Introduction
At Crux Investor, our Analyst’s Notes are guided by a healthy scepticism. We cautiously approach official company statements, carefully reading what is said and what is left unsaid. Our analysis aims to look beyond promotional narratives to identify omissions, inconsistencies, and overlooked risks that may otherwise go unnoticed.
Back in February 2021, we published a Note under the headline “How to Read Drill Results & Numbers to Watch Out For”. Our analysts pointed out that, under the NI 43-101 Disclosure Rule, company news releases in which technical information is disclosed have to be:
- Balanced and not misleading
- Understandable to a reasonably informed investor
- Consistent in its use of standardised terms and definitions
- Based on reasonable assumptions which are clearly explained
- Unbiased and identify the potential risks and uncertainties
- Signed off by a professional (QP) who takes responsibility for the information
In practice, this is often not the case.
Companies in such press releases go to great lengths to present the information in the best light. It is understandable that the projects are promoted by highlighting the positives; management will not necessarily misrepresent the facts, but tend not to give the whole truth and nothing but the truth. For this reason, the reader should be sceptical and focused on what IS NOT being said. That article gave several examples of what to look for in drill results reports.
This Analyst’s Notes will expand on the reporting of drill results to the reporting of mineral resource estimations (“MRE’s”) and has taken the latest updated MRE of African Gold as an example. African Gold is an Australian-listed company and, as such, does not have the obligation to publish a full technical report supporting its MRE, unlike Canadian-listed companies. Not having such an obligation provides additional opportunities to avoid providing full disclosure. Crux Investor is of the opinion that the ASX authorities should look into this limited obligation as it is not to the advantage of the market.
Background to African Gold
The company's market capitalisation (ASX:A1G) on 11 July 2025 at a share price of A$0.215 was A$113 million; with diluted Enterprise Value at approximately the same level A$115 million (US$76 million). With the share price having had a remarkable rerating since mid 2024 (see below) all options are in the money. With the June 2025 cash injection from Montagne Gold (now with a 17.5% interest) and a private placement of A$2.7 million, the company has enough money to cover the annualised cash burn of the last nine months of A$1.5 million many times over. The low burn rate is partially explained by drilling expenses settled by issuing shares since November 2024.
The company focuses on the Didievi gold project in Ivory Coast. This 391 km2 property was acquired early in 2021. It is located in the Oume-Fetekro Birimian greenstone belt. The belt hosts Allied Gold’s Bonikro/Hire (+3 Moz) and Endeavour’s Agbaou (+1 Moz) gold mines to the south of Didievi and the Fetekro discovery (+2.5 Moz) announced by Endeavour Mining to the north. The project is therefore at a very good address. Below is the location of Didievi, which is shown relative to other mineral properties held by the company in 2024.
The company has since withdrawn from Mali, but after entering in October 2022 into an earn-in agreement with Reunion Mining Limited, the company owning the Syama mine, on their Syama shear zone project. Reunion can earn 80% of this project by spending US$0.5 million and commencing a feasibility study (“FS”) within five years and completing this within another three years. Transfer of the 80% ownership will follow a positive Decision to Mine, which must be made within 90 days following the completion of the FS. The annual reports of Reunion following 2022 are quiet on this project. Given the five-year period, Reunion is under no pressure to start spending the US$0.5 million.
In late 2020, Africa Gold announced it had entered into an option agreement to acquire 80% of the Kouroufaba gold project, which comprised land holdings consisting of the Didievi permit and three other permit applications totalling 1,534 km2 covering Birimian greenstone belts in Central Côte d’Ivoire (see the map above). Diviedi had previously been drilled by Equigold, Lihir Gold, and Newcrest, but none of them deemed it attractive enough to hang onto the project.
Below is the tenement map for Didievi in the 2024 Annual Report.
In 2021, initial drilling was carried out at the Blaffo Guetto target, which previous exploration work had proven to contain gold mineralisation. Drilling the area in the east underlain by granite proved to be barren of gold, allowing the company to drop 25% of the permit area upon renewal, as required.
According to the 2022 Annual Report, drilling during the year at Blaffo Guetto “delivered some spectacular broad, shallow and high-grade gold intercepts from surface”. A statement based on the results of one particular hole: DDD044.
During 2023, auger drilling was carried out along the Poku trend to identify drill targets. Within the trend, an anomaly with 900 m strike length grading ≥ 0.15 g/t Au was defined, with a peak value of 0.845 g/t Au. During that year, trenching tested the Kouassi target located 14 km NNE of Diviedi.
Away from Diviedi, at Konahiri Nord (refer again to Figure 3), composite track sampling during 2023 identified an anomaly with a +100 ppb Au on the permit's eastern border over a strike length of 18 km.
At Agboville, large outcropping pegmatite intrusions were identified with coincident lithium, caesium, tantalum and rubidium anomalism in stream sediment samples. Given the decline in lithium prices, it is understandable that this project seems to have lower priority now.
On 30 July 2024, a maiden MRE was announced for Blaffo Guetto with 0.45 Moz at a very attractive grade of 2.9 g/t Au, classified as Inferred. This MRE is probably the reason for the re-rating of African Gold’s share price.
During the March 2025 quarter, the company got a good endorsement from Montagne Gold, which intended to take a 19.9% strategic stake in the company by a share swap of A$6.7 million and A$0.9 million share placement. Africa Gold also announced a “large, high-potential gold system” at the Poku Trend, but the intersections highlighted only graded 0.5 g/t Au, which can only be of interest if amenable to heap leaching. The focus, however, remained on Blaffo Guetto with borehole results over wide intervals reported for four holes (with big gaps in borehole numbering).
Montagne Gold's strategic investment was completed on 13 June 2025, but the stake is 17.5%, not the 19.9% previously announced.
On 23 June 2025, an updated MRE was announced, with 0.90 Moz in Inferred resources at a grade of 2.5 g/t Au. The sections below will review this MRE and raise a number of concerns.
