Doug Ramshaw

Why I Don't Like $MAI.V Since several people asked why I don’t like $MAI.V after a tweet I made last week, I decided to write this. It turned out much longer than I expected, but I found a lot of "interesting" details in the technical reports. I really encourage you to start reading technical reports yourself, they can be quite telling. In my opinion, MAI has two of the worst technical reports I’ve come across. They’re easily in my top five. I originally stopped doing due diligence on the company about a year ago after spotting two or three major red flags. This time around, I found even more. When two of a company’s mines raise this many concerns, I usually stop there. So no, I don’t have a view on the geological upside at the Pan Mine. Ask a geologist. To be clear, I’m not making a call on the stock’s future. If you believe management can address the issues and execute, that’s your decision. But as it stands, the company has two lower-tier mines in the U.S., producing around 100,000 ounces per year at an estimated $1,900 AISC blended. Even at $4,000 gold and a 25 percent tax rate, that’s about $195 million in free cash flow annually. Then there’s Gold Rock, which, in my view, is another mediocre asset. With a $500 million market cap, it’s not cheap compared to other miners, at least in my opinion. Things really start to fall apart when you look at their Mexican portfolio. Let’s just say there’s a reason La Fortuna isn’t mentioned in the new investor presentation. As for Cerro de Oro, it looks more like the illusion of a mine than a real project. Santana? I don't even know what's going on there. There has been issues for many years, and management has denied there's a water issue for a long time now. La Fortuna, we'll look at that first. Recent CEO Comments on La Fortuna "...and we rarely talk about La Fortuna despite what is a phenomenal deposit. It's a three and a half to four gram open pit, six to one strip, and it has fantastic economics." They might very well be right, but lets examine the "technical reports," and then you can call me a moron for not being long afterwards. La Fortuna, Where Is My Core? CSA Global did the tech report. The most glaring issue: "the physical evidence for the majority of the drill database effectively no longer exists. The 1994 San Fernando drilling campaign accounts for 59% of the data used in the resource estimate, but the core is mostly gone." If you ain't got no core, you have no QRD, so you don't know geotechnical stability or rock mechanics with any confidence. Therefore, you can't do proper mine design or slope design. This lead to: "On 10 July 2016, Scott Zelligan, P.Geo., visited the Project, accompanied by Miguel Cardona, P.Eng., Vice President of Exploration for Minera Alamos. Level 2 of the underground adits was visited and investigated in order to validate rock descriptions of the mineralized zone, as limited core is currently available." And then: "The density was set at 2.65 t/m³, based on a conservative rounding down from averaged density studies." So let me get this straight: they averaged mineralized rock (2.72 t/m³) and waste rock (2.61 t/m³), then just picked a number in the middle that felt safe; 2.65 t/m³. This is an estimation of an estimation. Great. This is a paper resource validated by a bloke entering the adit and kicking some rocks around. How they managed to produce an NI 43-101 compliant resource where the core was largely gone is beyond me, but okay! The Pickaxe Problem "San Fernando's program included detailed mapping and sampling of underground workings, including the verification sampling of selected CRM underground channel sample locations. The CRM sample locations selected for verification sampling were chosen at random. San Fernando's samples were cut by pick rather than moil and were semi-continuous across the same CRM channel widths. The samples were fire assayed with an AA finish by Rossbacher Labs in Vancouver. Sample weights were less than CRM's, in part accounting for the obvious grade variances (San Fernando's were higher grade). Gunn, of Dupont Exploration Canada, carried out check sampling on the Property and concluded that the variation between the San Fernando and CRM assays may be due to nugget effect and that large samples are necessary for accurate results." A tool will follow the path of least resistance, so if your mineralized rock might be softer it breaks away more easily; your sample bag might end up filled with a disproportionate amount of gold bearing material. Trust me bro, it's just the nugget effect though! As you can read above, at La Fortuna, this resulted in samples that were "higher grade" than those taken with proper tools. Independent Check Assays: A Comedy of Errors "For holes LF1 to LF78, independent check samples were sent to Min-En Laboratories Ltd in Vancouver for analysis (technique not available/described). For holes LF79 to LF119, check samples were sent to Bondar-Clegg in Vancouver for analysis by fire assay using a 30g aliquot, with an LDL of 5 ppb. Fluor (1995) states that 31 check samples were sent to Min-En from Bondar-Clegg, and 192 check samples (both rejects and pulps) were sent to Bondar-Clegg from SGS. Currently, data is only available for one set of 150 of