.

EDITORS NOTE—I don’t wait for oil to go up in my portfolio.  I go out and find energy stocks that have good looking stock charts and even better looking financials.  Like my solar “royalty” stock that has already increased its dividend in 2016.  You just have to know where to look. LOOK HERE to get the symbol of this 2016 winner.

Previous articleHow Many Energy Related Companies Have Upped Their Dividend in 2016 Already
Next articleThe Only Business Model That Works in Energy Now