Jump to content

December Statistics


holograph

Recommended Posts

I obtained the December updates from the NGS, but it looks like they didn't post any December GEOCAC recoveries. The statistics page has been updated with the current GC.com stats, but there are no changes to the NGS stats, and hence no changes to the maps, either.

 

I'll check the NGS datasheets again in about a week to see if they catch up on the December recoveries.

Link to comment

Deb had a lot of vacation to use before the end of the year, so my bet is that she has been off for quite a while and has not had time to update the recoveries. She hasn't answered an email I sent her over two weeks ago.

A lot of civil service workers who are too busy to take vacation time off during the year tend to take "use-or-lose" annual leave at year's end, when the workload tends to be slow in most offices anyway. I wonder if that's also the case with Cheryl Malone, who has a couple of large lists from me of PIDs listed in the wrong county.

 

-ArtMan-

Link to comment

PapaBear:

 

How did you view the December uploads? NGS does not have "2007" in the LOAD DATE drop-down box. [Why do they always wait until Feburary or March to get the website modified? Grrrr. Don't they believe the new year is coming?) The Power Squadron and I duplicate the recoveries of some stations during the first few months of every year, because we cannot view each other's finds.

 

Back to the 2007 entries you saw. Are you accessing an archived file?

 

-Paul-

Edited by PFF
Link to comment

PapaBear:

 

How did you view the December uploads? NGS does not have "2007" in the LOAD DATE drop-down box. [Why do they always wait until Feburary or March to get the website modified? Grrrr. Don't they believe the new year is coming?) The Power Squadron and I duplicate the recoveries of some stations during the first few months of every year, because we cannot view each other's finds.

 

Back to the 2007 entries you saw. Are you accessing an archived file?

 

-Paul-

I did not do an archived file download, I merely accessed the PIDs (individually) for which I had submitted logs in December 2006, e.g. http://www.ngs.noaa.gov/cgi-bin/ds_mark.prl?PidBox=KU4047 and similar. Friday moning my logs were not there. Friday evening they were there.

 

I think perhaps there were two separate issues:

 

1) Getting the logs into the database - that appeares (based on my admitedly limited sample) to have been done sometime Friday.

 

2) getting the download function to work for 2007 - based on what you say, it still does not work.

Link to comment

As of yesterday, there were 33 Nov and 392 Dec GEOCAC recoveries posted in January, none later than Dec 18. Here is the breakdown by state. I may wait a couple more days before doing the update.

 

St Nov Dec
AL   1  15
AZ   6 217
CA   6  32
CO   0  24
IL   0   6
MA   0  11
MD   0   3
MO   0   1
MS   0   2
NC   0  14
NJ   0   6
NY   1   8
OH   1   4
OR   0   1
PA  18  13
SC   0  29
VA   0   1
VT   0   2
WI   0   1
WV   0   2

Link to comment

As a quick, slightly related note, geocaching.com just broke 90,000 PIDs found. Actual PIDs found (I would guess) is more like 85,000. Congrats to all.

You really think there are that many bogus entries? One in twenty?

 

How would you estimate this? You could pull a random sample of all PIDs, but how could you be sure which found entries are legit?

 

-ArtMan-

Link to comment

As a quick, slightly related note, geocaching.com just broke 90,000 PIDs found. Actual PIDs found (I would guess) is more like 85,000. Congrats to all.

You really think there are that many bogus entries? One in twenty?

 

How would you estimate this? You could pull a random sample of all PIDs, but how could you be sure which found entries are legit?

 

-ArtMan-

A while back there was a debate on what exactly thoes stats mean. When it says "recovered", does it mean "found", or does it include all types (found, not found, destroyed, post a note) of logs?

 

 

Patrick

Link to comment
A while back there was a debate on what exactly those stats mean. When it says "recovered", does it mean "found", or does it include all types (found, not found, destroyed, post a note) of logs?

 

Patrick

Ah, good point.

 

In the last 7 days, 654 benchmarks have been logged by 332 users.

Overall, 90103 benchmarks have been recovered in 125753 logs.

This is the way the stats are presented on the benchmarking home page. "Logged" seems to me to suggest any kind of log — found, unfound, or note. "Recovered," on the other hand, seems to suggest that the mark was actually found. That doesn't make much sense to me, however, it seems that an apples-to-apples comparison is intended here.

 

Anyone want to count them? ;-)

 

-ArtMan-

Link to comment

I've updated the statistics page with the latest datasheets from the NGS. I'll update the maps sometime later today or tomorrow.

 

We should congratulate AZcachemeister for submitting a record number of recovery reports in a single year. Shorbird also exceeded previous records, and you all put out an amazing effort this year!

