Enno Peters
This is an older blog post, you will find one on more recent data here
The above presentation contains oil production data for selected wells in the Permian, in both New Mexico and Texas, until December. Unlike in the last update, I was able to estimate the production for most individual wells, based on the reported lease production. Having this data allowed me to show now also underlying decline.
A few remarks/findings:
- The total production overview shows the oil production from both horizontal (New Mexico + Texas) & vertical wells (Texas). December shows quite a big drop, but as always for Texas, I do expect revisions, especially for this month.
- The “well quality” tab shows the productivity of only horizontal wells in the Permian. I find it interesting to see that wells appear to have improved in recent years, on average. If you you select the grouping “Quarter of first flow”, you can see that this process of improving initial production has continued steadily. However, these improvements (similar as for the other basins) seem to be mostly located in the first 12 months, after which the profiles converge. The profiles are now quite similar as to the ones seen in the Eagle Ford (I have not compared the difference in gas production).
- If these impressions are correct, this would mean that the improvements are equal to a one-time gain, of which the size can be estimated from the difference in the first part of the cumulative profiles. E.g. we can see that the average horizontal well from 2010/11 does about 100 kbo by year 5. A typical 2014 well hits about the same production profile by end of year 2. By that time it has done about 100 kbo, vs about 65 kbo for the 2010/11 profile. Based on this I would currently expect these 2014 wells to do about 100 + (100-65) = 135 kbo by year 5. Wells starting in 2015 appear so far to do another 20 kbo on top of this.
- Another conclusion we could draw from this (assuming it is true), is that this higher initial production would require less new wells to keep production up, especially initially. But given that their (e.g. comparing 2015 with 2014 wells) total production over a number of years may not be much greater, still almost the same number of wells are needed as in the past for the same amount of total output. This is a different conclusion than what I’ve seen in some media reports, which claimed that these improvements in initial production would require far less wells to keep production steady. I would be interested to hear what you think about these 2 ideas.
A few more technical remarks:
- For the wells in New Mexico, the data is very accurate, whereas for Texas I had to estimate well production data based on lease production data. I have further fine-tuned this algorithm in several ways, which makes me belief that the well profiles shown here are quite accurate. Despite these improvements, the change in these well profiles compared with the last update doesn’t appear to be large (as you can check for yourself). There are still a number of wells that I had to exclude though : there are 5117 horizontal wells in the selected Texas leases that I’m using here, while for 4184 of those wells the production is shown. The reason why I had to exclude some of these wells is that the algorithm couldn’t handle the lack of detailed well information available for the leases in which they are. I’ll try to improve this further in the future, but there are limits to what is possible.
- The total production overview (“how much is produced” tab), includes 2859 horizontal wells from New Mexico, 4184 horizontal & 13373 vertical wells in Texas. Because I had to omit some wells/leases, the total production shown now for the Permian is a little less than last update, but now the underlying decline can be shown.
I expect to have another update on the US up by next week Friday.
====BRIEF MANUAL====
The above presentation has many interactive features:
- You can click through the blocks on the top to see the slides.
- Each slide has filters that can be set, e.g. to select individual or groups of operators. You can first click “all” to deselect all items. You have to click the “apply” button at the bottom to enforce the changes. After that, click anywhere on the presentation.
- Tooltips are shown by just hovering the mouse over parts of the presentation.
- You can move the map around, and zoom in/out.
- By clicking on the legend you can highlight selected items, and include or exclude categories.
- Note that filters have to be set for each tab separately.
- If you have any questions on how to use the interactivity, or how to analyze specific questions, please don’t hesitate to ask.
News, trends and data for the US upstream industry
Novi Energy Newsletter
Sign up today for free and get a weekly email with News and insights on important developments in the major tight oil & gas basins.
21 Comments
How much of the yearly improvements are from a greater percentage of hz wells? Is it possible to filter out verticals?
Also, just intuitively why do you think well quality has gone up so consistently year by year? Would think that Permian is rather well known, even more than the Bakken or EF. And then if there was some learning, why not a step change and then flat? Why so year by year, the improvement. Just looking for some guesses.