The Biaffo Guetto Prospect
Background
The gold mineralisation of the Blaffo Guetto prospect occurs as a set of closely spaced steeply dipping lodes controlled by the set of shear zones and their splays. 41 lodes have been interpreted on the cross-sections and delineated by constraining the defined intersections into 3D wireframes. This interpretation is based on the geological mapping of the prospect area and logging of the drillholes, which are distributed approximately on a 50 m x 40-30 m grid. Notably, the number of boreholes drilled pre-2021 was 203, and post-2021 is only 34. It raises the question of how the mineral resources could increase so dramatically from July 2024.
Below in Figure 4 is a plan view of Biaffo Guetto followed by a schematic cross-section showing the lode structures, some of which are very narrow and generally trend NE-SW. However, very confusingly, the legend of the plan view shows that the somewhat darker pink colour represents “mineral resource contours.” This does not make sense and is not repeated in the cross-section.
MRE Methodology
The methodology and results of the MRE are presented in a news release dated 23 June 2025 under the title “Blaffo Guetto’s Inferred Resource Surges 119% to 989,000 oz within 12.4 million tonnes at 2.5 g/t Au”
The gold lodes were initially interpreted on the cross-sections by defining the mineralised intervals where grade is not less than 0.2 g/t Au and thickness >2 m. These intercepts were correlated between cross-sections, presenting a set of continuous mineralised zones. “Correlation of the defined intersections between cross-sections was supported by geological interpretation of the host greenstone succession. Interpreted lodes were constrained by the 3D wireframes”.
An important observation is made as follows: “constraining the mineralised zones by the 3D wireframes has revealed the presence of the low-grade and waste parts within the mineralised lodes, indicating the patchy and discontinuous structure of the gold distribution within the host mineralised zone”. Crux Investor finds this a very concerning fact, putting a large question mark behind the validity of the wireframes.
The samples constrained by the wireframes have been composited to 1 m composites. The top-cut values were defined for each lode by finding a ragged tail on the Cumulative Frequency diagram and analysing the impact of high-grade cutting (capping) on the mean of the data population. A table shows the impact of grade cutting on the mean and variance (not standard deviation, as is the norm). Calculating the coefficients of variance (CVs = standard deviation/mean) from the numbers shows these to be generally low, which is good.
Each of the 41 lodes (wireframes) was estimated separately using the samples belonging to the estimated wireframe. Estimation was made into 20 m x 20 m x 1 m parent blocks that were infilled into the wireframes, constraining the lodes. The parent blocks were partitioned into 2 m x 2 m x 1 m, the size of the sub-cells reflecting the narrow nature of many lodes.
Mineralisation was constrained by wireframes, and the Au grade was estimated using the Inverse Distance of Power 3 (“ID3”) method. Part of the lodes were estimated using Ordinary or Simple Kriging.
The drillhole data and the block model was unfolded (flattened), using an equal thickness flattening algorithm of the Micromine 2024© (Crux Investor observation: such detail does not add anything to the reader’s insight as it refers to a black box methodology. The report should rather explain the principles of the methodology). Blocks and the drillhole data were flattened to 5 m equal thickness layers. Geostatistical data analysis and estimations were made in the unfolded space, which, after completion, was transferred back to the original (UTM) coordinates.
The search ellipse, used for the ID3 estimation, was defined in the unfolded space. Ellipsoid radii were 70 m (main axis, Azi 140o) x 40 m (semi-, Azi 230o) x 3 m (minor, vertical), which are consistent with the variogram ranges estimated for the lodes 177 and 178. Crux Investor records that no semi-variograms are presented in support of these chosen radii and to demonstrate the degree by which the curves are supported by data points.
As validation of the MRE, the estimated block grades were compared with corresponding drillhole sample grades. The method compares drillhole data and block grades by grouping them into large panels drawn along the strike of the lodes and, where appropriate, in the vertical direction. Crux Investor much prefers cuts through the block model that compare the block grades with borehole intersection grades. This will give the reader a much better impression of the reasonableness of block grades and extensions of such blocks to neighbouring borehole intercepts. There are numerous examples of MRE exercises extrapolating high-grade blocks to be immediately next to low-grade intersections.
The mineral statement - all in the Inferred category, but without a precise effective date - at a cut-off grade of 0.8 g/t Au is reproduced in Table 1 below.
Crux Investor Commentary to the MRE
The discussion of the MRE is, in the opinion of Crux Investor, deficient for the following reasons:
- The exact nature of the wireframing is nowhere properly explained and demonstrated by cross sections with borehole intersections;-
- No variograms are presented in support of the search radii;-
- No cuts through the block model are presented, comparing borehole intersection assay with block grades. This is, in the opinion of Crux Investor, the best method to verify the validity of the MRE;-
- Nowhere is it explained how the very attractive average MRE grade reconciles with the generally poor drill hole intersections reported in previous press releases (see the discussion below);-
- Nowhere is it explained how an additional 34 holes above the 203 holes available for the MRE one year earlier could result in a more than doubling of mineral resources. Less than 16% additional holes cause a doubling in resources?
The MRE in the Context of Previous Press Releases
What is striking about the discussion of the MRE is the absence of cross sections with boreholes intersecting the lodes and providing context to reported intercepts and how well these correlate. Crux Investor has reviewed all ASX announcements since the release of the previous MRE on 30 July 2024. Only four announcements were released with drill results, almost exclusively showing the results on plans and longitudinal sections. The 7 October 2024 press release is one of only two announcements that include a cross-section, but reporting on only one borehole (see below).
Note how poorly the lode interpretations are supported by the borehole assay values, with the lode to the east completely devoid of gold. A longitudinal section in the same ASX announcement shows four pierce points with results that are highlighted, surrounded by 17 other pierce points with results lower than 1 g/t Au.
In the 30 January 2025 ASX announcement, a longitudinal section is shown with a weird combination of intersections as bars and pierce points, which is repeated in the press release dated 31 March 2025 (see below).
