Wednesday, February 28, 2018

Pretium - everything is great

Good news, Pretium at the BMO Conference is telling us that everything is great (link), no need to worry. Production at Brucejack is swell, and everything is going to be super-good after the mill expansion. For fun, they are looking for the source porphyry, because, why not.

We also got the Brucejack ramp up production figures, you know, the ones where grade and ounces produced went down. We actually get to see the 'official' head-grades, and they were worse than Otto calculated (link).

Tonnes up, production down - Pretium going for the win!

They also included a nice long section of the Brucejack block model, which was interesting.

The reserves aren't nice and consistent, there are distinct area of high and low mineralization, but there is an awful lot of the Proven Reserve blocks that are green and yellow (i.e. 5-10 g/t Au).

This isn't uncommon, but if Pretium want to maintain consistent production level, they'll need to be very good at balancing the production from high and low grade areas (mine scheduling), and for me, the erratic nature of the mineralization is the reason why Pretium gave such a huge range for the H1, 2018 production. They just can't quite nail down the grades.

TL:DR version - it is going to be hard to consistently maintain a head-grade at 14.5 g/t Au.

I apologize for the next section, it contains a very naughty word. If you are sensitive, please go here (link)

It would have been interesting to have seen a a slide on reconciliation to see why they had missed their targets, was it due to:
  • Poor reconciliation - mined graded not equaling reserve grade
  • Slow development - unable to access proposed 2017 production areas and forced to mine peripheral, low grade zones.
We all know it is grade reconciliation, hence the reason they are doing a new grade control program and drilling the stopes on 5-7m centers. However, the slide that concerned me the most was this one (slide 17)

According to Paths, Pretium are retards, and are actually using the Reserve figures in their own technical reports (who the feck does that?!?), and are projecting that after the mill is expanded, the 2019 production will be:
  • ~580,000 oz Au per year 
  • A head grade at the Reserve Grade - 14.5 g/t Au*
  • AISC @ ~$570/ounce
*if you plug the numbers into excel you get a recoverable grade of 13.2 g/t Au or a recovery rate of 91%

So, to do this, all they need to do is solve the grade control issues, improve mine scheduling, improve recovery, and increase the underground development. Nice and easy.

If they don't manage this, at 3,800 tonnes per day each 1 g/t decrease in head-grade means a drop in production by 44,000 ounces, or $57M decrease in revenues. That is a lot of money, especially if you want to do this:


  1. I'm starting to have an issue with this management. Things are looking like chit vis-a-vis the original numbers and is the answer to spew increased production and assume that alone will right the ship? That's a recipe to lose money faster whilst exhausting the resource. Next numbers will tell much. At this rate, and assuming no improvements, this is one pale elephant growing white.

  2. Very fitting:

  3. Looking for the source porphyry? All they have to do is walk 500m down the hill and they're standing on KSM - a pretty big (albeit borderline uneconomic) porphyry.....

    1. KSM is "uphill" so to speak. VOK is "high up" in terms of epithermal elevation so a porphyry source would be somewhere below it. I don't think they are looking for the porphyry to mine it, they are looking for the plumbing because presumably that can tell them more about the structural controls, and the associated dyke conduits may also contain more continuous mineralization.

    2. From the presentation they are planning a couple of long exploration holes to explore the flow dome intercepts to depth.

    3. @Tom of course they aren't going to mine it. But its pretty evident that KSM is likely part of a related system. That whole area has been faulted and tilted so many times its pretty unlikely that the source porphyry is sitting directly below VOK still perfectly in place.

    4. They mentioned structural geology indicators, likely what they have is a system of jointing formed in reaction to the porphyry intrusion as they move to the other side, they can see the orientation of the joints rotating in orientation relative to the expected porphyry so essentially they know where it will be, just mater of intersecting it. It could well be a two km long extension of the deposit, although thinning and narrower but increasing in grade at depth. The significance is lost on people, most do not recognize this as a fantastic drill target with a very high probability of success.

