Latest from WillP on Stockwatch. I've repeated what Buffalo Bill previously posted. I suggest we all chip in and pay for WillP's SI subscription. We need more rational people here!
stockwatch.com
Author: WillP -- Date:1999-02-08 08:50:10
Subject: Diamond Analysis
OK, all...chew on this for awhile.
We know the sizes of the four largest stones: Just under 11 carats,
8.4 carats, 6 carats and 5 carats. There are 21 other stones larger
than 1.0 carats...weighing 37.44 carats. A total of 1387 diamonds
were recovered that didn't fall through a 1x9 millimetre screen.
Recovery of stones below 2 millimetres seemed to be very sloppy.
So...it's fairly easy to construct a diamond distribution curve...and
here is one. The weight categories were selected to match the raw
diamond value chart I sent earlier. The value is the average value of
a gem of equivalent quality to the three largest diamonds, based on
the value chart.
The four largest stones sizes are known...they don't appear in the
table. The purpose of this exercise is to determine the valuation of
the remaining stones.
This isn't rocket science.
Number .. Size Range . Avg.Wt. Value Total Value
... 2 ..... 3.00-3.99 .. 3.30 ... 1552 .. $10,243
... 4 ..... 2.00-2.99 .. 2.30 ... 1328 ... 12,218
... 6 ..... 1.50-1.99 .. 1.65 .... 930 .... 9,207
... 9 ..... 1.00-1.49 .. 1.15 .... 830 .... 8,591
... 4 ..... 0.90-0.99 .. 0.93 .... 556 .... 2,068
.. 13 ..... 0.68-0.89 .. 0.75 .... 423 .... 4,124
.. 29 ..... 0.46-0.67 .. 0.53 .... 340 .... 5,226
.. 21 ..... 0.38-0.45 .. 0.41 .... 249 .... 2,144
.. 66 ..... 0.25-0.37 .. 0.29 .... 183 .... 3,503
.. 82 ..... 0.18-0.24 .. 0.20 .... 141 .... 2,312
. 257 ..... 0.10-0.17 .. 0.12 .... 108 .... 3,331
. 652 ..... 0.05-0.09 .. 0.06 ..... 83 .... 3,247
. 238 ..... 0.02-0.04 .. 0.03 ..... 43 ...... 307
The total value of all of these stones...assuming they were all of the
exact same quality as the three largest...would be $66,521. Now,
we know that the value of all of these stones, totaling about 196
carats, was approximately $16,750, or one quarter the value of the
total parcel.
So...if you deal with just one value class...average, each diamond
was worth $85 US per carat. However...if you use two quality
classes...crap and equivalent to the top three stones...you easily
compute that fully 25% of the stones had to be in this category.
That is, 49 carats were so classified. Now if you add in the 25
carats from the top three stones...74 carats out of 226 or just
under 33% would account for all of the value.
Now some of you are going to whine...but there are 5000 different
classes of diamond valuation. So what...mathematically? If the
sample is large enough...you get the same result. I can simulate as
large a bulk sample as you want. It yields the same result
mathematically. Go ahead, try it with three classes...with five...with
seventy-four.
So now, as I have alluded...it is an easy exercise to generate a bulk
sample...either with statistics or by iteration on a computer.
Using the lowest, most conservative values throughout...diamond
distribution curve based on 1.14 carats per tonne, gem quality at
25% not 33%...I consistently generate a value per carat of
between $145-$175 US with the median at $160 for a 6,000
tonne sample. The median largest gem contained in these parcels
was 31.7 carats.
Now remember...these numbers were generated with many
conservative assumptions: #1. There is a real and high probability
that the average grade will be 20-50% higher than 1.14 carats per
tonne. #2. In assessing gem quality...the entire contribution of the
three largest stones was ignored. That's not realistic. #3. I
artificially deflated the distribution curve at larger sizes. This is my
own personal safety hedge.
A simple way to consider it all is...there were 4 large stones...three
of gem quality. These 3 contributed on average 25% each to the
value of the parcel. There should have been only one...statistically.
So the value per carat should be halved...to $150 per carat. This is
a good conservative number to play with...not the 85 US per carat
that totally ignores the fact that the stones existed.
And a final thought...remembering all the theory of diamond
formation...transportation...and destruction...
