Grading Stamp Book Leaflets

The Exceptable Method by

R Gordon Schmidt

The United States Specialist
March 1976

United States Booklet Pane Scarcity
Originally, Stamp Books and Leaflets
Unique to the Flat Plate Printing Era
1900 to 1923 also 1918 AEF and 1928 Lindy

Written by R Gordon Schmidt, BIA 7962

Interpolated by John P McGowan, BIA 7787
Web Site: www.epopstamps.com
email: usleaflets@epopstamps.com

PURPOSE OF THIS ANALYSIS
The first work in identifying Stamp Book Leaflet positions was that conducted by George H. Beans (See Everybody's Philatelist, George H. Beans) and published in EVERYBODY'S PHILATELIST. Stampers should be indebted to Mr Beans because of his position nomenclature and early promotion of stamp book leaflet collecting.

Since they first appeared April 16, 1900 collectors have searched for a method of evaluating the scarcity of one position or another, mainly to determine the value of their holdings. Stampers and investors today realize the true potential value in leaflets as evidenced by the rapidly increasing stamp catalog prices. Mr Beans also was impressed with the monetary aspects of flat plate leaflet collecting. He is said to have found stamp book leaflets enjoyable and even a profitable line.

In a later attempt to arrive at a true valuation of flat plate leaflets by position, J. C. Harrigan attempted to match position scarcity, condition, and issue. (Harrrigan, 1930). W. A. McIntire published most of Harrigan's original work, clarifying points earned, and including a discussion of more recent flat plate issues. (McIntire, 1940.)

Harrigan and McIntire based position scarcity on the theoretical probability of a particular position occurring. They arrived at the relative ranking as shown in TABLE I, below. The scarcity value for a given position was determined through a very superficial investigation of probabilities. This article delves more deeply into the probability of a position occurring and gives, in the authors opinion, a truer picture of the scarcity of flat plate position leaflets. Also, this work includes positions from the 180-subject Flat Plate which were not considered by Harrigan and McIntire.

ASSUMPTIONS
There are two assumptions used in this work which should be mentioned and they are not so far from the real world as to be unacceptable to this analysis.

Assumption I.
Harrigan and McIntire presented the point that to be a position leaflet full guide line segments had to be exhibited on the leaflet.

This article agrees that a full guide line segment is desireable but defines a position leaflet as any leaflet that can be located on the plate from which it was produced.

Example: Positions B and C in Harrigan and McIntire's view cannot occur simultaneously because one must be spoiled if the other exists. This article assumes that both occur on every plate thus diluting the value of positions B and C.

Harrigan and McIntire would arrive at exactly one-half the number of existing leaflets that this article has, and therefore twice the value for scarcity.

TABLE III is used as a basis of comparison for all issues of the 360-subject plate. The estimate of the total number of leaflets produced for each position is given in TABLE IV. These include those issues exclusively from the 360-subject plate. The total quantity of leaflets issued is from Harrigan's work and Bush's data.

Assumption II.
Harrigan made a survey of the occurrence of positions and found that, ". . . while they (Bureau Employees) theoretically aim to cut on the guidelines (center of gutters), they actually cut to the right of it on the average . . . ." Therefore, more H than I positions are found, more K than L, and more N than O leaflets.

This work assumes that 60% of the time the knife cut to the right of the guideline and 40% of the time it cut to the left of the guideline. As the sheet was turned for its second cutting the knife cut above the guidline, which was to the right, 60% of the time and cut below the guideline, which was to the left, 40% of the time.

With these assumptions in mind let's take a look at the 360-subject plate.

The 360-Subject Plate
Sixty leaflets are produced from the 360-subject plate. That means that each leaflet number (1 thru 60) has the probability of 1/60 or 0.01667 of occurring. It is not difficult at all to determine the total probability of a position occurring if it occurs on every sheet every time. ie Position D: it occurs on every sheet every time and hence has a probability of 1/60 or 0.01667. By Assumption I, positions B and C also occur on every sheet every time and therefore position B, position C, and position D have the same probability of 1/60 or 0.01667.

However, no other position can be treated as easily. What about position A? It occurs in at least 31 places on every sheet but it also could occur in combination with other positions up to an additional three leaflets.