      Regards, Paths
      (twitter @resourcepathway )

    5. Hey look, it's a porphyry cluster! Not unusual if there is a structural corridor controlling emplacement which often is the case in a volcanic arc. I didn't say a porphyry is sitting "directly below VOK still perfectly in place", just that the KSM porphyries are unlikely to be related to Brucejack.

    6. "They mentioned structural geology indicators, likely what they have is a system of jointing formed in reaction to the porphyry intrusion as they move to the other side, they can see the orientation of the joints rotating in orientation relative to the expected porphyry so essentially they know where it will be, just mater of intersecting it." -- Paths

      If you could vector to a porphyry center on structures then exploration would be so much easier!

    7. The planed holes (slide 20) are going to be ~1500m long and the intercepts in the flow dome, impressive appear to only show 2-3 structures. It will be interesting to see what they get in the gap between the VoK and the Flow dome areas.

      Slide 21 confused me, they have basically sampled every rock outcrop on the project, but didn't show any results. probably just a coincidence.

  4. "L'audace, l'audace, toujours l'audace."

  5. it feels like Strathcona was right and the resources are skewed by over-weighting the high-grades. This feels like a reserve grade of 9 - 10 grams is going to be a more reasonable estimate and once they try to move to a higher production rate, the grade will probably decrease by a gram because the push for production will cause more low grade to be dumped into the mill. I find the lack of transparency disconcerting and drilling miles of RC grade control is going to slow down production more than it is going to refine the head grade (IMO)

    1. That is my opinion, the mine will still produce a lot of gold, but below expectations.

      if we plug the numbers in, assuming a 9.5 g/t head-grade and a recovery of 95%:
      At 3800 tpd they should produce around 400,000 ounces of gold.

      Not bad, but ~180Koz lower than presented with an AISC around $750-800/oz.

      Upside - there will be much better quarters as mining gets into the high-grade zones
      Downside - if they miss the production figures for Q1, 2018 (they'll know that now), will they be tempted to high-grade to make the H1, 2018 guidance, and fuck themselves for a short-term gain?

  6. Adding grade control drilling and increasing the availability of stopes cannot be a forced reaction to a chain of events. It is simply part of the ramp up. Four years ago I predicted this would occur, as I explained in a post on Stockhouse on Dec. 6, 2013. If I can figure this out four years in advance, it can hardly be an unexpectedly required action of the mine operator.

    This will obviously work well, anyone who doubts this, is not familiar with the many successes seen in open pit mining. I can well imagine that perhaps they never will get to the estimations, and are always 10 to 15% behind, until they revise their parameters. However, that is no serious issue to the big picture. They have a bulk mineable deposit, where high production is easily achieved at low costs. Reaching the centre point of the guidance range, that is 175k oz.'s should be the real guidance target, and surpassing it should not surprise anybody who has been paying attention.

    My head grade calculation is different 580,000 oz x 31.1 gr/oz./(3800 x 365 days) = 13.00 g/t after diluted and recovery. So it looks to be a conservative and reasonable adoption of the reserve estimates.

    twitter @resourcepathway

    1. You could figure it out, yet they didn't bother to include it in the feasibility study or mine plan! Perhaps you were being a bit more careful and they were a bit too sure of themselves? And perhaps now other are being a bit more careful and you are being are a bit too sure of yourself.

    2. I used a 360 day mill utilization, as I don't expect that any mill can operated with no shut-down over the course of a year.

      The issue is, if they keep missing their production figures, as they have for the initial 6 months, where they haven't been able to maintain the head-grade.

    3. the TL:DR version - the mine will make money, it should produce a lot of gold, but Pretium may be giving themselves a reputation of over promising and under delivering.
      That is something that the market may not forget, unlike Pretium still not disclosing the fatality they had.