Do you really wish to base your investment decisions on the
principle of ignoring something that was duly found and assessed
by at least 4 independant diamentaires?
I think 3 of four stones over 4 carats being fine gems was a bit of a
fluke. But quite possibly it's not. I don't know enough about
diamonds.
But the more astute of you out there will realize...I know a little bit!
What I do know alot about is life...and one lesson I've learned is it
never pays to ignore reality...and the fact that three stones weighing
25 carats were worth over $50,000 US...is reality.
:-)
Cheers!
Top
Reply
Author: WillP -- Date:1999-02-08 16:30:13
Subject: Lies, Damned Lies, and
Statistics.
So...more chewing. It's almost supper time here.
There were three large gems in Winspear's 200 tonne bulk
sample. Therefore the best guess of the overall grade of the
ore is three especially fine gemstones per 200 tonnes.
But what is it really? Who knows!
However...let's have some fun with chance and probability!
Given an actual grade of three fine gemstones per 200
tonnes...what are the odds of finding exactly three such
diamonds in 200 tonnes? Well, there's about a 22% chance.
With an actual grade of five large fine gemstones per 200
tonnes? You would find only three 14% of the time.
What about the other way?
With a real grade of two such stones per 200 tonnes, you
would find three 17% of the time. Lower it to one stone and
the chance of finding three in 200 tonnes lowers to 6%.
At one large fine gemstone in 400 tonnes...you would
expect to find three in a 200 tonne sample just under 2% of
the time.
Try not to misuse this data. For instance, if the actual grade
was 7 stones in 200 tonnes, the chances of finding only 3 is
5%. This does NOT mean that the probability that the actual
grade is 7 large gems in 200 tonnes is 5%.
I may address this question from the standard error
approach in my next effort. But I digress.
The probability that it was an absolute total fluke that there
were large gems worth $2000 per carat in a 200 tonne
sample? Well, if the actual grade is one in 1000 tonnes...or
0.2 in 200 tonnes, the chance of finding exactly three is
0.02%. That is...two in ten thousand.
Of course...even at that paltry rate...one stone in a thousand
tonnes...the contribution to the ore value is still just under
$17 US per tonne in addition to the 97 from the remainder
of the parcel.
Was it a fluke? Yes. Most probably.
To what extent? In what direction?
Had Shakespeare lived in our age he may have
penned..."To long or to short, that is the question."
You be the judge.
Ya rolls the dice and ya takes yer chances.
Cheers!
WillP
Top
Reply
Author: WillP -- Date:1999-02-08 22:45:50
Subject: Unique?
Is Snap Lake unique? Unprecedented in the annals of
diamond exploration?
Sure it is...just like no two snowflakes are allegedly the
same.
But seriously...no it's not. Not from a diamond grade or
evaluation standpoint. Unusual? Yes. Unique...no.
Earlier I posted a combination of real and projected
diamond distributions for Snap Lake. Now I'm going to
compare it with real data from a Southernera release from
South Africa. The SUF parcel consisted of 263 carats, the
WSP parcel 226 carats. The bracketed number is SUF's
numbers adjusted down to 226 carats.
Size Range ... WSP ..... SUF ... ADJ
10 ct+ ......... 1 ....... 1 ... (1)
8 to 10 ct ..... 1 ....... 1 ... (1)
4 to 8 ct ...... 2 ....... 1 ... (1)
3 to 4 ct ...... 2 ....... 2 ... (2)
2 to 3 ct ...... 4 ....... 4 ... (3)
1 to 2 ct ..... 15 ...... 32 .. (27)
0.5 to 0.99 ct. 46 ...... 67 .. (57)
Anything seem terribly out of whack there? Not to me. With
a 1.8mm cutoff, SUF had an average stone size of 0.33
carats. WSP has one of 0.16 with a 1.2 mm cutoff. If you
remove the 238 stones smaller than SUF's cutoff (and their
7 carats) you get an average stone size of 0.2 carats.
Contrast these values with Aber's A154-S which had an
average stone size of 0.1 carats.
How's this for a comment picked from the SUF
release...."The diamonds are almost entirely of gem quality."
Hmmm. Interesting.
Yes, but valuations such as Winspear's are unprecedented,
no?
No.
Somewhat old data from A. J. Jansy gives the following
producing mines with a per carat value of $200 or more:
Camutue. Jagrsfontein. Letseng. Majhgawan. Kimberley.