the latter, and it's not clear whether the set available is from rejects or pulps." So they have a spreadsheet of numbers, but they don't actually know if those numbers came from the crushed rock (rejects) or the powdered rock (pulps). This makes it impossible to know if the error is in the lab or the sample prep. They have collected some data, but it's not reliable enough to trust the grades as a true representation of the deposit. Any resource model built on this data is at risk of being wrong; likely overestimating grade and economic potential. So, taken all this into consideration, what level of confidence would you place in the deposit? Yes, Measured, of course; same conclusion the technical guys came to. The authors still classified a massive portion of the deposit as Measured (the highest level of confidence). They classified 1.75 million tonnes (containing 167,000 oz Au) as Measured. The core is missing, so we checked the tunnel walls The old sampling was done with a pickaxe and came back higher than usual We don't know if the check samples were dust or rocks We guessed the density by rounding down We call this a Measured Resource Cheers How Did They Justify This? Despite the physical core being missing, the QP deemed the 1990s paper data sufficient to support a Measured classification because they decided the paperwork looked good. The report asserts that the 1994 San Fernando drilling campaign (which represents 59% of the data) was "thoroughly modern in its methods, including QAQC procedures" and that the "data is reliable." The other argument? Since Castle Gold drilled six twin holes in 2008 to verify 1990s data, this somehow verifies the 121 historical drill holes. Combining all of this and using the tight drill spacing of 15m, they argue that geological continuity is confirmed even if the core is gone. According to CIM Definition Standards (referenced in almost every NI 43-101 report), a Measured Mineral Resource requires that: Geological evidence is derived from "detailed and reliable exploration, sampling and testing" It must be sufficient to "confirm" (not just assume) geological and grade continuity We are not done yet though; the metwork is also interesting. They assume 90% gold recoveries; the crush gravity and float is standard but to make economics better they will install a "Dual X-ray DEXRT sorter." They assume they can take low-grade/waste rock (0.8–2.0 g/t), scan it with X-rays, reject the waste, and upgrade the rock to 3.5–4.0 g/t Au with 80% recovery. It's difficult to claim your resource is "reliable" when the primary physical evidence (drill core) is missing, the sampling method (pickaxe) is biased and the historical assay results are a mess as well. It's standard practice to downgrade legacy holes or disregard them entirely. Oh and the tonnage you use to calculate grade is guessed as well. Cerro de Oro: The "Company Maker" The new CEO recently said during a town hall that Cerro De Oro is a "company maker" and explicitly stated that the costs in the PEA were "informed not by some database of project data that Consultants come up with" but by actual operating costs from their Santana mine. We have technical studies, to protect us from promotional management teams. But okay. Based on the 2022 Cerro de Oro report, the situation is mental in its own way but there is a bit less cave exploration here. The entire economic model is built on 100% Inferred Resources though (the lowest confidence category), largely because they didn't measure the density of the rock; they just "borrowed" the numbers from a different mine. WTF? So, tonnage is volume x density; if the density is wrong then your production forecast is bullshit. The standard practice is to send core samples to a lab to measure the specific gravity of your rock. However in the tech report: The Company reported that there are no density measurements in the drill-hole cores, the available density is measured only on surface samples, which are not representative of the deposit. The author used average values of 2.55 t/m³ for oxides and 2.70 t/m³ for fresh rock, which are acceptable values considering average values obtained in similar deposits, such as Castle Gold's El Castillo mine. How did they justify this? It's not noted but I surmise they chose this mine because the El Castillo mined used to be own by Castle Gold Corp. "Mr. Koningen was a founding member and VP Mine Development of Castle Gold Corp. He is now the CEO of Minera Alamos and that mine is 300+ km away btw. If the rock at Cerro de Oro is even 5% lighter than the rock at El Castillo, 5% of the gold resource instantly evaporates. Just because two mines are in the same region or have similar geology does not mean the rock has the exact same specific gravity. Weathering profiles and porosity vary wildly between deposits. Inferred resources are considered "too speculative geologically to have economic considerations applied to them" yet the entire PEA is based on gold that may or may not be there. The Metallurgical Testwork The metwork was done on RC chips as well which is not ideal as it underestimates the time required for cyanide to penetrate real rocks; RC chips are <2mm, this is dust and leaches very easily. They plan to stack 22mm rock so let's