  7. My forecast is after a few months, once this is starting to be better understood, market participants have enough information to come to terms with this situation (some never will). What will happen is many of the overly critical, oh so negative doubters, will have been humbled, (by the great humbler, stock market reality in the real world).

    The fellow at Exploration Insights will be recognized as having gone in print as having glaring deficiencies in his understanding of geostatistics. Some subscribers who used his article to help build what is now a huge short position will been through a serious and damaging short squeeze, all of which should no be any surprise to anybody who actually paid attention during their geostatistics classes. I suspect that it would be impossible to find any owner of these short positions who even has a working understanding of geostatitics issues. I remain firm in my view, that there has never been even one single strongly negative comment made online about Pretium, that does not involve or include technical errors and/or faulty assumptions regarding the use of geostatistics. I have only had one observation/criticism, that being the Snowden report did not actually show the variogram curve plotted along side the scatter of data points, there may have been proprietary issues, but I do trust that is was done correctly.

    There should be more geoscience people willing to make a stand here, and say no, this is a very much misunderstood situation, an example of online group think gone off the rails to excess, aided by critics who find it fashionable to criticize Pretium. I'm not referring to Angry Geologist, however, who has attempted to be somewhat balanced, and who never has repeated the most incorrect talking points, such as "the block model will not work, Strathcona will be correct", and the most serious of the nonsense.

    Paths, Stockhouse
    midpath@resourcepathway, Twitter

    1. Man you are working overtime for your shill bosses! I don't see (most) critics saying to short the company, most understand there are some high grade blocks so in any given quarter or even a run of quarters they could come back with spectacular production and that would blow up a short position.

      "Failure" doesn't have to be absolute as you are attempting to frame the criticism, it could be a matter of degrees (a main one is ability to repay debt, but it could be something less drastic like they just have a "regular" mine with decent grades but nothing to write home about, making Pretium pretty much fully valued at the current price for example). And we're already seeing a small degree of it ... the failure to assign proper grade to 1200 level blocks outside tightly spaced infill drilling. The company has even admitted to this so you don't need to provide any apologetic BS as a counter.

      They had some problems with MIK at Pirquitas, it didn't model the transition zone correctly and by the time they had the mine "ramped up" the resource was nearly depleted. Pirquitas had the same advisor (Snowden), and even the same exploration manager (Warwick Board), as VOK. Maybe people would have greater trust if their prior fiasco hadn't left such a bitter taste.

      Simply put, these guys are using a very risky approach here, the MIK doesn't give you very sharp cutoff shapes, they don't really understand the structural controls, and they apparently haven't tried any risk mitigation such as running a conditional simulation to see if revisions in the mine sequence could be warranted especially early in the mine life when debt repayment is paramount. Instead they are going to try blast hole sampling as a selective control, when they previously claimed that a sample tower of even a bulk sample (which is much more representative then blast hole cutting) was not going to provide a good representation of block-by-block grades. I mean, they will be able to confirm the very high grade blocks with this method, but the problem isn't the very high grade blocks. It's the mid to lower grade blocks.

    2. They would have stayed with their FS plans while starting up, too many things happening at once to make changes. More flexibility and regular planning revisions will start to occur. They are entitled to use three month averaging, they would only confuse most shareholders if they gave grades of specific locations, and frequent reconciliation numbers. It's longhole mining in a random network of stockwork veining, of course there is no sharp cut off. Any other mining method would have been worse. I think they would have gone ahead with the sample tower, and it would have worked, problem was Strathcona was unfamiliar and fearful and anxious for their reputation, and wanting to publish their doubts, that is what stopped the sample tower work from going ahead, I think. The problem is the modelling of the high grade distribution, the low grades model well over large ranges.

      I'm no shill. So far, I seem to be about the only mining person messaging online, who has made correct assessments, predictions, and explanations across time. I was not over bullish before, and then reasonably bullish, correctly now. I have been very correct on this at turning points while many others have been wrong. Being accurate, being correct, and also making a good effort to help other investors who certainly deserve some assistance, at least whatever I might be able to help with, in this difficult to analyze situation.


