9:07 pm

April 15, 2005

Kirk,

I recommend that you obtain a copy of John Madl's book on the Model 1886. The survey completed by Bill Porter has errors in it. John with the assistance of John Hawk (a former assistant in the CFM research office) conducted an extensive survey and the production numbers they recorded are more refined than Porter's numbers.

Bert

WACA 6571L, Historian & Board of Director Member

12:33 am

February 5, 2016

Bert,

You always have something good to add. I just ordered the book.

Have another question: Is there a formula/matrix where I could enter the characteristics of a rifle to find out how many guns of that specific type were made?

Mathematically I think I know how to do it based on % probabilities but would like some thoughts.

Thanks

Kirk

6:58 am

April 15, 2005

Kirk Fitzgerald saidHave another question: Is there a formula/matrix where I could enter the characteristics of a rifle to find out how many guns of that specific type were made?

Mathematically I think I know how to do it based on % probabilities but would like some thoughts.

Thanks

Kirk

Kirk,

In answer to your question, No, there is not. In order for something like to work, it would require a very detailed listing of the entire production run for the subject model. Additionally, the more items that you add to the matrix, the more complex the formula would need to be to accurately calculate (extrapolate) the potential production number. In my surveys, I limit the total number of variable per formula to just two. Beyond that, it gets very complicated.

Bert

WACA 6571L, Historian & Board of Director Member

6:37 pm

February 5, 2016

Burt,

I have spent the last several days wrestling with trying to get a answer to how uncommon is my rifle. I understand your point about the level of confidence being reduced after the use of two variables in production number percentages. Bottom line I want to know if my rifle is possibly a one-of-a-kind.

Here is the whole story. The Factory Letter states:

Rifle

45/70

1/2 octagon barrel

Plain trigger

Plain pistol grip

Lyman front and rear sights

Shotgun butt

Rubber butt plate

2/3 magazine

By any standard the gun is 90%+ condition

From the Porter survey in can get most of the production data. But factory data does not address, at least consistently, factors like plain pistol grip, shotgun butt, rubber butt plate and Lyman sights. I would guess that surveys of existing guns would provide some insight, not absolute data as to potential numbers. I know that I will not get a mathematically derived answer to how rare is the gun. However with any additional data based on surveys of existing guns I may arrive at an intrusive probability range of rarity.

Help/data from anyone would be appreciated.

Cheers Kirk

10:00 pm

April 15, 2005

Kirk,

I am very confident that your Model 1886 riffle is not a "one-of-a-kind", or even close to it. Of the features you listed, none of them are rare, or even uncommon.

1. 1/2 octagon barrel... many thousands were made

2. Plain pistol grip... at least a few thousand were made

3. Shotgun butt... tens of thousands were made

4. Hard rubber butt plate... tens of thousands were made

5. 2/3 magazine... at least a few thousand were made

6. Lyman front & rear sights... many thousands were made.

The odds are very high that at least a few hundred rifles were made just like yours.

Bert

WACA 6571L, Historian & Board of Director Member

10:13 pm

February 5, 2016

Lets deal with real numbers.

Production percentage numbers for the Model 1886 are as follows:

2/3 Magazine = 2.0% for a total of 240 rifles with this feature.

1/2 Octagon Barrel= 2.2%

Pistol grip=4.5%

These are enough to make my point. I need not use the other variables i.e. caliber, ETC.

Starting with a known production number of 240, and assuming normal distribution of features the following is offered:

240 X 2.2% octagon barrel=5.28 (possible productions guns) X 4.5% pistol grip=0.2 (possible production guns).

It can be seen that, without going into all the other known variables and observed or anecdotal variables, i.e. rubber butt plates were less used that metal , ETC the answer is that there is a high probability that another such configuration does not exist.

For this discussion I don't care if there is better survey data than Porter,s out there where we an argue tenths of a percent and production numbers. This data is good enough for government work.

I believe this makes a convincing case for a one-of-a-kind gun. Thoughts all?

Additionally if someone has a same configuration gun let me know.

Cheers Kirk

12:16 am

November 5, 2014

Hi Kirk-

I do not presume to be a math professor (OTOH apparently maybe I do... But I'm not ). I am sure the specific combination of features present in your rifle are rarely encountered together...