Despite the highly promotional and speculative interpretations above, what is evident is the numerous low-grade pierce points between the impressive grades. However, with numerous parallel lode structures, this does not demonstrate that these impressive grades are in the same structure/lode.
The ASX announcement of 31 March 2025 promotes the high-grade trend further by providing the cross section below of the “high-grade terrace/tubular zone.”
The interpretation is based on two boreholes, with the company ignoring, for unexplained reasons, borehole 61 immediately north of borehole 60. There is no geological support for a “terrace” or “tube”, and no explanation is provided for what could control this. Crux Investor observes that the same cross-section profile would be achieved if the company intersected mineralisation controlled by a NW-SE cross-cutting structure. However, this would imply high grades over a limited strike extent along the main structural NE-SW control.
This kind of cross-cutting control is very much implied in the plan below, extracted from the latest corporate presentation. Crux Investor has annotated the slide with two blue lines connecting the holes highlighted as having outstanding results.
Finally, Crux Investor has attempted to find a press release with a comprehensive set of results, not just highlighted results. The JORC code itself prescribes under point 17:
Reporting of selected information such as isolated assays, isolated drill holes, assays of panned concentrates or supergene enriched soils or surface samples, without placing them in perspective is unacceptable.
And elaborates on this in its Table 1:
Where comprehensive reporting of all Exploration Results is not practicable, representative reporting of both low and high grades and/or widths should be practiced to avoid misleading reporting of Exploration Results.
The most recent ASX announcement with such information is dated 30 January 2025 under the headline: “First Assays from Transformational 10,000 m Drill Program Deliver Outstanding Results”. Please refer below to Table 2 and 3, which reproduce the results in this announcement and which Crux Investor has annotated by drawing blocks around reported wider intercepts. Please note that only ONE intersection has an excellent intersection. All others are either very narrow and/or very low-grade. Note that several intersections are shown as having 0.0 g/t Au! In any case, wide intersections should not be announced as the type of mineralisation is supposed to be high grade in narrow, sub-vertical lodes.
Conclusions and Recommendations
As for drill results, the announcement of mineral resources should not be accepted at face value, but critically reviewed to determine whether or not it is reasonable. It is up to the reporting company to make the case that the MRE is well supported by being consistent with the geological controls of the mineralisation, that the averaging method is suitable, the search parameters are well supported and the result verified in a manner that allows the reader to visually compare block grades and borehole intersection grades and to what extent block grades have been extrapolated beyond where there is support for these.
Some specific issues to consider when reading the explanation of the MRE methodology:
- Is the MRE consistent with exploration results previously announced?
- Does the wireframing reflect the geological control(s)?
- Has grade capping for precious metals been performed to an extent that drops the CV to preferably below 2.0? Companies are reluctant to cut grades aggressively as it will affect the average grade they will be able to report. However, averaging techniques such as ordinary kriging require a grade population that is not excessively skewed. A reluctance to cut grades should make the reader extra cautious. It is even more concerning when the company elects to use averaging methods such as multiple indicator kriging (“MIK”) that do not cut grades at all. History has shown that it leads to a clear overestimation of average grades. Examples are Brucejack, Sukari and Moose River.
- Looking at the various averaging techniques to estimate block grades, the reader should look beyond the technical jargon and realise these methods are all similar, just use a different approach to weighing the influence of assays on the average grade. The MRE discussion must make the case that the input parameters used for the weights are well supported. The results of the variography in the form of tables do not support this. Preferably, many visual examples of the variography should be given for the search directions.
- The input parameters for the reasonable prospect for eventual economic extraction (“RPEE”) should be well motivated. Low unit cost assumptions are typically provided for mining, processing, and general and administration costs. Often, the royalty obligation is ignored. The result cannot be considered credible when metallurgical recoveries are assumed based on little to no testwork.
- Finally, it is up to the MRE report to demonstrate a reasonable correlation between block grades and the assays of the samples used to estimate these. The more cuts through the block model together with borehole results to illustrate this, the better.
Introduction
At Crux Investor, our Analyst’s Notes are guided by a healthy scepticism. We cautiously approach official company statements, carefully reading what is said and what is left unsaid. Our analysis aims to look beyond promotional narratives to identify omissions, inconsistencies, and overlooked risks that may otherwise go unnoticed.
Back in February 2021, we published a Note under the headline “How to Read Drill Results & Numbers to Watch Out For”. Our analysts pointed out that, under the NI 43-101 Disclosure Rule, company news releases in which technical information is disclosed have to be:
- Balanced and not misleading
- Understandable to a reasonably informed investor
- Consistent in its use of standardised terms and definitions
- Based on reasonable assumptions which are clearly explained
- Unbiased and identify the potential risks and uncertainties
- Signed off by a professional (QP) who takes responsibility for the information
In practice, this is often not the case.
Companies in such press releases go to great lengths to present the information in the best light. It is understandable that the projects are promoted by highlighting the positives; management will not necessarily misrepresent the facts, but tend not to give the whole truth and nothing but the truth. For this reason, the reader should be sceptical and focused on what IS NOT being said. That article gave several examples of what to look for in drill results reports.
This Analyst’s Notes will expand on the reporting of drill results to the reporting of mineral resource estimations (“MRE’s”) and has taken the latest updated MRE of African Gold as an example. African Gold is an Australian-listed company and, as such, does not have the obligation to publish a full technical report supporting its MRE, unlike Canadian-listed companies. Not having such an obligation provides additional opportunities to avoid providing full disclosure. Crux Investor is of the opinion that the ASX authorities should look into this limited obligation as it is not to the advantage of the market.
Background to African Gold
The company's market capitalisation (ASX:A1G) on 11 July 2025 at a share price of A$0.215 was A$113 million; with diluted Enterprise Value at approximately the same level A$115 million (US$76 million). With the share price having had a remarkable rerating since mid 2024 (see below) all options are in the money. With the June 2025 cash injection from Montagne Gold (now with a 17.5% interest) and a private placement of A$2.7 million, the company has enough money to cover the annualised cash burn of the last nine months of A$1.5 million many times over. The low burn rate is partially explained by drilling expenses settled by issuing shares since November 2024.
