The Admiral 2 cent green perforated 12 horizontally (#133) seems to be one of the more popular faked items in all of Canadian philately. George C. Marler, in his classic Admiral tome, describes examining a lot of 1500 copies and not finding a single legitimate copy. He further goes on to describe some aspects for authenticating this issue.
As is often the case with Marler, these descriptions are not always completely clear and require some reference to other parts of his book for further enlightenment. His three criteria for spotting the legitimate #133 include: (1) a break in the outer left frame line beside the second leaf from the bottom, (2) two "breaks" in the right side of the portrait oval at the level of the King's shoulder and 6 or 7 lines above it, (3) a small mark in the margin about 1mm right of the right numeral box. The book includes an image to illustrate this that does not show point 3 above. So it would appear that point 3 is optional in legitimising this stamp.
Since the #133 coils were all made by attaching strips of 20 from the post office sheets of #107, the specific type characteristics described above can be identified as coming from specific plates (165 to 172) of the regular sheet issue. Marler's description of this type does not include the margin dot (point 3 above) although he does indicate that it "appears on these plates". He also describes the second break in the portrait oval more specifically as being "seven" lines above the break at the shoulder. It should be noted: (1) that the left frame break is always a partial break that effects only the outer of the three engraved lines that make-up the frame (which appears as a single thick line on almost all printed copies), (2) in my experience, the "breaks" in the right oval are not consistent in that some (particularly the upper break) appear more as light areas rather than complete breaks in the oval line.
The relevant criteria mentioned are all illustrated in the image below. A couple of other recent
ebay auction images showing criteria 1 and 2 are also shown.



What is interesting is that Marler uses these criteria to address faked coils created from the imperforate issue or the booklet panes for the 2 cent green. Since these issues were not printed with plates with this type, these criteria are indeed an easy way to spot these sorts of fakes. Indeed, several current
ebay auctions clearly show fake #133 stamps that do not have these criteria.
The elephant in the room that everyone seems to ignore is that the sheet stamps can also be used to fake these coils in a very similar way to how the printers created the original coils in the first place. Trimming the left and right perforations from sheet stamps printed using the appropriate plates would create quite reasonable fakes as long as copies with wide margins were used. The stamps from the sides of the sheets with straight edges would seem to be particularly useful for this purpose. To my way of thinking, this makes the width of these coils another significant criteria for validating a legitimate copy.
Of course, as most of us will recognise, most coils created by the post office from sheet stamps in this era show a marked difference in their widths, so it may be difficult or impossible to determine a legitimate #133 from a cropped #107. Since the legitimate coils were created from the sheet stamps, and since perforation placement varies, there may be no way of actually differentiating modern fakes from legitimate coils created in this way.
This is the situation as I see it: the best examples of mint #133 coils should be sufficiently wide and centered enough to virtually preclude trimmed side perforations and they should also include Marler's third criteria (the margin dot) to provide a very high but still no absolute authenticity. It should be noted that many legitimate coils will not exhibit these characteristics. Finally, since the current Unitrade catalogue value for a VFNH #133 ($300) is only a little more than double the price of the #107 ($135), and since these differences become progressively less significant when retail values are concerned, the profit to be gained by faking mint versions of these stamps is not very motivating to fakers. Used copies are a completely different story. To my thinking, the gross difference in catalogue prices would lead to a multitude of fake used copies of this coil stamp and dated cancelations on the stamp or tied to a cover would be the only ideal (though still not perfect since these can also be faked) way of authenticating them.
The main take-away here is that the breaks in the left frame and right portrait oval are not enough to distinguish a good fake for this issue. Despite the long post, I am not an Admiral issue expert and I would encourage comment form others who are about this issue. Thanks if you read this far.