So... we already realise 1st execution Polarouter numbers are very low... Its also fairly clear that the Subreferences were produced sequentially in batches, i.e. 20217-1, then 20214-1, then 20217-2, 20217-3, 20217-4, and so on. The full known ranges can be found in the Reference Tables: https://universalgenevepolerouter.com/reference-tables/reference-table-steel-gold-plated/ This shows there were probably somewhere between 58 and 253 of each Subreference made. But if we apply a few assumptions, we can probably make a better estimation than this. The assumptions would be: 1) The subreferences were ordered/produced in equal number, i.e. same number of 20217-4 were made as 20214-6, etc.; and 2) There are no gaps in the serials between the Subreferences; e.g. if the last 20217-1 serial was 1647674, then the first 20214-1 serial was 1647675. Dropping this data into a quick spreadsheet, we get what i've screenshotted below. By editing only the two cells outlined in orange, we can test which models with the above assumptions fits the data. The result I get, is that it fits our data for any whole number between 167 and 174 (i.e. between 167 and 174 of each subreference). Id argue to add one more assumption - that UG would order in a nice round number. So, in summary, if all of that holds, that would mean that 170 of each 1st Execution Polarouter subreference was produced. More importantly, that means there are a lot left for us to find!
Can you explain where you got all this data? Is it in some other string somewhere? I do not see it on your website, but maybe I am looking in the wrong place. What is the difference between "Serial Min" and "Modelled Min", etc?
If you happen to have samples of serial numbers for each of your subreferences you can apply some statistics to estimate total production numbers using the MVUE equation: Population Maximum = Sample Maximum + (Sample Maximum / Sample Size) – 1 Read more about the history behind the theory here: https://en.wikipedia.org/wiki/German_tank_problem https://www.statisticshowto.com/german-tank-problem/
The green highlighted columns (real data) is sitting at the link i posted. Scroll across to see all columns. Serial Min/Serial Max is from current real data range (real examples found so far). Modelled Min/Modelled Max is the theoretical data range. This changes according to how you set your start value for 20217-1 Modelled Min (in orange) and the Serial Range (in orange).
Amazing, thanks. It's worth emphasising that this analysis is based only on what you've named the first execution Polarouters. I can't remember if you've explained the differences on this forum before but I've linked your article below for completeness. So the theoretical maximum for each ref of 170 applies only to the first execution. https://universalgenevepolerouter.com/polarouters-vs-polerouters-1st-2nd-execution-s-and-h-cases/ Based on the current data on your website, the approach taken to the serials/production of the second execution Polarouters seems to be quite different. Would you agree or do you think that it's too early to reach any conclusions because you need more data? Perhaps more second execution watches of a particular ref were produced if there was greater demand?
2nd execution Pola are much less clear. The subreference ranges for these seem to overlap, so definitely need more data. Perhaps there are smaller batches within these ranges that have not revealed themselves yet... Or perhaps something else.
The proportion of these that have survived would be fascinating to know. There may have been 170 but I wonder how many are really still out there? What's the highest number of serial numbers you have for a particular individual first exec pola ref? It wouldn't surprise me at all if the survival rate of the steel SAS Pola was one of the highest, as these watches were connected with a significant event and a career and have understandably been handed down within families. I think 15 SAS Polarouters have surfaced to date. Do any of the other refs get close to 15?
Most are around 5-7 of each, but as low as 3, and as high as 10. SAS 20217-6 are the highest number found so far at 15. It could be as you say, that more have survived as they are more likely to be handed down and kept in families as they commemorate something special. Or, it could be that they stand out the most as something special, so are more likely to be googled and sent to auction, rather than just being sold for $5 at a fleamarket. It could also be a combination... Just going to keep plugging away and collecting/adding serials when we find them.