The company focuses on the Didievi gold project in Ivory Coast. This 391 km2 property was acquired early in 2021. It is located in the Oume-Fetekro Birimian greenstone belt. The belt hosts Allied Gold’s Bonikro/Hire (+3 Moz) and Endeavour’s Agbaou (+1 Moz) gold mines to the south of Didievi and the Fetekro discovery (+2.5 Moz) announced by Endeavour Mining to the north. The project is therefore at a very good address. Below is the location of Didievi, which is shown relative to other mineral properties held by the company in 2024.
The company has since withdrawn from Mali, but after entering in October 2022 into an earn-in agreement with Reunion Mining Limited, the company owning the Syama mine, on their Syama shear zone project. Reunion can earn 80% of this project by spending US$0.5 million and commencing a feasibility study (“FS”) within five years and completing this within another three years. Transfer of the 80% ownership will follow a positive Decision to Mine, which must be made within 90 days following the completion of the FS. The annual reports of Reunion following 2022 are quiet on this project. Given the five-year period, Reunion is under no pressure to start spending the US$0.5 million.
In late 2020, Africa Gold announced it had entered into an option agreement to acquire 80% of the Kouroufaba gold project, which comprised land holdings consisting of the Didievi permit and three other permit applications totalling 1,534 km2 covering Birimian greenstone belts in Central Côte d’Ivoire (see the map above). Diviedi had previously been drilled by Equigold, Lihir Gold, and Newcrest, but none of them deemed it attractive enough to hang onto the project.
Below is the tenement map for Didievi in the 2024 Annual Report.
In 2021, initial drilling was carried out at the Blaffo Guetto target, which previous exploration work had proven to contain gold mineralisation. Drilling the area in the east underlain by granite proved to be barren of gold, allowing the company to drop 25% of the permit area upon renewal, as required.
According to the 2022 Annual Report, drilling during the year at Blaffo Guetto “delivered some spectacular broad, shallow and high-grade gold intercepts from surface”. A statement based on the results of one particular hole: DDD044.
During 2023, auger drilling was carried out along the Poku trend to identify drill targets. Within the trend, an anomaly with 900 m strike length grading ≥ 0.15 g/t Au was defined, with a peak value of 0.845 g/t Au. During that year, trenching tested the Kouassi target located 14 km NNE of Diviedi.
Away from Diviedi, at Konahiri Nord (refer again to Figure 3), composite track sampling during 2023 identified an anomaly with a +100 ppb Au on the permit's eastern border over a strike length of 18 km.
At Agboville, large outcropping pegmatite intrusions were identified with coincident lithium, caesium, tantalum and rubidium anomalism in stream sediment samples. Given the decline in lithium prices, it is understandable that this project seems to have lower priority now.
On 30 July 2024, a maiden MRE was announced for Blaffo Guetto with 0.45 Moz at a very attractive grade of 2.9 g/t Au, classified as Inferred. This MRE is probably the reason for the re-rating of African Gold’s share price.
During the March 2025 quarter, the company got a good endorsement from Montagne Gold, which intended to take a 19.9% strategic stake in the company by a share swap of A$6.7 million and A$0.9 million share placement. Africa Gold also announced a “large, high-potential gold system” at the Poku Trend, but the intersections highlighted only graded 0.5 g/t Au, which can only be of interest if amenable to heap leaching. The focus, however, remained on Blaffo Guetto with borehole results over wide intervals reported for four holes (with big gaps in borehole numbering).
Montagne Gold's strategic investment was completed on 13 June 2025, but the stake is 17.5%, not the 19.9% previously announced.
On 23 June 2025, an updated MRE was announced, with 0.90 Moz in Inferred resources at a grade of 2.5 g/t Au. The sections below will review this MRE and raise a number of concerns.
The Biaffo Guetto Prospect
Background
The gold mineralisation of the Blaffo Guetto prospect occurs as a set of closely spaced steeply dipping lodes controlled by the set of shear zones and their splays. 41 lodes have been interpreted on the cross-sections and delineated by constraining the defined intersections into 3D wireframes. This interpretation is based on the geological mapping of the prospect area and logging of the drillholes, which are distributed approximately on a 50 m x 40-30 m grid. Notably, the number of boreholes drilled pre-2021 was 203, and post-2021 is only 34. It raises the question of how the mineral resources could increase so dramatically from July 2024.
Below in Figure 4 is a plan view of Biaffo Guetto followed by a schematic cross-section showing the lode structures, some of which are very narrow and generally trend NE-SW. However, very confusingly, the legend of the plan view shows that the somewhat darker pink colour represents “mineral resource contours.” This does not make sense and is not repeated in the cross-section.
MRE Methodology
The methodology and results of the MRE are presented in a news release dated 23 June 2025 under the title “Blaffo Guetto’s Inferred Resource Surges 119% to 989,000 oz within 12.4 million tonnes at 2.5 g/t Au”
The gold lodes were initially interpreted on the cross-sections by defining the mineralised intervals where grade is not less than 0.2 g/t Au and thickness >2 m. These intercepts were correlated between cross-sections, presenting a set of continuous mineralised zones. “Correlation of the defined intersections between cross-sections was supported by geological interpretation of the host greenstone succession. Interpreted lodes were constrained by the 3D wireframes”.
An important observation is made as follows: “constraining the mineralised zones by the 3D wireframes has revealed the presence of the low-grade and waste parts within the mineralised lodes, indicating the patchy and discontinuous structure of the gold distribution within the host mineralised zone”. Crux Investor finds this a very concerning fact, putting a large question mark behind the validity of the wireframes.
The samples constrained by the wireframes have been composited to 1 m composites. The top-cut values were defined for each lode by finding a ragged tail on the Cumulative Frequency diagram and analysing the impact of high-grade cutting (capping) on the mean of the data population. A table shows the impact of grade cutting on the mean and variance (not standard deviation, as is the norm). Calculating the coefficients of variance (CVs = standard deviation/mean) from the numbers shows these to be generally low, which is good.