 

If there are additional 2006 recoveries posted by the NGS, they will eventually show up in the 2006 column, even after we start seeing 2007 statistics.

 

edit: Oops, I see shorbird's count for the year was just below seventhing's previous record. But it is still an amazing accomplishment!

 

edit:

The maps have now been updated.

 

geocac_map_t.gif delta_map_t.gif

Edited by holograph
Link to comment

I'm not sure what you mean. The statistics from GC.com logs were obtained Jan 4. The statistics from NGS were obtained yesterday. The NGS statistics include all new GEOCAC recovery reports -- there were 643 recoveries from December. 25 January recoveries sneaked in, too.

 

edit: m&h, I see that your geocaching.com stats were shown as zero. It turns out that the ampersand in your name was causing problems. That has been corrected and the stats have been updated.

Edited by holograph
Link to comment

Sorry to hijack your topic a little, holograph. As always, your work is cool.

 

ArtMan, I recall a little bit of discussion about this in regards to predicting when geocaching.com would reach 100k "recovered" marks, which I think is marks that are logged as Found.

 

In regards to the error rate of the number of PIDs that have Geocaching.com Found logs: I might have overstated the issue a little, but I bet that the error rate is something like 3-4%. This isn't the error rate in overall Found logging, which includes duplicate logs on the same stations, it's the error rate in overall stations described as Found. In going through the photo gallery and scanning around the database for the Agency benchmark list, I've run acros a lot of incorrectly logged stations, mainly ones logged by caching muggles who don't know what they're looking for. These are the usual problems (a wrong disk on a highly trafficked mountain, an RM recovered as the main station, and disks found while camping, that haven't been checked up on by a benchmarker). The error rate in the photo gallery, etc. is lower than 3-4%, but that's only the disks that have a photo I can compare to the person's log. In the photoless logs, there's often hints that they didn't get the right station, and then you can guess as to what fraction of the "Found the station" bare logs should be added in as erroneous. This ties into the "quality logs" discussion held a little while ago.

 

By comparison, I would guess that there's probably a 1.5-2% +/- error in the overall number of logs submitted to the NGS over the same period of time as the Geocaching.com logs cover, mainly due to errors by the USPSQD and to a certain extent GEOCAC. I've sent in 2 or three erroneous reports that I know of (later corrected, and mainly NF, so...) in 870 NGS logs, so I've probably sent in a few that I'm not aware of. Also, I've had several duplicate logs/misnumbered logs that I've straightened out, from incorrectly submitting on the website. Not all of these are corrected.

 

Stupid snow!!! This is usually the time of year that I can get out and recover while ya'll are stuck in the house, and we've had 18 inches here in Albuquerque, and it's still in the ground in the places where I want to benchmark.

Edited by BuckBrooke
Link to comment

Stupid snow!!! This is usually the time of year that I can get out and recover while ya'll are stuck in the house, and we've had 18 inches here in Albuquerque, and it's still in the ground in the places where I want to benchmark.

...snow?! what snow...we've been waiting for snow in PA to help soften up the grass & pack down the briars so we gan go out into the woods for some tri-station hunting!

 

Question for Holograph (forgive me if the answer is already out there somewhere) just out of curiosity - what is an average amout of total recoveries (us, the Sqd., and all the other agencies) the the NGS gets in a month?

Link to comment

Ha! There's a new challenge! To turn my home county (maroon?) - (whatever that color is for more than 100 finds) :D !

(Oops. That ain't gonna work. Who was it who had the county lists? Only 173 benchmarks in Morris County? There's no way that 101 of 173 are going to be findable. I'm running a 50% DNF rate. Oh, well. It was just a thought.)

Edited by Harry Dolphin
Link to comment
EZ0422 is given a position of 0°N 0°E, which seems outside of NC.

 

Not an error. We claim the equator as part of Tarheel Territory. :)

 

Actually, Neweyess (from Charlotte, NC) went down there on a business trip and he recovered the Equator. But now he can't figure out how to file a report!

 

When he returned, he gave me a miniature statue of the Equator monument. It must be impressive in person!

 

-Paul-

Link to comment

Actually, if you call up the datasheets for those stations, you see that the lat, lon is missing on the datasheet, which isn't supposed to happen. All datasheets are supposed to have coordinates.

 

We should point that out to the NGS. They may have a database problem.

 

edit: I just did a quick scan of all the datasheets that I downloaded, and only found that one pid, EZ0422, to be missing the lat, lon.

Edited by holograph
Link to comment
I just did a quick scan of all the datasheets that I downloaded, and only found that one pid, EZ0422, to be missing the lat, lon.