Koidu.
One of those...had a per carat value of $400.
As far as values per tonne go, remember the 23rd
Congress? 720 per tonne. International? 480 per tonne.
Now there were a few things bothering me about rejecting
Winspear's three largest gemstones without question. Those
things are:
#1... It is not statistically accurate to reject data without
valid cause.
#2... The mere fact that they existed and existed in two
separate areas suggests that no valid cause exists to
*entirely* reject them.
#3... Of most importance is the suggestion that inferior large
diamonds are far more likely to be destroyed than are large
fine gems. Why? Don't know. Could hazard a few guesses
though. What makes them bort? Think flaws. Think
inclusions. Think things causing weakness.
So...in the spirit of accuracy and prudence, kindly allow me
to offer my best guess.
I feel the 6000 tonne bulk sample will return a diamond
value of $160-$200 US per tonne. Of this I have an 85%
level of confidence at the lower level. Downside risk is
present, but relatively small. My 95% confidence mark
would be around $120 per tonne. Upside potential is great.
Consider what I have chosen to ignore:
#1. GRADE. Two pits gave values of 0.86 and 1.47 carats
per tonne. The first pit exhibited markedly lower diamond
content than did pit two, which seemed closer to all the
other holes drilled around Snap Lake. (If you feel that there
is no corelation between macro diamond counts and richer
portions of kimberlite...review all of the 100 odd CF results
released over the years. Compare holes that hit larger stones
(0.10 ct and up) with holes that didn't. There's a strong
correlation...even in a remarkably consistent pipe like
A154-South.)
#2. THREE LARGE STONES. It's never prudent to ignore
reality. One can attempt to discount it...but its presence is
always there haunting you. It's highly unusual to discount all
three of them. The most valuable...yes. Maybe even two?
But all three. Why didn't someone point to the 5 carat stone
and scream..."You've taken my three best and left me with
this...bort!!! I want you to at least make this a fine gem in
return!" It would only be fair, no?
#3. LARGE STONES IN GENERAL. As I mentioned
earlier, I significantly deflated the ability of large stones to
appear in my model. (Larger stones are those greater than
10 carats.) Nature may not have operated under a similar
constraint.
#4. CAUSTIC FUSION RESULTS. Buried in the release
was mention that the NW dyke results had three stones
weighing between 0.47 and 0.75 carats, in 1200 kg of ore.
The bulk sample contained approximately 0.4 such stones in
a representative (average) 1200 kilogram sample.
Statistically relevant? No. Worthy of being ignored? No. All
it is is an indication of at least 1.2 tonnes of ore exceeding
expectations of the mini-bulk sample.
The risks? #1. TONNAGE. The project needs 10 million
tonnes, especially at my lower valuations to be highly
profitable. If one examines the available drill maps of Snap
Lake...count the squares. Each one is roughly one million
tonnes. There are ten contiguous squares with kimberlite
intersections, and no misses anywhere within the area. There
are more squares with hits but tyey are not yet "attached" to
the contiguous area. Assessed risk: Low. Risk at 20 million
tonnes: High. Risk at 50 million tonnes: Very high.
#2. GRADE. Addressed above. There's a chance that the
Pit 1 grade of 0.86 carats per tonne is more representative.
Seemingly refuted by CF results. Assessed risk: Low.
#3. VALUE. Addressed above. Much hinges on gems of
good quality being present. Risk at $120 US per tonne:
Low. Risk at $180 US per tonne: Moderate. Risk at $300
US per tonne: High. Risk at $420 per tonne: Extreme.
#4. PERMITTING. Nothing need be dammed, dyked, or
drained. The 3-D's are bad words in the NWT. With Ekati
and Diavik gone before....assessed risk: Very low.
#5. CAPITAL COST. The MRDI scoping study suggested
a capital outlay of $105 million. Add to that at least $50
million for current exploration, feasibility, and permitting. The
cash has to come from somewhere. Debt is good. Equity is
not. Risk of significant dilution: High.
#6. TAKEOVER. If Snap Lake is indeed a bonanza, and is
poorly valued by the market...look out. Risk: High.
Well..it's late. I hope I have presented this in an objective
fashion.
There's so much to consider, eh? :-)
All the best. |