Each of the 41 lodes (wireframes) was estimated separately using the samples belonging to the estimated wireframe. Estimation was made into 20 m x 20 m x 1 m parent blocks that were infilled into the wireframes, constraining the lodes. The parent blocks were partitioned into 2 m x 2 m x 1 m, the size of the sub-cells reflecting the narrow nature of many lodes.
Mineralisation was constrained by wireframes, and the Au grade was estimated using the Inverse Distance of Power 3 (“ID3”) method. Part of the lodes were estimated using Ordinary or Simple Kriging.
The drillhole data and the block model was unfolded (flattened), using an equal thickness flattening algorithm of the Micromine 2024© (Crux Investor observation: such detail does not add anything to the reader’s insight as it refers to a black box methodology. The report should rather explain the principles of the methodology). Blocks and the drillhole data were flattened to 5 m equal thickness layers. Geostatistical data analysis and estimations were made in the unfolded space, which, after completion, was transferred back to the original (UTM) coordinates.
The search ellipse, used for the ID3 estimation, was defined in the unfolded space. Ellipsoid radii were 70 m (main axis, Azi 140o) x 40 m (semi-, Azi 230o) x 3 m (minor, vertical), which are consistent with the variogram ranges estimated for the lodes 177 and 178. Crux Investor records that no semi-variograms are presented in support of these chosen radii and to demonstrate the degree by which the curves are supported by data points.
As validation of the MRE, the estimated block grades were compared with corresponding drillhole sample grades. The method compares drillhole data and block grades by grouping them into large panels drawn along the strike of the lodes and, where appropriate, in the vertical direction. Crux Investor much prefers cuts through the block model that compare the block grades with borehole intersection grades. This will give the reader a much better impression of the reasonableness of block grades and extensions of such blocks to neighbouring borehole intercepts. There are numerous examples of MRE exercises extrapolating high-grade blocks to be immediately next to low-grade intersections.
The mineral statement - all in the Inferred category, but without a precise effective date - at a cut-off grade of 0.8 g/t Au is reproduced in Table 1 below.
Crux Investor Commentary to the MRE
The discussion of the MRE is, in the opinion of Crux Investor, deficient for the following reasons:
- The exact nature of the wireframing is nowhere properly explained and demonstrated by cross sections with borehole intersections;-
- No variograms are presented in support of the search radii;-
- No cuts through the block model are presented, comparing borehole intersection assay with block grades. This is, in the opinion of Crux Investor, the best method to verify the validity of the MRE;-
- Nowhere is it explained how the very attractive average MRE grade reconciles with the generally poor drill hole intersections reported in previous press releases (see the discussion below);-
- Nowhere is it explained how an additional 34 holes above the 203 holes available for the MRE one year earlier could result in a more than doubling of mineral resources. Less than 16% additional holes cause a doubling in resources?
The MRE in the Context of Previous Press Releases
What is striking about the discussion of the MRE is the absence of cross sections with boreholes intersecting the lodes and providing context to reported intercepts and how well these correlate. Crux Investor has reviewed all ASX announcements since the release of the previous MRE on 30 July 2024. Only four announcements were released with drill results, almost exclusively showing the results on plans and longitudinal sections. The 7 October 2024 press release is one of only two announcements that include a cross-section, but reporting on only one borehole (see below).
Note how poorly the lode interpretations are supported by the borehole assay values, with the lode to the east completely devoid of gold. A longitudinal section in the same ASX announcement shows four pierce points with results that are highlighted, surrounded by 17 other pierce points with results lower than 1 g/t Au.
In the 30 January 2025 ASX announcement, a longitudinal section is shown with a weird combination of intersections as bars and pierce points, which is repeated in the press release dated 31 March 2025 (see below).
Despite the highly promotional and speculative interpretations above, what is evident is the numerous low-grade pierce points between the impressive grades. However, with numerous parallel lode structures, this does not demonstrate that these impressive grades are in the same structure/lode.
The ASX announcement of 31 March 2025 promotes the high-grade trend further by providing the cross section below of the “high-grade terrace/tubular zone.”
The interpretation is based on two boreholes, with the company ignoring, for unexplained reasons, borehole 61 immediately north of borehole 60. There is no geological support for a “terrace” or “tube”, and no explanation is provided for what could control this. Crux Investor observes that the same cross-section profile would be achieved if the company intersected mineralisation controlled by a NW-SE cross-cutting structure. However, this would imply high grades over a limited strike extent along the main structural NE-SW control.
This kind of cross-cutting control is very much implied in the plan below, extracted from the latest corporate presentation. Crux Investor has annotated the slide with two blue lines connecting the holes highlighted as having outstanding results.
Finally, Crux Investor has attempted to find a press release with a comprehensive set of results, not just highlighted results. The JORC code itself prescribes under point 17:
Reporting of selected information such as isolated assays, isolated drill holes, assays of panned concentrates or supergene enriched soils or surface samples, without placing them in perspective is unacceptable.
And elaborates on this in its Table 1:
Where comprehensive reporting of all Exploration Results is not practicable, representative reporting of both low and high grades and/or widths should be practiced to avoid misleading reporting of Exploration Results.
The most recent ASX announcement with such information is dated 30 January 2025 under the headline: “First Assays from Transformational 10,000 m Drill Program Deliver Outstanding Results”. Please refer below to Table 2 and 3, which reproduce the results in this announcement and which Crux Investor has annotated by drawing blocks around reported wider intercepts. Please note that only ONE intersection has an excellent intersection. All others are either very narrow and/or very low-grade. Note that several intersections are shown as having 0.0 g/t Au! In any case, wide intersections should not be announced as the type of mineralisation is supposed to be high grade in narrow, sub-vertical lodes.
