Insights from 30 Years of DCSki Data

I’ve been doing quite a bit of coding and making enhancements to DCSki lately. Some of those changes have been visible, such as the new ability to Like forum posts, while others have been behind the scenes.

One of my recent projects was building some internal analytics tools that let me explore DCSki’s history in ways I haven’t really been able to before. Because DCSki has preserved its articles and forum discussions going all the way back to the 1990s, there’s a surprisingly rich collection of data sitting underneath the site.

Naturally, once I had the tools, I started poking around. Here are a few things I discovered.

DCSki currently has 1,750 published articles, dating back to May 1997. Collectively, they’ve accumulated more than 13.4 million views.

The volume of articles has varied dramatically over the years. The busiest year was 2003, when DCSki published 129 articles. There were also 108 in 1998, 104 in 2007, and exactly 100 in 2008.

Those numbers reflect how the site itself has evolved. In the early years I published a much larger stream of shorter news and conditions updates. These days there are fewer articles, but I tend to spend considerably more time researching and writing each one.

The most-viewed article in the entire archive surprised me: “Winter Scenes: Some Shots from the Region,” published in 2007, with more than 87,000 views. It only has three little images, so that's kind of weird, but the Internet works in weird ways sometimes.

Next are John Sherwood’s “How Resorts Make Snow” from 2003, with more than 54,000 views, and Jim Kenney’s “The Toughest Slope in the Mid-Atlantic” from 2002, with more than 51,000.

One thing the numbers make especially clear is the longevity of some stories. Articles about snowmaking, difficult slopes, ski patrol, first aid, driving to West Virginia ski areas, and old firsthand reports continue attracting readers many years after they were written. I’ve also just started collecting more detailed article traffic data, so over time I’ll be able to see when an old story suddenly gets rediscovered and starts attracting readers again.

Then there’s the Forum.

There are now more than 100,000 forum posts preserved in DCSki’s database. Together, they provide a pretty remarkable running record of Mid-Atlantic skiing over the years.

One of the things that jumped out at me while exploring the data was the longevity of the community. There are people still regularly visiting and contributing to DCSki today whose accounts date to 1999, 2000, 2001 and the early 2000s. Some have contributed thousands of posts along the way.

That longevity has given the Forum an interesting kind of institutional memory. There are discussions about long-gone ski areas, major resort changes, memorable winters, lift installations, ownership changes and countless other subjects — often containing firsthand observations written while those events were unfolding.

The analytics also give me some new ways to look at which discussions seem to resonate with readers. There are already useful signals such as how many times a thread has been viewed, how many replies it has received, how recently it has been active, and how quickly a discussion is growing. The new Like feature adds another useful signal: some human curation from DCSki readers themselves.

That opens up some interesting possibilities. For example, I’d eventually like DCSki to be able to automatically surface relevant Forum discussions alongside articles. If there are several potentially related threads, combining subject relevance with views, replies, recency, activity and Likes could help determine which discussions are most useful to highlight.

There’s an interesting intersection between articles and reader discussion, too. DCSki articles have accumulated more than 4,000 comments over the years. The most-commented article appears to be “Porte Crayon Ski Resort Considered” from 2003, with 133 comments. A 2000 story about the National Weather Service predicting a return of winter weather generated 75 comments, while a 2004 Canaan Valley and Timberline trip report drew 65.

The author statistics produced a few interesting discoveries as well. Apparently I’ve written 969 of DCSki’s 1,750 articles, accounting for a little over 6 million views. The late, great Connie Lawn contributed 173 articles, Jim Kenney 138, John Sherwood 112, and Robbie Allen 66. Looking through the complete list brought back a lot of names and stories I hadn’t thought about in years.

All of this has reinforced something I’ve always thought was pretty special about DCSki: the history is still here.

An article written in 1998 is still available. A Forum discussion from the early 2000s can still be read today. Through multiple generations of servers, databases, programming languages and site redesigns, I’ve taken painstaking efforts to convert and bring all of that history forward rather than leaving it behind.

Now I’m finally building some tools that let me step back and see what nearly three decades of that history looks like as a whole.  These tools are also enabling me to write some of the feature stories you've seen hitting the home page recently, which dive into the history of Mid-Atlantic skiing and provide some (hopefully) relaxing reading and reminiscing as we anxiously await the return of winter.

I agree about the value of this site for "posterity".  Most other ski forums have either shuttered or evolved in ways that eliminated a large volume of early inputs.  An ironic thing about the digital era is that we humans collect all this info, photos, etc, only to mostly lose track of it within hours or days.  A lot of my earliest digital ski images have been lost in cyberspace, but some are still retained here in DCSki articles.  Whereas, I have paper photo albums that are 50+ years old, but unfortunately relatively few old ski photos.

It's interesting that peak views of DCSki content seem to be from about 20 years ago.  I never made the switch to facebook, reddit, etc., so ski forums remained relevant for me.  What online sources are most popular now to obtain info that is similar to what is presented on DCSki?  One more thought, I like the civility of DCSki.  I've avoided facebook and the like because of crudeness and lack of moderation. 

A key reason why older stories often have much higher view counts is simply that they've been around longer to gain those views.  A story from 20 years ago has had 20 years of incoming traffic from Google searches, etc.  I've seen some (much) older stories suddenly go viral simply because they were on a topic that suddenly was of interest to people, and Google or other search engines were happy to direct readers that way.  Anecdotally, I think current stories are getting more views in their first couple weeks of publication than equivalent stories from 10, 15, or 20 years ago, but that would be an interesting test to run.

Facebook is a bit of a mess.  The algorithms "choose" what you see and what you don't, and are optimized around business principles that I don't believe align with most of our personal principles.  And they're just overrun with spammers and bots.  I do maintain a DCSki group on Facebook, and there can be good discussions there, but I prefer the quality and true community feel of discourse that happens here in the forums.  The challenge for independent sites like DCSki is that people have to go out of their way to visit them.  There are now a few big social media players (Facebook, Instagram, TikTok, etc.) that have hooked people into visiting many times a day, every day.  As they doom scroll, content is pushed directly in front of their faces, optimized to be addictive.  So if you have content there, it's basically being delivered directly to someone's "virtual house"; they don't have to go anywhere to see it.  So you get more views (if the algorithm decides to show your content, which is a big if), but generally from less engaged readers.

What's really frustrating for content producers like me is that Facebook/Meta and its peers do everything possible to prevent outbound links from content posted within their walled gardens.  For example, if I post a link to a new DCSki story in the DCSki Facebook group (a group that I manage and moderate, and that people joined because presumably they want to see that kind of content), Facebook intentionally limits its viewership -- effectively "shadow-banning" the content -- because they don't want users clicking out of Facebook where they will no longer see Facebook ads.  It's incredibly frustrating because really violates the principles that the Internet was originally formed on.  But it is what it is.  I can only continue trying to put out good, relevant content for fellow snowsports enthusiasts and hope that enough will find it compelling enough to visit DCSki directly.