Seeking Alpha

S. E. Wells

 
View as an RSS Feed
View S. E. Wells' Comments BY TICKER:
Latest  |  Highest rated
  • 2 New Financing Deals Change The Status Of Rubicon's Phoenix Project From Possible To Practical: Is It Time To Buy? [View article]
    Well, the mineralization is the same in the lithological and geochemical fingerprints as at the Campbell mine with the exception that the mineralization at Campbell is more in sulfides that need to be autoclaved. F2 is slightly better than BCD/Couchenour, not as good as the high grade zone at RL and a dead ringer for Campbell in terms of the gold content of the core samples.

    But, it's really not about types of mineralization, etc. It's about systemic deficiencies in the predictive methods in very highly variable deposits; i.e., it's about the statistics. If you look at AMC's latest newsletter, Digging Deeper, there is a one page explanation of what those deficiencies are. AMC did the first PEA for F2. Shame they didn't have a change or heart sooner.
    Mar 17 07:16 AM | Likes Like |Link to Comment
  • 2 New Financing Deals Change The Status Of Rubicon's Phoenix Project From Possible To Practical: Is It Time To Buy? [View article]
    PVG isn't going to have the grade underprediction problem I don't think. They chose not to use linear methods for the high grade in their deposit from the get go. I know they had problems with Strathcona saying there weren't any resources at all but Snowden put paid to that notion with the bulk sample results, etc. As for examples. Let's look at ones right in the Red Lake camp right next door to F2. I've done extensive analysis of the drilling reported for the Campbelll mine. When you normalize the drilling grade, F2 and Campbell drilling are pretty much dead ringers for each other over hundreds of thousands of meters of drilling. The last year the Campbell mine was spiked out separately that I've found was 2005. That year, linear methods for Campbell yielded P&P grades of 8.2 gpt (F2 is at 8.1 gpt). That year Campbell produced at 15.9 gpt. The 43-101 inferred grade for BCD/Couchenour is 10.2 gpt. If you look at the figures GG has published for tonnage and lom estimates, you will see they actually expect the BCD/Couchenour to produce at over 15 gpt. I saw a report from a TD analyst that said GG geologists told a group of analysts that they expect actual production from BCD/Couchenour to be around 17 gpt.

    Then, if you look at the Red Lake mine 43-101 figures published on GG's website, you will see that their 43-101 predicted grade is 9.94 gpt. In 2012, their actual mine head grade at Red Lake was 19.52 gpt.

    There are more examples but I think the Red Lake ones are most pertinent. One thing all these examples have in common is the use of linear estimation techniques for 43-101 purposes in deposits where the coefficient of variance is over 300%. That yields predicted results that are no where even remotely close to the grades at which production actually occurs, or in the case of BCD/Couchenour is planned to occur.
    Mar 16 08:35 AM | Likes Like |Link to Comment
  • 2 New Financing Deals Change The Status Of Rubicon's Phoenix Project From Possible To Practical: Is It Time To Buy? [View article]
    Yes there is a reason to think grades will be higher. To the best of my knowledge, and I've searched diligently, there is not a single example of a mine with a very high coefficient of variance in their drilling, such as F2 has, that has not produced at much higher grades than the grades predicted by linear estimation methods like inverse distance or indicator kriging. Those are the standard methods used in Canada but they do not work acceptably well for deposits like F2. When I say acceptably well, I mean they don't even come close to meeting the confidence intervals established by the 43-101 standards or jroc standards.

    I'd turn the question around. Is there a single deposit that you can point to that had 43-101 predicted grades that were even within hailing distance of being correct when the coefficient of variance in the drilling results is over, say, 300%? I don't believe you can find one. I sure haven't been able to do so.
    Mar 10 07:54 AM | Likes Like |Link to Comment
  • 2 New Financing Deals Change The Status Of Rubicon's Phoenix Project From Possible To Practical: Is It Time To Buy? [View article]
    No, I'm not long it. I plan to be before they go to production because I think the production grades will be higher than projected. But, this management has a really nasty habit of diluting shareholders, to my mind unnecessarily. I sold my shares before the gold stream and before the placement because I did not trust them to do what they said they were going to do the way they said they were going to do it. I got long it near the lows and sold about 2 weeks before they did the financing deals. I'd had enough of being diluted. So, when I think it's perfectly clear that they will get to production without further dilution, I'll probably buy some. But, quite frankly, some of the money that I had earmarked for this will be going toward Pretivm. To my mind, Pretivm handled the question of what the likely grade will be to near perfection. Yes, they will have to dilute too but they will dilute from a position of strength, not weakness.
    Mar 6 05:36 PM | Likes Like |Link to Comment
  • 2 New Financing Deals Change The Status Of Rubicon's Phoenix Project From Possible To Practical: Is It Time To Buy? [View article]
    I too expect a few grams higher. I think probably in the 12 to 15 gpt range. If you run detailed analyses of the drilling done at the Campbell mine( this data used to be in the public domain and available on GG's site) you find that the drilling is essentially a dead ringer for drilling done at the Campbell mine. As far as I know, the last year for which a resource estimate was compiled separately for Campbell was 2005. That year the Campbell mine had a 43-101 P&P grade of 8.2 gpt. That year it produced at 15.9 gpt. The mining methods used that year at the Campbell were the same as those proposed for F2.

