map man
12-06-2015, 23:42
A few months ago Laurie Potteiger of the ATC posted a bar graph showing how many NOBOs completing thru-hikes in the 2010-2014 seasons started from Springer on each date. She did this to illustrate the times of the year that NOBOs might want to stay away from to avoid overcrowding. From this bar graph I was able, by blowing it up to fill my screen and doing a lot of squinting, to figure out the numbers for each date -- 2549 thru-hikers in total. Here is a link to that bar graph:
http://www.whiteblaze.net/forum/attachment.php?attachmentid=29669&d=1421990260
In addition, mountain squid (Tim) here at WB has been maintaining a database for years showing the intended start dates for WB thru-hiker hopefuls who have chosen to register their intentions here at WB. I looked at the years 2010-2014 in these databases, and subtracted any SOBOs, sections hikers or hikers doing FLIPS so that I could tabulate just NOBO thru-hiker start dates (there ended up being 2275 hikers in all), so I could directly compare them to the ATC calculation of start dates for those same years. Here is a link to the page for mountain squid's databases:
http://www.whiteblaze.net/forum/content.php/28-whiteblaze-thru-hikers
Comparing these two databases should give an idea of what start dates have a higher or lower percentage of completers. So here is a list showing the percentage of hikers who have started in a certain date range compared to the total in all date ranges . I used the same date range groupings that mountain squid uses. The first number is intended start date range percentages for WBers and the second number (in parentheses) is start date percentages for completers:
1.2% (.4%) January or earlier
10.1% (7.6%) February
15.1% (11.5%) March 1-7
14.3% (13.6%) March 8-14
17.5% (16.8%) March 15-21
13.4% (16.4%) March 22-31
12.8% (15.8%) April 1-7
10.8% (12.6%) April 8-30
4.8% (5.3%) May or June
By dividing the second number by the first number we get a "factor" showing relative likelihood of completing a thru-hike, with 1.00 being the number representing the overall average:
.33 -- January or earlier
.75 -- February
.76 -- March 1-7
.95 -- March 8-14
.96 -- March 15-21
1.23 -- March 22-31
1.23 -- April 1-7
1.17 -- April 8-30
1.10 -- May or June
Or to put it another way, if we assume that around 25% of NOBO thru-hiker hopefuls end up completing the trail, which is a ball park figure for what the ATC estimates is the general completion rate for NOBOs, then completion percentages for the start date ranges can be estimated:
8% -- January or earlier
19% -- February
19% -- March 1-7
24% -- March 8-14
24% -- March 15-21
31% -- March 22-31
31% -- April 1-7
29% -- April 8-30
28% -- May or June
Am I saying that it is "easier" to complete a thru-hike with a start date after March 21 than it is before then? No, I am not. For instance, if hikers with more experience, or hikers who know themselves to be fitter than average (groups with a better chance of completing, perhaps) also choose in greater numbers to start later then that would have an effect on completion rates by start dates. But for those who start at really busy times in early and mid-March in the belief that the earlier they start the better their completion chances, these numbers should give them pause. And perhaps that might encourage more people to start a little later and disperse some of the big numbers of people starting at traditionally popular times the first three weeks of March.
Now for the disclaimers: Yes, there is potential for sample bias any time you have a partial population stand in for an entire population. There isn't a whole lot of potentail for that in the ATC's numbers based on 2000 miler applications because if you compare their numbers with the physical count of NOBO thru-hikers that Baxter State Park does, it sure looks like the ATC numbers represent at least 75 to 80 percent of thru-hikers.
In comparison, our WB databases for 2010 to 2014 of hikers intending to thru-hike probably only represent around 20%-25% of all NOBO thru-hike starters. That is still a pretty darn large sample. And even if you make the case that WB thru-hike registrants skew older and perhaps more male than the overall population of thru-hikers, that doesn't automatically mean that older hikers or male hikers make significantly different choices about when to start. The only exception that seems likely to me is that perhaps college-age hikers are more likely to start in May, after the school year ends, and that could potentially mean that May starters could be under-represented in the WB databases, though I doubt if the distortion is dramatic.