Nony,
“How much of the yearly improvements are from a greater percentage of hz wells? Is it possible to filter out verticals?”
The total production overview shows both horizontal & vertical (only from the Texas side) wells. However, the “well quality” tab only shows the well productivity from horizontal wells, so they are indeed filtered out in that view.
Enno,
Really appreciate your work, I’ve been doing similar analysis. Your Permian numbers seem a bit out of line compared to EIA:
http://www.eia.gov/todayinenergy/detail.cfm?id=24932#tab_1
Cheers
Thanks Che,
I had a conversation about this topic, with the same reference, earlier. Our conclusion then was that the EIA includes both horizontal and vertical wells in this analysis, whereas I only include horizontal wells (in the well quality tab).
I have included more than 50% of the horizontal wells (maybe closer to 80%) in the Texas side of the Permian (100% of the hz wells in the New Mexico side), I’m still trying to enlarge the scope. Also, I have omitted gas focused wells in all the basins, and therefore not included any wells from “gas leases” in Texas.
I belief these are the main reasons for the differences with the results in the EIA article you referred to.
Enno,
Thank you for your reply. Insofar the hz well issue, it would seem that proportion of vertical wells drilled since 2012-2013 has dropped to less than 10% (at least in Delaware and Midland): http://info.drillinginfo.com/permian-basin-production/ (a bit dated but think probably relevant to current situation as well). This increase in hz well count is probably the main reason why vintage performance in Permian has been going up so much.
I’m curious on the “gas leases” what proportion of your total sample were those?
Thx
Che,
“This increase in hz well count is probably the main reason why vintage performance in Permian has been going up so much.”
Exactly, that is also my impression – at least it is a major contributing factor.
“I’m curious on the “gas leases” what proportion of your total sample were those?”
So far I’ve excluded gas leases from both the EF & Permian (Texas), mostly because of convenience, and to keep the comparison a bit more apples to apples, even though there are of course still big difference between horizontal wells in the different areas. I don’t have a handle yet on how many horizontal wells are in those gas leases, I hope to be able to say something about that in the coming 3 months.
Also a bit dated, but nevertheless interesting:
http://www.artberman.com/less-than-2-percent-of-permian-basin-is-commercial-at-30-oil/
Would be more inline with the EIA data which basically says Permian is worse than EF/Bakken.
Another way to check is to go through the company filings and compare the number of wells added to increases in production. It’s as messy (if not more) as TRC data but it also seems to show that well performance in Permian is considerably worse.
Che,
I do think that the well productivity shown in this presentation is quite representative of the actual performance of horizontal wells in the Permian. Although the TRRC data is not easy to work with, once you have all the data there is enough information to make reasonable estimates of well productivity in most leases.
As shown in the latest US update (Well quality tab, select grouping by basin), Permian wells indeed seem to underperform the Bakken & EF so far, in general. It may still be that there are certain areas where wells are better, or that there are other factors that I am not considering that change the economics (depth, costs, infrastructure, etc). Still, the general picture I have is that the Permian hasn’t been a game-changer, yet.
(If you’ve already discussed this, mea culpa)
There is a pretty well known effect of the reported overall production rising over time in Texas. Quite substantially and for several months. What is going on when that happens and how does this effect affect your shale profiles here?
Nony,
Indeed I expect that this issue will also affect the data I report here. But first I need to collect a few more months of revision data before I can comment on in what ways, and by how much, this has an impact. For now I ignore the latest month available (for which I can already determine that it is very incomplete), and make a note that earlier months are still subject to revision.
OK. I am interested if it applies more to certain plays or operators. How it impacts well quality comparisons over years, etc. Intuitively would think it would mean near term underreporting, so perhaps Permian and PXD (already showing year by year improvement) are improving even more than shown. But I don’t really know.
Pioneer, the mother fracker.