    I do realize that it's not a certainty that the grades will be much higher. But, neither is it a certainty that the price of gold will go up. What I do know for a fact is that there is not a single case out there where this type of deposit did not produce significantly in excess of its projected 43-101 grade once in production. I know because I've looked. I've looked a lot.
    Mar 6 05:25 PM | 2 Likes Like |Link to Comment
  • 2 New Financing Deals Change The Status Of Rubicon's Phoenix Project From Possible To Practical: Is It Time To Buy? [View article]
    Is there some reason for you to believe the 43-101 projected grades are accurate? If not, then why use them in your analysis? If so, why bother?

    Look I do understand that the 43-101 numbers are the 43-101 numbers but there is literally no case extant where a project with the kind of coefficient of variance in the drilling that exists at F2 has failed to produce at much higher grades than projected by linear estimation techniques such as those used in the Preliminary Economic Assessment.

    An investor looking at RBY as a potential investment should realize that there are three reasons it might make sense. 1 - the price of gold might go up 2 - the mined grades might be higher than predicted and possibly much higher 3 - RBY has a lot more property in Red Lake that is highly prospective and they may find more gold on those other properties.

    In my view, it's numbers 2 and 3 that make it make sense. If the only thing that happens in the above list is that the price of gold goes up, there are many, many miners that will benefit as much or more than does RBY.
    Mar 6 12:03 PM | 6 Likes Like |Link to Comment
  • Royal Gold - Is The New Streaming Deal Beneficial For Its Shareholders? [View article]
    I'm sure it was getting some nominal return but if that return were adjusted for real inflation it was almost assuredly a negative rate of return.
    Feb 14 07:57 AM | Likes Like |Link to Comment
  • Royal Gold - Is The New Streaming Deal Beneficial For Its Shareholders? [View article]
    I agree that your calculations are correct. I think we just use the term royalty differently. My only point really was that the streaming deal of 6.3% with RGLD paying 25% of spot is equivalent to a royalty of 4.725%, etc.
    Feb 12 03:37 AM | Likes Like |Link to Comment
  • Royal Gold - Is The New Streaming Deal Beneficial For Its Shareholders? [View article]
    Where the real kicker is for RGLD is when RBY's F2 produces at MUCH higher grades than the 43-101 resource estimate predicts. The predicted grades are around 8.1 gpt and I think it most likely that RBY's F2 willl produce at between 12 - 15 gpt. If so, there is HUGE upside for RGLD in this deal. Enough to induce me to buy RGLD I think.

    If you look closely at the standard 43-101 predictive methodologies used in Red Lake, they don't even come remotely close to the actual grades at which production occurs. F2 is an extreme nugget effect deposit. Linear estimation techniques like inverse distance and/or ordinary kriging often cannot deal well with the type of deposit that F2 is and yield grade predictions that are low. But, check out GG's actual production at Red Lake compared to their 43-101 resource estimates, etc. to check this possiblity for yourself.
    Feb 11 06:32 PM | 1 Like Like |Link to Comment
  • Royal Gold - Is The New Streaming Deal Beneficial For Its Shareholders? [View article]
    The 6.3% royalty figure needs to be reduced as they will be paying RBY 25% of spot. So, the 6.3% becomes a 4.725% royalty and the 3.15% becomes a 2.3625% royalty.

    That said, there is a provision for adjustment downwards of the amount RGLD pays for the gold it gets from RBY related to the amount of debt senior to RGLD's stream RBY raises.
    Feb 11 06:25 PM | 1 Like Like |Link to Comment
  • Rubicon Minerals: An Interesting Bullish Case But Major Risk Remain [View article]
    There could be technical reasons. The one that leaps to mind is that it's physically impractical to take a sample that's scientifically valid because one or more of the domains is physically inaccessible without spending a lot of their remaining cash to get to it.

    My feeling is that it's too late to do this as the water around the question of actual grade has just been muddied beyond the ability of bulk sampling to clear the matter up. That's one reason that I think only actual production over some period of time will do the trick.

    They did make some mention that this might be attempted but I think the current environment and the need to conserve cash means it probably won't happen.
    Dec 9 02:55 PM | 1 Like Like |Link to Comment
  • Rubicon Minerals: An Interesting Bullish Case But Major Risk Remain [View article]
    That's the real conundrum here I think. The latest PEA contemplates a very different mining plan than the prior plan and that different mining plan is what necessitates the additional funding. Optimizing total return over the life of the mine is a different matter than optimizing the return over the ounces identified at this stage of development. I would like to think that the switch in mining plan is an attempt at the former rather than the latter. However, I would also feel more comfortable that the new mining plan is an attempt at the former if management had more skin in the game. That's because of prior management miscues relating to bulk sampling where they took a bulk sample that wasn't scientifically valid (because it did not accurately represent the domains present in the deposit). At this point, I don't know if getting to production with the cash they have on hand is really not desirable or whether it's simply beneath their dignity as big deal mining executives. I'd feel much better if I thought their reasoning in this matter was being adequately tempered by having significant real dollars of their own at risk.