Conclusions and Recommendations
As for drill results, the announcement of mineral resources should not be accepted at face value, but critically reviewed to determine whether or not it is reasonable. It is up to the reporting company to make the case that the MRE is well supported by being consistent with the geological controls of the mineralisation, that the averaging method is suitable, the search parameters are well supported and the result verified in a manner that allows the reader to visually compare block grades and borehole intersection grades and to what extent block grades have been extrapolated beyond where there is support for these.
Some specific issues to consider when reading the explanation of the MRE methodology:
- Is the MRE consistent with exploration results previously announced?
- Does the wireframing reflect the geological control(s)?
- Has grade capping for precious metals been performed to an extent that drops the CV to preferably below 2.0? Companies are reluctant to cut grades aggressively as it will affect the average grade they will be able to report. However, averaging techniques such as ordinary kriging require a grade population that is not excessively skewed. A reluctance to cut grades should make the reader extra cautious. It is even more concerning when the company elects to use averaging methods such as multiple indicator kriging (“MIK”) that do not cut grades at all. History has shown that it leads to a clear overestimation of average grades. Examples are Brucejack, Sukari and Moose River.
- Looking at the various averaging techniques to estimate block grades, the reader should look beyond the technical jargon and realise these methods are all similar, just use a different approach to weighing the influence of assays on the average grade. The MRE discussion must make the case that the input parameters used for the weights are well supported. The results of the variography in the form of tables do not support this. Preferably, many visual examples of the variography should be given for the search directions.
- The input parameters for the reasonable prospect for eventual economic extraction (“RPEE”) should be well motivated. Low unit cost assumptions are typically provided for mining, processing, and general and administration costs. Often, the royalty obligation is ignored. The result cannot be considered credible when metallurgical recoveries are assumed based on little to no testwork.
- Finally, it is up to the MRE report to demonstrate a reasonable correlation between block grades and the assays of the samples used to estimate these. The more cuts through the block model together with borehole results to illustrate this, the better.
Introduction
At Crux Investor, our Analyst’s Notes are guided by a healthy scepticism. We cautiously approach official company statements, carefully reading what is said and what is left unsaid. Our analysis aims to look beyond promotional narratives to identify omissions, inconsistencies, and overlooked risks that may otherwise go unnoticed.
Back in February 2021, we published a Note under the headline “How to Read Drill Results & Numbers to Watch Out For”. Our analysts pointed out that, under the NI 43-101 Disclosure Rule, company news releases in which technical information is disclosed have to be:
- Balanced and not misleading
- Understandable to a reasonably informed investor
- Consistent in its use of standardised terms and definitions
- Based on reasonable assumptions which are clearly explained
- Unbiased and identify the potential risks and uncertainties
- Signed off by a professional (QP) who takes responsibility for the information
In practice, this is often not the case.
Companies in such press releases go to great lengths to present the information in the best light. It is understandable that the projects are promoted by highlighting the positives; management will not necessarily misrepresent the facts, but tend not to give the whole truth and nothing but the truth. For this reason, the reader should be sceptical and focused on what IS NOT being said. That article gave several examples of what to look for in drill results reports.
This Analyst’s Notes will expand on the reporting of drill results to the reporting of mineral resource estimations (“MRE’s”) and has taken the latest updated MRE of African Gold as an example. African Gold is an Australian-listed company and, as such, does not have the obligation to publish a full technical report supporting its MRE, unlike Canadian-listed companies. Not having such an obligation provides additional opportunities to avoid providing full disclosure. Crux Investor is of the opinion that the ASX authorities should look into this limited obligation as it is not to the advantage of the market.
Background to African Gold
The company's market capitalisation (ASX:A1G) on 11 July 2025 at a share price of A$0.215 was A$113 million; with diluted Enterprise Value at approximately the same level A$115 million (US$76 million). With the share price having had a remarkable rerating since mid 2024 (see below) all options are in the money. With the June 2025 cash injection from Montagne Gold (now with a 17.5% interest) and a private placement of A$2.7 million, the company has enough money to cover the annualised cash burn of the last nine months of A$1.5 million many times over. The low burn rate is partially explained by drilling expenses settled by issuing shares since November 2024.
The company focuses on the Didievi gold project in Ivory Coast. This 391 km2 property was acquired early in 2021. It is located in the Oume-Fetekro Birimian greenstone belt. The belt hosts Allied Gold’s Bonikro/Hire (+3 Moz) and Endeavour’s Agbaou (+1 Moz) gold mines to the south of Didievi and the Fetekro discovery (+2.5 Moz) announced by Endeavour Mining to the north. The project is therefore at a very good address. Below is the location of Didievi, which is shown relative to other mineral properties held by the company in 2024.
The company has since withdrawn from Mali, but after entering in October 2022 into an earn-in agreement with Reunion Mining Limited, the company owning the Syama mine, on their Syama shear zone project. Reunion can earn 80% of this project by spending US$0.5 million and commencing a feasibility study (“FS”) within five years and completing this within another three years. Transfer of the 80% ownership will follow a positive Decision to Mine, which must be made within 90 days following the completion of the FS. The annual reports of Reunion following 2022 are quiet on this project. Given the five-year period, Reunion is under no pressure to start spending the US$0.5 million.
In late 2020, Africa Gold announced it had entered into an option agreement to acquire 80% of the Kouroufaba gold project, which comprised land holdings consisting of the Didievi permit and three other permit applications totalling 1,534 km2 covering Birimian greenstone belts in Central Côte d’Ivoire (see the map above). Diviedi had previously been drilled by Equigold, Lihir Gold, and Newcrest, but none of them deemed it attractive enough to hang onto the project.
Below is the tenement map for Didievi in the 2024 Annual Report.
In 2021, initial drilling was carried out at the Blaffo Guetto target, which previous exploration work had proven to contain gold mineralisation. Drilling the area in the east underlain by granite proved to be barren of gold, allowing the company to drop 25% of the permit area upon renewal, as required.
According to the 2022 Annual Report, drilling during the year at Blaffo Guetto “delivered some spectacular broad, shallow and high-grade gold intercepts from surface”. A statement based on the results of one particular hole: DDD044.