Finally, thank you Lorrie Potteiger and mountain squid for all the time and effort you put into tabulating or maintaining these charts or databases. They shouldn't be held responsible for the way I chose to use the numbers produced by their labor.
http://www.whiteblaze.net/forum/attachment.php?attachmentid=29669&d=1421990260
In addition, mountain squid (Tim) here at WB has been maintaining a database for years showing the intended start dates for WB thru-hiker hopefuls who have chosen to register their intentions here at WB. I looked at the years 2010-2014 in these databases, and subtracted any SOBOs, sections hikers or hikers doing FLIPS so that I could tabulate just NOBO thru-hiker start dates (there ended up being 2275 hikers in all), so I could directly compare them to the ATC calculation of start dates for those same years. Here is a link to the page for mountain squid's databases:
http://www.whiteblaze.net/forum/content.php/28-whiteblaze-thru-hikers
Comparing these two databases should give an idea of what start dates have a higher or lower percentage of completers. So here is a list showing the percentage of hikers who have started in a certain date range compared to the total in all date ranges . I used the same date range groupings that mountain squid uses. The first number is intended start date range percentages for WBers and the second number (in parentheses) is start date percentages for completers:
1.2% (.4%) January or earlier
10.1% (7.6%) February
15.1% (11.5%) March 1-7
14.3% (13.6%) March 8-14
17.5% (16.8%) March 15-21
13.4% (16.4%) March 22-31
12.8% (15.8%) April 1-7
10.8% (12.6%) April 8-30
4.8% (5.3%) May or June
By dividing the second number by the first number we get a "factor" showing relative likelihood of completing a thru-hike, with 1.00 being the number representing the overall average:
.33 -- January or earlier
.75 -- February
.76 -- March 1-7
.95 -- March 8-14
.96 -- March 15-21
1.23 -- March 22-31
1.23 -- April 1-7
1.17 -- April 8-30
1.10 -- May or June
Or to put it another way, if we assume that around 25% of NOBO thru-hiker hopefuls end up completing the trail, which is a ball park figure for what the ATC estimates is the general completion rate for NOBOs, then completion percentages for the start date ranges can be estimated:
8% -- January or earlier
19% -- February
19% -- March 1-7
24% -- March 8-14
24% -- March 15-21
31% -- March 22-31
31% -- April 1-7
29% -- April 8-30
28% -- May or June
Am I saying that it is "easier" to complete a thru-hike with a start date after March 21 than it is before then? No, I am not. For instance, if hikers with more experience, or hikers who know themselves to be fitter than average (groups with a better chance of completing, perhaps) also choose in greater numbers to start later then that would have an effect on completion rates by start dates. But for those who start at really busy times in early and mid-March in the belief that the earlier they start the better their completion chances, these numbers should give them pause. And perhaps that might encourage more people to start a little later and disperse some of the big numbers of people starting at traditionally popular times the first three weeks of March.
Now for the disclaimers: Yes, there is potential for sample bias any time you have a partial population stand in for an entire population. There isn't a whole lot of potentail for that in the ATC's numbers based on 2000 miler applications because if you compare their numbers with the physical count of NOBO thru-hikers that Baxter State Park does, it sure looks like the ATC numbers represent at least 75 to 80 percent of thru-hikers.
In comparison, our WB databases for 2010 to 2014 of hikers intending to thru-hike probably only represent around 20%-25% of all NOBO thru-hike starters. That is still a pretty darn large sample. And even if you make the case that WB thru-hike registrants skew older and perhaps more male than the overall population of thru-hikers, that doesn't automatically mean that older hikers or male hikers make significantly different choices about when to start. The only exception that seems likely to me is that perhaps college-age hikers are more likely to start in May, after the school year ends, and that could potentially mean that May starters could be under-represented in the WB databases, though I doubt if the distortion is dramatic.
Finally, thank you Lorrie Potteiger and mountain squid for all the time and effort you put into tabulating or maintaining these charts or databases. They shouldn't be held responsible for the way I chose to use the numbers produced by their labor.