Nony,
So, what are your comments about this article? I expect a little more from you, as you pride yourself on critical thinking 🙂
What I see is a nice and interesting story. However, on the key points, how well these wells are actually doing production-wise & economically, I find the article highly misleading.
1. “Pioneer began to have some success drilling horizontally in a south Texas formation known as Eagle Ford.”
Check out its wells in the Eagle Ford, on which I have very good data. It looks like its 75 horizontal wells there, in oil leases, will on average not get to 200 kbo. Subtract at least 25% for royalties, and then with the $8+ million well cost I doubt whether these wells were an economical success. I don’t see new wells after 2014.
2. “The three are expected ultimately to produce a million barrels each.”
What do you think about this, when you look at Pioneers actual well profiles?
3. “and more than 20,000 drilling sites that could hold as much as 10 billion barrels of crude.”
So the average well is going to return 500 kbo? Looking at the well profiles in the Spraberry formation (Midland & Martin), that would not be my expectation.
4. “Such efficiencies have made it possible for the company to amass cash with crude prices as low as $35 a barrel.”
I would love to see a calculation that shows how these wells are cash flow positive with $35 a barrel.
5. “Pioneer’s cash reserves rose to $1.61 billion in the first quarter of this year, up from $383 million a year earlier.”
That sounds pretty good. But of that $1.61 billion, is not 99% ($1.597 billion) from a share issuance in Q1 ?
In comparison, cash from operations was $1.359 billion in the last 5 quarters.
During that time, they spend $2.943 billion on oil & gas properties & equipment.
I’ll reread and try to make some comments. Thanks for detailed reactions.
I’m not sure that I “pride myself”. I just do what I do. But will try to “have a take”.
a. I do think it is interesting to look at the good companies or the success stories. I support you when you cut through anecdotal remarks and show overall cum curves and the like. That said, the Permian (and PXD) do show dramatic year over year improvements in well quality. It is worth looking at and thinking about also, not just the hype destruction, but also the hype. Capisce? I mean run the well quality cums by year in the Permian and it is a very dramatic chart–just like some of your hype destruction charts are dramatic.
b. I though the story was kind of interesting with the human aspect to it. Also, I did not realize that Sheffield had just been hanging out so long like he had. (Guess you could say same of Hamm–he used to be president of the stripper well organization!) The part at the end was very interesting. I think “he got lucky” is a very reasonable hypothesis!
c. (Eagle Ford): The story is talking about EF wells in 2011. They did about 150,000 bbls in 42 months. Maybe some gas too (this is the EF, so you can’t dismiss that aspect of it.) Don’t know their royalty spend, but at $100/bbl, those wells may have been money makers at the time–which is what the story discusses.
As far as currently, if you read the PXD conf call transcripts/slides, they talked about de-emphasizing the EF for last few quarters and even took a capital charge associated with moving some equipment out of the field and to the Permian.
c. (Point 3, total resource): That sure sounds like the old story at $100/bbl. I wonder if it is substantially lower at sub-50. Actually, I really don’t know how you can evaluate EUR claims based on the first few years (from cornie or doomer slant). Using YOUR TOOL, I get a pretty clear story of PXD Permian wells getting better and better every year. in 2015, they were hitting ~115,000 at 12 months. I guess I could see wells like that doing 500,000, no?
Now, maybe wells need to start getting worse as they drill out of the sweet spot. But also, maybe they keep improving techniques. In the Bakken, it seems like the two effects kind of neutralize each other, looking at it basin-wide. Don’t know how it plays out for PXD. Not making a big claim one way or the other. I just wouldn’t be too sanguine about the “hater” view or the “hyper” view. Obviously, they have incentives to push how great they are and the Permian is. But also obviously, they have a lot more access to data and analytics than we do.
d. (Point 4) both you and the article are being a little imprecise. And you are a bit putting words into the articles mouth. But I don’t even know what it means for a well to be cash flow positive. By it’s nature, a well is a large upfront investment (negative cash flow), followed by positive cash flows. Doing a discounted cash flow analysis gives you the NPV. But it’s never going to be cash flow positive, the month that you drill it. It might even not be positive NPV. But you have to do the math to see.