    I think the market perception of RBY is predicated on the PEA prediction of likely production grades. If POG were to go to 1100 which I think is a distinct possibility, at the predicted grades, I don't see much reason to own more of it. (I have a tiny 3K shares at present). I would like to see the likely production grade issue resolved sooner rather than later. For instance, at 1100 POG the NPV is around 200 million but at around 14 gpt the NPV at 1100 POG would be around 1.3+ billionish. A lot rides on what the actual production grade would be in this POG environment. So, it really does boil down to a question of whether you can trust management and I agree that one of the important reasons to trust management; i.e., is their money at risk alongside mine, is missing. The old management, unsupplemented by Mr. Lalonde, was very prone to technical miscues. One can hope that Mr. Lalonde has put paid to that but hope isn't much of a basis for an investment.

    On the other front, I do think it might be possible for juniors to borrow instead of issuing equity as this is a topic that Sprott has weighed in on and has said they are in the business of doing, given adequate collateralization. (See Can Juniors Borrow Instead of Issuing Equity? at mining dot com).

    Hope this helps.
    Dec 9 12:40 PM | 4 Likes Like |Link to Comment
  • Rubicon Minerals: An Interesting Bullish Case But Major Risk Remain [View article]
    All in all, this is a good analysis of Rubicon. I've followed this company for several years.

    The only thing I can really add is that in addition to management's statements that the resource delineation was "conservative" there is a systematic bias towards under prediction of production grade in deposits like F2 that have an extremely high coefficient of variance in their drilling results (i.e., high nugget effect) when the estimates are made using inverse distance weighting. This bias results in things like the Red Lake mine having a 2012 Proven and Probable grade of 9.57 gpt when their actual production for 2012 was at 19.52 gpt. The last year that I know of where Campbell mine 43-101 grades were spiked out separately was 2005 when the P&P grade was 8.2 gpt and their actual production was at 15 gpt. The Couchenour/BCD 43-101 inferred grade was around 10.1 gpt whereas GG's geologists told analysts they expect actual production to be around 17 gpt. These figures are highly consistent with the under prediction of grade in the bulk sample areas at F2 that were reported in this article. In short, the systemic deficiencies of inverse distance weighting with top cut in heterogenous, high nugget effect deposits is another factor that a potential investor should consider. Personally, I've never found an instance where the typical methodology used in Canada (inverse distance with top cut) ever came even remotely close to accurately predicting the actual production grade in a deposit like F2 that has an extremely high coefficient of variance in its drilling results. Pertivm, another company mentioned by the authors, has realized this deficiency and took another approach to estimating their mineral resources and reserves using multi indicator kriging. So, bottom line, not only were the predictive methods used by SRK very conservative, those methods seem to be extremely biased toward under predicting grades in deposits like F2. Another thing to take into account even if the topic is technical and requires some research.
    Dec 9 10:53 AM | 4 Likes Like |Link to Comment
  • Rubicon Minerals: The Key Is Exploration Success At Its Red Lake Property [View article]
    The author doesn't seem to really know much about the Red Lake camp. First off, in 2005 the Campbell mine had a 43-101 predicted grade of 8.2 gpt. That year they produced over 200K ounces of gold and their actual grades were 15gpt. There isn't a single case extant in the Red Lake camp where a 43-101 predicted grade is even remotely close to the grade the mine is actually producing at. Production is at much higher grades than the 43-101 process predicts.

    For instance, the BCD/Couchenour project of GG has a 43-101 predicted inferred grade of around 10.1 gpt. But, Goldcorp's geos have told analysts that they expect the actual production there to be around 17gpt.

    There really isn't much reason to take the 43-101 process predicted grades in Red Lake seriously. In deposits with a very high coefficient of variance in the drilling results, the 43-101 process usually gets the tonnage about right but usually understates the grade.

    The author doesn't seem to actually know much about Rubicon at all.

    I think the actual production at F2 for RBY will probably be in the 12 gpt to 15 gpt range. There are risks for RBY. They should have gone into production sooner on a smaller scale to demonstrate the actual production grade and produced their way to a larger operation.

    Now, they've put themselves in the position of having to raise additional funding at an extremely unfavorable time to even begin production. Not a smart move.

    I think the primary risk to an investment in RBY at this time boils down to two things. The price of gold and what the capital they need is going to cost shareholders.
    Sep 12 03:56 PM | 8 Likes Like |Link to Comment
  • Rubicon Minerals - A High-Grade Underground Gold Mine In Canada [View article]
    You're right. Until the pudding eating has delivered the proof it is just potential upside, exactly like the price of gold going up is just potential upside until it actually does go up. I think the "nugget effect" is a little bit more predictable than the price of gold is though as it's a physical, statistical phenomena and the price of gold is more dependent on what people believe.
    Jul 15 02:21 PM | Likes Like |Link to Comment
COMMENTS STATS
141 Comments
517 Likes