There is a general consensus that longer laterals are the way to go in the unconventionals, but does the data suggest this practice is working, particularly in the Bakken? This entry deals with some of the data involved in this issue.
First, a look at how some of the top operators are drilling their Bakken wells.
ND is split into 1 square mile sections in the section township range system (STR). 1 square mile section = 640 acres, 2 sections = 1280 acres etc. For the rest of this entry I have defined two categories of Bakken laterals.
Short laterals: One-section laterals such as the EOG example. The surface to bottomhole distance can be up to about 7500′ if the well is drilled diagonally across the entire section.
Long Laterals: Two-section laterals such as the Whiting and Continental examples. Most of the new wells coming online are drilled in this fashion. I classify these as anything greater than 7500′, however the vast majority are up near 10,000′ since they span the greater part of both sections.
So, is there a production benefit to the longer laterals? This is something that we at DI-Energy Strategy Partners have been studying extensively. Here are some charts. The first one I binned the short and long laterals, ran a count and an average 6-month cum, then trellised by year.
These next charts are the same info just presented a bit differently. The top chart shows the long lateral, 6-month oil cums are increasing steadily with time while the short lateral 6-month cums peaked with the 2008 wells. The bottom chart shows the percentage breakdown of long vs short wells over each year since 2006. In 2010, almost 3/4 of the producing wells were long laterals.
Finally, a table of the top 5 operators for each lateral type. This is out of a sample of about 1800 Bakken producers with first production dates from 2006 to present.
What conclusions can be drawn here? For the majority of the modern Bakken long laterals have not been outproducing short laterals on a 6-month oil cum basis. It was not until 2010 that the long laterals overtook the short laterals. The same trend is seen with 12 month cums. However, the steady improvement of long lateral performance shown above is telling. As more long laterals are drilled, performance improves. This fits perfectly in line with the “learning curve” theory adopted by most and quanitfied by some of our past studies.
The data shows an industry switch to long laterals, but is this ultimately the best way to go? 10,000′ laterals are a tremendous engineering feat. Especially when considering a 30+ stage frac job. And I have seen quite a few engineering talks discussing heel/toe production disparities. The idea that most of the production comes from the heel suggests a series of short laterals can be an interesting alternative.
Factors such as marginal cost to drill the long lateral, shortages in frac crews and rigs, and urgency to hold sections play large roles in this as well. Leave a comment or email with some ideas/thoughts/feedback etc.