The overall cash increase is interesting. I wonder how much of that is from equity raised. Didn’t they do a stock offering? I don’t think you can give the wells credit for that. [You might say the wells didn’t drain it as fast as they used to, but then you need to get into the math and show that.] They are growing, still, so some outspend of CFFO into capex may be justified. (Not saying it is, you have to do the numbers. Just that, you do have to acknowledge that growing companies may need to invest more than their cash flow.)
e. The story is still playing out, but so far the whole Einhorn bombshell seems to be fizzling. It’s not like PXD has been exposed by his announcement and is now imploding Enron-like. More the opposite.
Nony,
Thanks for your comments.
“That said, the Permian (and PXD) do show dramatic year over year improvements in well quality. ”
Yes, that looks pretty impressive. But it also appears that these improvements may just be happening in the first year or so, and if that is the case, it would just be a total gain of less than 40 kbo, compared with wells a few years earlier. That is of course nothing to sneeze at, but given that these well profiles are worse than what we see in the EF, it is not sufficient for me to see what the latest hype in the Permian is about.
If these gains would be more sustainable, that would be a different matter. So, as always, I’m curious to see how it plays out. If you group by quarter, you can see the well profiles strongly converge between 12 & 24 months.
” get a pretty clear story of PXD Permian wells getting better and better every year. in 2015, they were hitting ~115,000 at 12 months. I guess I could see wells like that doing 500,000, no? ”
The way I try to estimate what these wells can be roughly expected to yield, is by looking at when they eat into the tails of wells one or more years earlier, and than comparing on the cumulative chart how much they have done extra by then. Next, you can see where those earlier wells end up with, and at the amount they have done extra in the beginning.
For example, by looking at all Pioneers wells, we can see that 2015 wells at month 12 are flowing at a similar rate as wells from 2012. At that time they have produced 70 kbo more than those 2012 wells. By month 50, those 2012 wells have returned about 100 kbo. So I expect 2015 wells to be doing somewhere not very far from (100+70) 170 kbo, by month 50. How much life they still have in them by then, is anybody’s guess, but they seem to flow then at around 25 bo/d, which is not highly profitable, and less than half what Bakken wells are doing for example. I also can’t tell you exactly what they’re going to return, but based on these estimates, an average of 500 kbo UR does seem to me quite inflated, yes. I would be surprised if they would be doing more than half of that (without additional capital, for e.g. refracing).
I honestly don’t understand why companies are allowed to mention such highly optimistic, unproven, EURs. Of course it has a great anchoring effect on investors, and worse, I suspect they are used for reserve booking & depreciation schedules as well (and management bonusses).
There is great value in companies being conservative in their accounting, and planning. The risks are asymmetric.
“The overall cash increase is interesting. I wonder how much of that is from equity raised. Didn’t they do a stock offering?”
That’s what I said, right? 99% of the cash they had end of Q1 came from a share offering during that quarter. That seems like something that would have been worth mentioning in the article.
Hi Enno
Stumbled across this blog today – great work you do here.
On the well profiles, it was notable to me that Pioneer do not mention ‘Type Curve’ or anything similar in their 10-K filing with the SEC, unlike the widespread usage in less official presentations.
I also did some calculations (don’t have them to hand – will return) on their proven developed reserve additions last year after stripping out the effects of price changes – even allowing for gas, I calculated a fraction of 1mmboe (think it was about 300-400mmboe/well… will have to check).
Again, great stuff.
Greasy.
Thanks for your comment Greasy, great to hear that you find it useful.
Why does this only go up to 900k bpd? Isn’t the Permian around 2.2mm bpd or am I missing something?
Jay,
The Permian is huge; I’ve only focused so far on some of the main shale formations, while ignored a very significant amount legacy production. In the coming months I try to increase my coverage of the Permian, while maintaining this focus on horizontal wells in the shale formations.
Got it, thanks.