During 2023, auger drilling was carried out along the Poku trend to identify drill targets. Within the trend, an anomaly with 900 m strike length grading ≥ 0.15 g/t Au was defined, with a peak value of 0.845 g/t Au. During that year, trenching tested the Kouassi target located 14 km NNE of Diviedi.
Away from Diviedi, at Konahiri Nord (refer again to Figure 3), composite track sampling during 2023 identified an anomaly with a +100 ppb Au on the permit's eastern border over a strike length of 18 km.
At Agboville, large outcropping pegmatite intrusions were identified with coincident lithium, caesium, tantalum and rubidium anomalism in stream sediment samples. Given the decline in lithium prices, it is understandable that this project seems to have lower priority now.
On 30 July 2024, a maiden MRE was announced for Blaffo Guetto with 0.45 Moz at a very attractive grade of 2.9 g/t Au, classified as Inferred. This MRE is probably the reason for the re-rating of African Gold’s share price.
During the March 2025 quarter, the company got a good endorsement from Montagne Gold, which intended to take a 19.9% strategic stake in the company by a share swap of A$6.7 million and A$0.9 million share placement. Africa Gold also announced a “large, high-potential gold system” at the Poku Trend, but the intersections highlighted only graded 0.5 g/t Au, which can only be of interest if amenable to heap leaching. The focus, however, remained on Blaffo Guetto with borehole results over wide intervals reported for four holes (with big gaps in borehole numbering).
Montagne Gold's strategic investment was completed on 13 June 2025, but the stake is 17.5%, not the 19.9% previously announced.
On 23 June 2025, an updated MRE was announced, with 0.90 Moz in Inferred resources at a grade of 2.5 g/t Au. The sections below will review this MRE and raise a number of concerns.
The Biaffo Guetto Prospect
Background
The gold mineralisation of the Blaffo Guetto prospect occurs as a set of closely spaced steeply dipping lodes controlled by the set of shear zones and their splays. 41 lodes have been interpreted on the cross-sections and delineated by constraining the defined intersections into 3D wireframes. This interpretation is based on the geological mapping of the prospect area and logging of the drillholes, which are distributed approximately on a 50 m x 40-30 m grid. Notably, the number of boreholes drilled pre-2021 was 203, and post-2021 is only 34. It raises the question of how the mineral resources could increase so dramatically from July 2024.
Below in Figure 4 is a plan view of Biaffo Guetto followed by a schematic cross-section showing the lode structures, some of which are very narrow and generally trend NE-SW. However, very confusingly, the legend of the plan view shows that the somewhat darker pink colour represents “mineral resource contours.” This does not make sense and is not repeated in the cross-section.
MRE Methodology
The methodology and results of the MRE are presented in a news release dated 23 June 2025 under the title “Blaffo Guetto’s Inferred Resource Surges 119% to 989,000 oz within 12.4 million tonnes at 2.5 g/t Au”
The gold lodes were initially interpreted on the cross-sections by defining the mineralised intervals where grade is not less than 0.2 g/t Au and thickness >2 m. These intercepts were correlated between cross-sections, presenting a set of continuous mineralised zones. “Correlation of the defined intersections between cross-sections was supported by geological interpretation of the host greenstone succession. Interpreted lodes were constrained by the 3D wireframes”.
An important observation is made as follows: “constraining the mineralised zones by the 3D wireframes has revealed the presence of the low-grade and waste parts within the mineralised lodes, indicating the patchy and discontinuous structure of the gold distribution within the host mineralised zone”. Crux Investor finds this a very concerning fact, putting a large question mark behind the validity of the wireframes.
The samples constrained by the wireframes have been composited to 1 m composites. The top-cut values were defined for each lode by finding a ragged tail on the Cumulative Frequency diagram and analysing the impact of high-grade cutting (capping) on the mean of the data population. A table shows the impact of grade cutting on the mean and variance (not standard deviation, as is the norm). Calculating the coefficients of variance (CVs = standard deviation/mean) from the numbers shows these to be generally low, which is good.
Each of the 41 lodes (wireframes) was estimated separately using the samples belonging to the estimated wireframe. Estimation was made into 20 m x 20 m x 1 m parent blocks that were infilled into the wireframes, constraining the lodes. The parent blocks were partitioned into 2 m x 2 m x 1 m, the size of the sub-cells reflecting the narrow nature of many lodes.
Mineralisation was constrained by wireframes, and the Au grade was estimated using the Inverse Distance of Power 3 (“ID3”) method. Part of the lodes were estimated using Ordinary or Simple Kriging.
The drillhole data and the block model was unfolded (flattened), using an equal thickness flattening algorithm of the Micromine 2024© (Crux Investor observation: such detail does not add anything to the reader’s insight as it refers to a black box methodology. The report should rather explain the principles of the methodology). Blocks and the drillhole data were flattened to 5 m equal thickness layers. Geostatistical data analysis and estimations were made in the unfolded space, which, after completion, was transferred back to the original (UTM) coordinates.
The search ellipse, used for the ID3 estimation, was defined in the unfolded space. Ellipsoid radii were 70 m (main axis, Azi 140o) x 40 m (semi-, Azi 230o) x 3 m (minor, vertical), which are consistent with the variogram ranges estimated for the lodes 177 and 178. Crux Investor records that no semi-variograms are presented in support of these chosen radii and to demonstrate the degree by which the curves are supported by data points.
As validation of the MRE, the estimated block grades were compared with corresponding drillhole sample grades. The method compares drillhole data and block grades by grouping them into large panels drawn along the strike of the lodes and, where appropriate, in the vertical direction. Crux Investor much prefers cuts through the block model that compare the block grades with borehole intersection grades. This will give the reader a much better impression of the reasonableness of block grades and extensions of such blocks to neighbouring borehole intercepts. There are numerous examples of MRE exercises extrapolating high-grade blocks to be immediately next to low-grade intersections.
The mineral statement - all in the Inferred category, but without a precise effective date - at a cut-off grade of 0.8 g/t Au is reproduced in Table 1 below.