The thing is that your statistical argument is based on the assumption that each variable sorts independently. For example, that guns ordered with half octagon barrels were no more (or less) likely to be ordered with half magazines than rifles with full octagon barrels. I suspect that certain features 'tended' to be ordered together, which if true would invalidate the pure statistical argument based on 'random assortment'.

A curious example from back in the 1980s was when President George Bush (1st), his wife Barbara, and their dog Millie all came down with the same thyroid condition (Grave's disease) within the space of a few years. There was great concern within the White House that this was some kind covert of Soviet plot. It turns out that (according to the CDC that was called in to investigate) the prevalence of Graves disease in the population (including dogs!!!) was high enough that the 'odds' of three people in the same household coming down with the condition within a few years of one another was only about 1:20,000. VERY unlikely, but not impossible... Such are the powers and pitfalls of statistics!!!

Any way you 'sort' it, however, it's a nice rifle in a rare configuration and POSSIBLY one of a kind.

Best, Lou

WACA 9519; Studying Pre-64 Model 70 Winchesters

3:44 am

November 5, 2014

Hi Kirk-

This is off-topic, but... In his book Roger Rule, got most everything right (IMHO), so there's not much to gain from studying these rifles beyond what must be gained by 'hands-on' experience (that books cannot teach).

Most of my ongoing 'surveys' are so obscure that I do not ask for help here. I do the 'homework' myself from the internet and fellow aficionados I know (both dealers and collectors), who understand that I am not trying to find rare guns for the purpose of trying to acquire them myself (something of which 'surveyors' of rare items are sometimes accused).

So some things I'm interested in at present are:

1) Ramped straight taper barrel M70s. Both .375 MAGNUM and every other chambering, i.e. '1st variation target models' (the rifles of which Rule said there were only 40 or so known). So far I've found some info on 23 ramped straight taper barrel M70s that aren't 375 MAGNUM and I think there were a good number more than 40 made...

2) Front sight heights on rifles with Lyman 31W (pre-war standard rifles) and/or Redfield "full gold bead" SGs. This is a curious one to me. The catalogs I've seen to date do not specify height. Rule says the "Redfield 255" (0.360" height) is correct on SGs (which is why Redfield 255 sights sell instantly on Ebay for big $ while the 254/256 sit forever). But others have reported different height 31W and Redfield gold bead sights on old 'beater' M70s, raising questions. Madis' sight book claims different height 31W front sights were used based on M70 chambering, which kind of makes sense since it was common practice in earlier times on other models, but the 'books' do not validate this. If anyone has info, I'd like to know!!!

3) Position of the front sling swivel of M70 SGs in relation to the fore end tip. On the earliest guns is was about 1/4", but at some point it got moved back to about 3/4", presumably b/c the wood kept splitting out. See pics below of a couple of mine [SN 3768 (1937) versus SN 50821 (1942)]. I am trying to narrow the SN range of the changeover.

The challenge with M70s is that I must assume that every one I see has been 'helped' somewhere along the way, so the only chance to get info is to look at 'population averages', which will eventually reveal which are the 'outliers' (unless all M70s have been 'helped' randomly). This is something all the people here who do Winchester surveys know!!!

Bloody statistics!!!

Best, Lou

WACA 9519; Studying Pre-64 Model 70 Winchesters

Most Users Ever Online: 628

Currently Online: Big Mac

55 Guest(s)

Currently Browsing this Page:

1 Guest(s)

Forum Stats:

Groups: 1

Forums: 16

Topics: 6883

Posts: 55781

Member Stats:

Guest Posters: 932

Members: 8950

Moderators: 5

Admins: 3

Newest Members:

barrelsandblades@gmail.com, bholmes@acadian.com, daduke357@yahoo.com, trozzo51@yahoo.com, willis1952@yahoo.com, Weaverfarms@skybest.com, Danshaffer1@gmail.com, hscauctionsmaine@gmail.com, john.janossors@gmail.com, Mtaylor@apogeenow.comAdministrators: admin: 4, Mike Hager: 169, Bert H.: 9632