If an M leaflet is produced then the one above it becomes an A leaflet, so it has a larger probability than a simple 31/60 or 0.51667. The author believes the best way to describe the involved probabilities is by referring to the position numbers, as adopted from Mr Beans pictorial of the 360-subject sheet (in reverse to the plate), and discussing each of the three complexes which occur.

The H-I Complex
If an H leaflet is produced, then an A leaflet is also produced at positions 17, 46, and 56. Conversely, if an I leaflet is produced, then A leaflets are created at positions 15, 45, and 55. By Assumption II, 60% of the time these leaflets would be position H and 40% of the time would be position I. The H-I complex occupies a total of six leaflets for a total probability of 6/60 or 0.1000.

Regardless which way the sheet is cut, three leaflets will be position A. Therefore, one-half the total probability of the H-I complex is associated with position A or 0.0500.

According to Assumption II, the other half of the probability is split between position H and position I in a 60%-40% divison so that the probability of a position H occurring is 60% of 0.0500 or 0.0300 and position I is 40% of 0.0500 or 0.0200.

The J-M Complex
The same arrangement used in the H-I complex may be extended to the J-M complex. If a position M leaflet is produced, then the corresponding position J becomes a position A leaflet. Conversly, if a position J is produced, then the adjoining leaflet becomes a position A leaflet.

According to Assumption II, the other half of the probability is split between position J and position M in a 60%-40% divison so that the probability of a position J occurring is 60% of 0.05 or 0.03 and position M is 40% of 0.05 or 0.02.

The total probability for the J-M complex is 16/60 or 0.2667 which are represented by positions 21, 22, 23, 24, and 27, 28, 29, 30, 31, 32, 33, 34, and 37, 38, 39, 40. One-half this possibility 0.2667 or 0.1333 is given to position A. Sixty percent of the remaining half 0.1333 is 0.07998 and 0.08 is assigned to position J. Forty percent of the remaining half 0.1333 is 0.05334 and 0.05 is given to position M

The K-L-N-O Complex
Now comes the hardest of the three complexes to evaluate. The K-L-N-0 complex occupies four leaflets for a total probability of 4/60 or 0.0667. The following list shows what happens as each position leaflet is produced

Based on Assumption II
the joint probabilities are represented by Figure I.

The joint probabilities are calculated by multiplying the percentages together, that is, 36% for position K comes from the 60% chance of the knife creating a line on the left and the 60% chance of the knife creating a line on the bottom, that is, 60% x 60% = 36%. The probability occupied by position K leaflet is 36% x 0.01667= 0.006. The remaining probability is distributed so that 64% of the time J, H, or A leaflets are produced at position 25. Similar logic is applied to position L, position N, and position O with resulting probabalities of 0.004, 0.004, or 0.00267 respectively.

The leaflets produced at the K-L-N-O positions include four position A. two position H or two position I, and two position J or two position M. Their probabilities compute to 0.01667, 0.01, 0.00667, 0.01, or 0.00667 respectively. Now that all 60 positions have been accounted for, Table II of total occurrence probability can be made.

TABLE III illustrates percentages and scarcity factors associated with each position analyzed.

The probability expressed as a percentage indicates the percent of the leaflets produced which may be expected to have occurred for any issue.

Example:
Seventy-one and seven thenths percent of the leaflets produced for any issue printed from the 360-subject plate can be expected to be position A, whereas only four tenths percent of the leaflets produced can be expected to be position L

The scarcity factor is simply the ratio between the probability of position A and any other position.

This means 268 times as many position A leaflets were produced than position O leaflets.

TABLE III is used as a basis of comparison for all issues of the 360-subject plate. The estimate of the total number of leaflets produced for each position is given in TABLE IV. These include those issues exclusively from the 360-subject plate. The total quantity of leaflets issued is from HarriganÕs work and BushÕs data

TABLE IV
QUANTITIES ISSUED BY POSITION FOR THE 360-SUBJECT PLATE.

So concludes this discussion of the 360-Subject Plate. A discussion of Scott number 279e, 331a, and 332a occurs later in the article, as they were produced from both the 360-Subject Plate and the 180-Subject Plate.

(To be continued)

(end of page)