Crux Investor Commentary to the MRE
The discussion of the MRE is, in the opinion of Crux Investor, deficient for the following reasons:
- The exact nature of the wireframing is nowhere properly explained and demonstrated by cross sections with borehole intersections;-
- No variograms are presented in support of the search radii;-
- No cuts through the block model are presented, comparing borehole intersection assay with block grades. This is, in the opinion of Crux Investor, the best method to verify the validity of the MRE;-
- Nowhere is it explained how the very attractive average MRE grade reconciles with the generally poor drill hole intersections reported in previous press releases (see the discussion below);-
- Nowhere is it explained how an additional 34 holes above the 203 holes available for the MRE one year earlier could result in a more than doubling of mineral resources. Less than 16% additional holes cause a doubling in resources?
The MRE in the Context of Previous Press Releases
What is striking about the discussion of the MRE is the absence of cross sections with boreholes intersecting the lodes and providing context to reported intercepts and how well these correlate. Crux Investor has reviewed all ASX announcements since the release of the previous MRE on 30 July 2024. Only four announcements were released with drill results, almost exclusively showing the results on plans and longitudinal sections. The 7 October 2024 press release is one of only two announcements that include a cross-section, but reporting on only one borehole (see below).
Note how poorly the lode interpretations are supported by the borehole assay values, with the lode to the east completely devoid of gold. A longitudinal section in the same ASX announcement shows four pierce points with results that are highlighted, surrounded by 17 other pierce points with results lower than 1 g/t Au.
In the 30 January 2025 ASX announcement, a longitudinal section is shown with a weird combination of intersections as bars and pierce points, which is repeated in the press release dated 31 March 2025 (see below).
Despite the highly promotional and speculative interpretations above, what is evident is the numerous low-grade pierce points between the impressive grades. However, with numerous parallel lode structures, this does not demonstrate that these impressive grades are in the same structure/lode.
The ASX announcement of 31 March 2025 promotes the high-grade trend further by providing the cross section below of the “high-grade terrace/tubular zone.”
The interpretation is based on two boreholes, with the company ignoring, for unexplained reasons, borehole 61 immediately north of borehole 60. There is no geological support for a “terrace” or “tube”, and no explanation is provided for what could control this. Crux Investor observes that the same cross-section profile would be achieved if the company intersected mineralisation controlled by a NW-SE cross-cutting structure. However, this would imply high grades over a limited strike extent along the main structural NE-SW control.
This kind of cross-cutting control is very much implied in the plan below, extracted from the latest corporate presentation. Crux Investor has annotated the slide with two blue lines connecting the holes highlighted as having outstanding results.
Finally, Crux Investor has attempted to find a press release with a comprehensive set of results, not just highlighted results. The JORC code itself prescribes under point 17:
Reporting of selected information such as isolated assays, isolated drill holes, assays of panned concentrates or supergene enriched soils or surface samples, without placing them in perspective is unacceptable.
And elaborates on this in its Table 1:
Where comprehensive reporting of all Exploration Results is not practicable, representative reporting of both low and high grades and/or widths should be practiced to avoid misleading reporting of Exploration Results.
The most recent ASX announcement with such information is dated 30 January 2025 under the headline: “First Assays from Transformational 10,000 m Drill Program Deliver Outstanding Results”. Please refer below to Table 2 and 3, which reproduce the results in this announcement and which Crux Investor has annotated by drawing blocks around reported wider intercepts. Please note that only ONE intersection has an excellent intersection. All others are either very narrow and/or very low-grade. Note that several intersections are shown as having 0.0 g/t Au! In any case, wide intersections should not be announced as the type of mineralisation is supposed to be high grade in narrow, sub-vertical lodes.
Conclusions and Recommendations
As for drill results, the announcement of mineral resources should not be accepted at face value, but critically reviewed to determine whether or not it is reasonable. It is up to the reporting company to make the case that the MRE is well supported by being consistent with the geological controls of the mineralisation, that the averaging method is suitable, the search parameters are well supported and the result verified in a manner that allows the reader to visually compare block grades and borehole intersection grades and to what extent block grades have been extrapolated beyond where there is support for these.
Some specific issues to consider when reading the explanation of the MRE methodology:
- Is the MRE consistent with exploration results previously announced?
- Does the wireframing reflect the geological control(s)?
- Has grade capping for precious metals been performed to an extent that drops the CV to preferably below 2.0? Companies are reluctant to cut grades aggressively as it will affect the average grade they will be able to report. However, averaging techniques such as ordinary kriging require a grade population that is not excessively skewed. A reluctance to cut grades should make the reader extra cautious. It is even more concerning when the company elects to use averaging methods such as multiple indicator kriging (“MIK”) that do not cut grades at all. History has shown that it leads to a clear overestimation of average grades. Examples are Brucejack, Sukari and Moose River.
- Looking at the various averaging techniques to estimate block grades, the reader should look beyond the technical jargon and realise these methods are all similar, just use a different approach to weighing the influence of assays on the average grade. The MRE discussion must make the case that the input parameters used for the weights are well supported. The results of the variography in the form of tables do not support this. Preferably, many visual examples of the variography should be given for the search directions.
- The input parameters for the reasonable prospect for eventual economic extraction (“RPEE”) should be well motivated. Low unit cost assumptions are typically provided for mining, processing, and general and administration costs. Often, the royalty obligation is ignored. The result cannot be considered credible when metallurgical recoveries are assumed based on little to no testwork.
- Finally, it is up to the MRE report to demonstrate a reasonable correlation between block grades and the assays of the samples used to estimate these. The more cuts through the block model together with borehole results to illustrate this, the better.
To read the FULL report, for FREE, please subscribe below.
- Notification By Email When Our Latest Notes Are Published With Immediate Access As Soon As They Go "Live"
- Suggest Future Companies To Be Analysed (Launching Soon)
- Additional Related Notes and "How To's" To Aid You On Your Investment Journey.
Already a subscriber? Sign in
Related companies