just hope it holds I guess. They assume a 68% Gold Recovery for the Life of Mine which is not egregious at all but the testwork is too limited to say right now. "Management's experience with similar low-grade operations guided the use of a conservative recovery estimate that is in the high 60s range until further testwork can be completed." Who needs testing when you got experience I guess. They also do something else that is quite interesting here: "The testwork program stages were somewhat unconventional and consisted of the following: Screening of sample to remove −10 mesh fines that were rejected (30%–40% of original sample). Crush remaining coarse sample material to minus 2" top size." No clue why; usually fines would help the recoveries but the sum here is that the actual chemistry of the heap leach has not been tested properly. For the transition/sulphide material (which often behaves very differently from oxides), they relied on a single bottle roll test performed in 2020: "A single sample of transition/sulphide mineralization was selected and sent to LTM laboratories in Hermosillo, Mexico, for standard bottle roll testwork to examine the potential for gold dissolution in August 2020. The sample consisted of RC chips taken from drill hole CR18-40, in the South zone of Project area." Compare this with the Viva Gold PEA where they performed column tests on master composites generated from 156 bags of material to test cement addition (agglomeration) to ensure the heap would not collapse or plug up. Here they tested drilling dust and removed the fines from their rocks? Then they based the sulphide assumption on a single bucket of chips. Just like La Fortuna there has been zero geotech drilling so they assume a "standard" 45 degree angle in their pit. "The QP is of the opinion that the analytical work performed by the various laboratories was suitable for use in the Mineral Resource estimation." "The assumptions, parameters, and methodology are appropriate for the Mineral Resource estimation, are consistent with the style of mineralization, and are applicable for an open pit and heap leach operation." "The QP has classified the current Mineral Resource estimation as Inferred for the oxidized portion of the Cerro de Oro deposit. Although drill spacing is locally sufficient for Indicated classification, there remain necessary revisions and updates to the geological logs; better definition of the limit between the oxides and fresh rock; understanding of mineralization controls; and bulk density measurements. The reported Inferred Mineral Resources are estimated with an average drilling grid of 85 m by 85 m." We Are Not Done Yet Though! The report states: "The drilling database includes gold assays for the Cerro de Oro deposit, but the lithological descriptions are not available." So since they have no logs the QP states: "The absence of the lithological descriptions in the drill holes prevents a better geometric control of the spatial distribution of the gold grades, and the limit between oxides and fresh rock is not clearly defined. These inconsistencies in the drilling database should be reviewed and repaired." As such: "The domains used for this estimation were primarily driven by the gold grade distribution. This is due to the lack of reliable geological data, preventing the inclusion of geological limits in the geometric definition of mineralization." Let's hope they don't include any sulphide rock in the oxide resource but should be fine regardless since we have a single sample of said sulphides. The "Cerro de Oro" Approach: Don't measure the density: Use the numbers from the CEO's old mine (El Castillo). Keep the resource "Inferred" because the geological data is messy and then design a pit and mine around it. Don't crush rock: Test the drilling dust (RC chips) and assume the big rocks will behave the same way. Don't test the walls: Assume 45 degrees. The CEO calls this a "company maker" partly based on the 0.2:1 strip ratio, but that ratio is derived from a pit shell designed on assumed slopes and assumed density. If those assumptions are wrong, the strip ratio; and the economics; change immediately. Stack 10 assumptions on top of each other and then you add the built-in inaccuracy of even a great PEA and your degrees of freedom are so large the reports are not worth the paper they are written on. These aren't minor gaps you can hand wave away with management experience. These are fundamental deficiencies that would make any serious investor or financier pause. The fact that Cerro De Orro managed to secure a $25M financing package is baffling to me. Both mines need significant work. Why did they buy the Pan mine if these assets are so great? Why did "Minera Alamos Signs Definitive Documentation for Cerro De Oro Construction Financing Package of up to US$25 million and Completes Initial Drawdown" in October 2023 and the mine is yet to be built? Permitting I guess Do with this information what you will. This information is available on SEDAR+ for anyone that wants to do some reading. I'm not criticizing anyone at $MAI.V, just highlighting what's disclosed in the technical reports. Make up your own mind. This is simply why, personally, I don't find $MAI.V to be an attractive investment.