Throughout this year, I have been writing articles each month on the Niobrara and using data provided by the EIA (Energy Information Administration) in its Drilling Productivity Report to estimate what the year-end production figures for oil should be in the region. Since last year is over and the EIA has given their forecast for December of the year, I figured it would be a good idea to go over my work and see if and how the data provided was useful for long-term oil investors.
In my prior works on the Niobrara, I had three different forecasts for oil production every month, a conservative scenario, a moderate scenario, and a liberal scenario. To stay in the middle, all of my numbers below, unless otherwise stated, refer to the moderate scenario data I looked at, which was, in my opinion, the most realistic outcome for production.
Off by a long shot
One of my core beliefs is that you need to be honest about your performance at all times. This means, to me, basking in your successes but humbly and easily admit your failures. In my articles on the Permian and Bakken, I was able to talk fondly about how close my model was to predicting production levels provided by the EIA for this year but in the case of the Eagle Ford, I made a mistake but so too did the EIA. Is the Niobrara akin to the Permian and Bakken or is it in the Eagle Ford category for the year?
Sadly, the answer is the latter. You see, in the graph below, you can look at how my oil production forecasts for the end of this year (with each month showing my year-end forecast at the time such that January's 92,803 barrels per day is the figure I estimated for oil production using January's data for December of this year). What you can see by looking at this is that, on an almost monthly basis, my revisions grew worse, eventually climbing 309,462 barrels per day from my January forecast until the EIA's data actually became available for this December.
*Created by Author
On the bullish side (expecting less production for the year), my forecast error amounted to 77%. Unfortunately, because oil production numbers in my model were never higher than the EIA's December forecast, I never had a bearish (expecting more production for the year) forecast error.
What about the average picture?
Thankfully, if we look at the average oil production forecast every month for the entire year, the picture is a little better. It's still terrible but not what it could be. In January, for instance, I estimated that oil production throughout the year would average 205,524 barrels per day in the Niobrara but the EIA's data for December shows that it will likely have been about 425,006 barrels per day throughout the year.
*Created by Author
If we look at the forecast error here, it's certainly smaller (for the bullish case) but, at 51.6% of the total production estimated by the EIA for December, it's still far from the mark and not pretty to behold. Just as in the case of the year-end production estimate method, I did not have a bearish forecast error.
What went wrong?
Beyond any doubt, I do deserve some blame here, especially when it comes to the rig count forecast I provided. Earlier in that year I figured that, under the existing conditions, the rig count would eventually fall to 0 and stay there but as the energy market recovered, so too did the rig count. According to Baker Hughes (BHI), it stands at 21 units as of the time of this writing.
While I certainly made a mistake here, however, the EIA was also terribly off in some key area. For instance, in January the decline rate in the region stood at around 14% according to their forecasts but by December it had dropped to just 7.07%. This is around the same level the EIA now believes it was in January of this year, meaning that they had to go back and materially revise their data. If a rate of around 7% had been chosen the first time around by the organization, the amount I estimated for the year-end approach would have been 107,977 barrels per day closer in January than it actually was and it would have been about 75,601 barrels per day closer for the average method compared to my initial forecast.
*Created by Author
Another error made by the EIA relates to their poor estimate regarding total oil production every month. Take, for instance, the case of January. In my January report, I saw oil production estimated by the EIA at 343,888 barrels per day but by this report (the one issued in November but showing estimates for December), that number has been revised higher to 464,379 barrels per day. That represents a spread of 120,491 barrels per day between the two and suggests that if the EIA had been closer to accurate, even with my error regarding the rig count, I would have been materially closer than I actually was.
Based on the data provided, it's clear that the picture in the Niobrara was far off throughout the year. In fact, it seems to have been worse than the Eagle Ford's data and shows that not everything provided by even official organizations like the EIA will end up being correct. This does not mean that the data should not be used but it does mean that investors should never take just one report for granted. That's why, every month, I made it a point to continue updating my readers on the Niobrara so that they could have access to appropriate estimates for the region in a timely manner and stay on top of changes in the domestic oil market.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.