    3. You are being highly disingenuous about the sampling towers and Strathcona. They knew what they were doing, and they wanted to do a block-by-block reconciliation while it was actually possible, but they were prevented from doing so by Pretium. It wasn't only to validate the block model, it was first and primarily to validate that the mining method would work given the grade characteristics. The reconciliation, which was very poor and highly unreliable, never got disclosed and therefore the discussions about the mining method and grade distribution were buried.

      Forget the low grades, they need to be thrown completely out and treated as waste. There are no 5-10 gram blocks, nor any 10-20 gram blocks. There are either blocks that are 1-2 grams or lower (the background grade of the mineralized portion of the stockworks), which are clearly not economic to mine underground, or there are high grade blocks, which in reality have very sharp boundaries that can be mined selectively, but of course without seeing the mineralization on a mining face (which you won't with long hole stoping) you only have the drill holes (doesn't matter what drill holes, core or RC infill, blast hole, etc. none of them will provide you with the right info).

    4. I didn't follow it all back then, but they said there was not an economic ore body, which was wrong, as it's been above cut off grade since the beginning. There never has been a reasonable question about the mining method, and a reconciliation of the grade on the small scale of the bulk sample area or even the level would not have helped much.

      There is a range of block grades, look at the drawing. The high grade blocks are often isolated, and form irregular units, not nice zones. The only sharp boundaries occur if they were to try to mine the individual veins, which change directions and move about all over the place. I do know more than enough about narrow vein mining, having worked about six years at mostly shrinkage stope mines.

      It cannot be mined selectively in a reasonable way. The drill steel is usually 8 ft, and the miner cannot know which way the vein is curving, so ends up drilling way into the waste and leaving ore in the other wall, etc. and the labour cost per tonne in those narrow inefficient difficult locations becomes absurd. The company has been doing everything correct, I think, and that suggests to me that all is going to work out ok there.

      Paths (stockhouse id),

    5. You point out accurately why long hole would be problematic for selective mining with no linear continuity. You also point out accurately that some high grade blocks are isolated ... and those on the periphery of the VOK are therefore more risky as I have stated. The blocks are statistical only, that is why they have a range of grades assigned. In reality, the actual grades are going to be much more skewed and only by mixing in a bunch of waste (because they are mining on a bulk scale not selectively) will they come up on average somewhere in the middle. If they mined selectively based on probability from conditional simulation, they would get just high grades, and a lot of areas would not get mined (even though they have a bit of high grade in those areas, there is likely to be too much waste i.e. 1-2g/t rock).

  8. Paths, you have never obviously worked at or operated a high tonnage rate underground mine so I find your arguments to be very academic. MIK underground is a disaster and the engineers of the fs were well aware of this year's ago but were handed the Snowden block model so used it. Perhaps head grade will settle down to about 7 g/t but those attempting to defend 14-16 g/t in the long run are fools. Doug Beattie ocotilloredux

    1. It will be interesting to see what impact the plant expansion will have on the head-grade. Will Pretium be able to keep it at ~9-10 g/t or will be see a drop down to 7-8 g/t once they ramp up to 3800 tpd?

    2. Continual geostatistics underground would be awkward and affect schedules, however, the plan all along was to go that route. That is the only way to operate as a success. The other option would be to ignore all that as being too much bother, and operate at break even. Obviously correct choices have been made.

      I am not attempting to defend 14-16 g/t, as the company has stated the results are below modeled so far. However, I see 10, 11, maybe 12 g/t as being a reasonable expectation, so I am inclined to believe the guidance that was given. That would be fine, especially at 3800 tpd. I can't see a basis for all the many negative views that have been expressed.

      Paths/midpath, stockhouse/twitter