 

When you eliminated EZ0422, the southernmost benchmark changed to one which I believe is correct, in the far southeast corner of the state.

 

While we're on the subject, please take a look at North Carolina's "westernmost" mark. Your chart shows FC0348, which is in Polk County, Tennessee. I've always felt the most western mark in North Carolina was FC1777 (18 EG 19) at N35 01 16.2, W084 18 54.1. It is 0.1 mile east of the Tennessee line. There are two others nearby (FC1775, FC1776). But I believe FC1777 wins, base on Longitude.

 

-Paul-

Link to comment

While we're on the subject, please take a look at North Carolina's "westernmost" mark. Your chart shows FC0348, which is in Polk County, Tennessee. I've always felt the most western mark in North Carolina was FC1777 (18 EG 19) at N35 01 16.2, W084 18 54.1. It is 0.1 mile east of the Tennessee line. There are two others nearby (FC1775, FC1776). But I believe FC1777 wins, base on Longitude.

 

You should let NGS know that the state is mis-coded on that datasheet. I'll exclude it from the NC extremes.

Link to comment

Mis-coding noted. That happens quite frequently. It seems the NCGS strays across the border on a regular basis. :ph34r:

 

I was joking with the SCGS rep earlier this week about protecting his "turf". He says South Carolina has set a few in NC, also.

 

While we were on that topic, I learned something interesting about marks which are directly on the border between two states. According to Alan-Jon Zupan, the state which comes first in alphabetical order gets the mark. Consequently, North Carolina is the registration when it involves the common boundary with South Carolina, Tennessee, or Virginia.

 

-Paul-

Link to comment

On the subject of extreme's, here's another nit-pick.

 

The northermost station in New Hampshire is Monument 475, not Monument 476 as listed. The reason you have what you have is because 475 is the boundary marker for the two states. It is listed in Maine (and you correctly list it as the westernmost station in Maine). I guess this is due to the "Alan-Jon rule" mentioned in PFF's post above.

 

Note: the 1:24000 USGS map of the area (Moose Bog quadrangle, 1989) incorrectly depicts the location of Monument 475. It is at the tri-point boundary intersection, not several hundred yards to the east as shown on the USGS map. If you looked at this map you would incorrectly infer that monument 475 is entirely within Maine. The older 15 minute map (here) is correct, as is the IBC map for this section of the border.

 

See the GC log I made for Monument 475 for some details: MON 475 IBC. I have some pictures of the 1858 state boundary stone at this point.

Edited by Papa-Bear-NYC
Link to comment

Oops, another one.

 

QH0624 "LOT" (Lat = 071° 30.045) is listed as the easternmost station in Vermont. I think there are two stations further east:

 

QH0623 "REF MON 517" Lat = 071° 29.982. This is on the VT/NH line and is listed in NH so you would have missed it.

and

QH0564 "BEECHER TABLET" Lat = 071° 29.091. This is unambiguously in VT and should get the prize.

 

I remember them from a datasheet search, but just now sorting on Latitude with GSAK for this county confirmed that BEECHER TABLET was the furthest east. GSAK obviously didn't pick up the one on the line since it's not listed in Essex county.

 

I've recovered REF MON 517 but not the other two. If I go back to the area I may search for both of them. They may be findable with GPS and a metal detector.

 

Hey! A new goal! A new list - recover all the extreme stations! Who will be first!! :unsure:

 

Edit Sorry to keep doing this. But for the southernmost station listed in vermont, look at this map: Map for MZ1193. Sure looks like this in in Massachusetts. Error in the datasheet, clearly.

 

Now, I promise to stop looking at these things.

Edited by Papa-Bear-NYC
Link to comment

In order not to clutter the "extreme" lists with a lot of unrecoverable or lost stations, I excluded "not found" stations from the extremes. If you click the "more stations" link at the category heading, you will see that QH0564 BEECHER TABLET is listed as the Easternmost station, but was excluded because its last report was "not found". If you find it and submit a recovery report, it will show up as the easternmost station in Vermont.

 

As far as the stations that sit on the boundaries, I have no way of determining which stations are exactly on boundaries. I simply list them in the state that is shown on the datasheet.

 

MZ1193 might be in the wrong state, but we also know that there are a few scaled locations that are wildly different than the actual location. You can't rely solely on the map. If you recovered it and could know for sure that it was listed in the wrong state then you could notify the NGS to get the datasheet changed.

 

In October of 2005, I generated a list of all stations whose datasheet coordinates placed them more than 1 km outside the county listed on the datasheet. The thread is here. The file is still available for download.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...