Go Back   CMP Forums > CMP Sales > CMP 1911
Register FAQ Members List Calendar Mark Forums Read

Reply
 
Thread Tools Display Modes
  #1  
Old 12-13-2018, 12:02 AM
Mercaptan Mercaptan is offline
 
Join Date: Dec 2018
Posts: 8
Default Projected Ship Dates (A statistical analysis)

Hey Folks!

Didn't get a number yet? Wondering when it might come?


Updated on 08JAN2019

Notice the break between 12/20 and 1/3. I have therefore broken the regression into two stages; 2018 and 2019. The 2019 projection assumes the same rate of number assignments we were seeing in November/December, an altogether reasonable assumption. Note that numbers likely to receive a 1911 (8000-9000) won't go out until April/May timeline!

Initial assessment: As you can see the regression for this is much tighter - we have more data and more lottery numbers assigned (versus those shipped). I did two regressions; one of the entire dataset (11/1 onwards) and another chunk after 11/10, where there was an inflection (rapid change of slope) that indicated an odd event. Because we are assuming they started at #1000, I have provided the estimate for #7500 (assumes 6500 pistols for sale). This means lottery numbers will continue going out into March!
  • CMP sent out at least 150 notices each on 11/5 and 12/4. Kudos guys and gals. We appreciate it!
  • Most work days are averaging around 75 notices sent out.


Already assigned a number and having fun waiting for your CMP 1911 like I am?


Updated as of 22JAN2019.

A long overdue update with regard to shipped pistols. There were no reports of pistols shipped between 12/27 and 1/9 that I could find. This delayed things slightly.

I've included two regressions. The solid black line is all the data, from 11/9 - 1/22. The dashed black line represents the peak ship volume (all pistols shipped between 12/1 and 12/20). I have extrapolated this ship rate with a center on January 16th using averaged ship data.

The "early" ship date is if CMP 1911 can keep up a December ship rate, which is unlikely. The "expected" ship date is using all data. My guess is... your pistol will ship about 3-9 days after the "early" ship date for the next several hundred orders.

Click on the image for a linear regression using the data I've sifted from this forum. Basically, we can approximately predict when your pistol will ship with this information, +/- a few days.

You'll note a couple of things about this data set:
  1. Few data points early-on
  2. Slow shipping rates prior to Thanksgiving
  3. More rapid fulfillment post-Thanksgiving

Because of this I broke the regression into two parts, the solid line (all data) and the dashed line (data after 12/1 only). This results in two projected ship dates in the bottom right corner; the expected (solid line) and early guestimate (dashed line).

Given time off with the upcoming holidays, I suspect the solid line equation will be more accurate in the near-future (through Q1 of 2019). If there are only 6500 pistols, I project somewhere between July to August for the last shipment.

Notes and Other:
I data-mined the following threads to create this information:


Data acceptance criteria:
  • Must have a full and complete lottery number (none of this 12XX business)
  • Must have a valid ship date (not receive date, not phone call, etc.)

Questions:
Hey, my pistol didn't ship based on what you said!
- Not really a question, more of a statement. Statistics are funny this way. Nothing is absolute, but this is a pretty good guess. No guarantees are made with this model. Please don't contact the beleaguered CMP and tell them they're behind schedule and use this as 'evidence.'

How can I help?
- Great! Please fill out your information with a lottery number and ship date (if applicable) in the stickied 1911 order thread. I'll update this model as more pistols ship, because more data = better predictions.

What's my exact ship date?
- The equations are in the figure above. If you want to go with the solid line estimate, perform the following calculation:

Quote:
Date = (Order Number+839501)/19.361
That date is a serial date (# days past 01 January, 1900), and you will need to convert it into a more reasonable format.

If you have excel, take this serial date and enter into a cell
Quote:
=TEXT(123456,"mm/dd/yyyy")
where 123456 is your serial date.

Last edited by Mercaptan; 01-23-2019 at 12:33 AM. Reason: Updated on 22JAN2019
Reply With Quote
  #2  
Old 12-13-2018, 12:13 AM
LHS905 LHS905 is online now
 
Join Date: Oct 2018
Location: Colorado
Posts: 29
Default

Nice!
Reply With Quote
  #3  
Old 12-13-2018, 12:27 AM
Quicksilvergoat Quicksilvergoat is online now
 
Join Date: Aug 2011
Location: DFW Metroplex
Posts: 572
Default

I’d like to see your analysis of when everyone will have their number.
Reply With Quote
  #4  
Old 12-13-2018, 12:33 AM
Todd Todd is offline
 
Join Date: Dec 2009
Location: Talent OR
Posts: 91
Default

Mercaptan, can we call you the mathman, j/k. Great work
Reply With Quote
  #5  
Old 12-13-2018, 01:49 AM
Mercaptan Mercaptan is offline
 
Join Date: Dec 2018
Posts: 8
Default

Quote:
Originally Posted by Quicksilvergoat View Post
Id like to see your analysis of when everyone will have their number.
Good call. I added a regression for this in the original post.
Reply With Quote
  #6  
Old 12-13-2018, 03:25 AM
Ohio Don Ohio Don is offline
 
Join Date: Jan 2014
Location: S Florida and SE Ohio
Posts: 4,147
Default

Now why would someone want a name that is known for a rotten egg smell in natural gas piping?

Interesting data analysis.
__________________
NRA Life member
NRA Certified Pistol Instructor,RSO
US Army Transportation and Medical Corps Museums Life member
Reply With Quote
  #7  
Old 12-13-2018, 05:52 AM
siegfried siegfried is offline
 
Join Date: Oct 2011
Posts: 28
Default

Looks like an excellent fit on the trend line, thanks for the excellent analysis Mercaptan. I am #1833 and it looks like my number is almost up in the non-mortality sense.
Reply With Quote
  #8  
Old 12-13-2018, 06:57 AM
fiftycal2004 fiftycal2004 is offline
 
Join Date: Mar 2016
Location: Midland, MI
Posts: 19
Default

This is excellent! I know some Ph.D.s that can’t get R squard values like that. CMP1911 is definitely staying on top of it!
Reply With Quote
  #9  
Old 12-13-2018, 08:56 AM
Quicksilvergoat Quicksilvergoat is online now
 
Join Date: Aug 2011
Location: DFW Metroplex
Posts: 572
Default

Quote:
Originally Posted by Mercaptan View Post
Good call. I added a regression for this in the original post.
Thanks
Reply With Quote
  #10  
Old 12-13-2018, 10:03 AM
cranehunter cranehunter is online now
 
Join Date: Oct 2011
Location: Monroe, WI
Posts: 862
Default

A lot of nice work there, but I am going to stick with my magic 8 ball. Just as accurate for something like this where can be changes made to the process at any time.
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -5. The time now is 09:33 PM.