gauvins

For some reason, I've looked at trends.google.com to get a feel for the upcoming season. Unsurprisingly, search for camping is down from the peak of May 2020. Interestingly, the recent peak for bike touring occurred in 2017. Someone would know what triggered the interest back then? (FWIW -- "France Travel" is setting records).


Relative search volumes: bike touring in blue; bikepacking in red, United States, 2004-2022 (https://trends.google.com)

79pmooney

I went to the Cycle Oregon Kickoff Party that January and learned that it was hitting Crater Lake from the north that December. Also course profile-wise, an excellent one to do fix gear. (Few long hills with a minimum of need to stop and flip the wheel.) 30 miles of gravel! So I set my Mooney up as a triple chainline, gravel worthy fix gear. (38,35c tires, 46-42-36 X 13, 17 and 21 gears. Debuted this with Forumites that summer.

Apparently that triggered a worldwide interest in touring the old way. Fix gear like they did it in the 1890s. With my obsession with getting fit enough to ride this monster, I completely missed that I had set a world-wide trend. Then CO canceled due to the fires and smoke. I fell into a depression and hid for months. Never learned until just now the trend I started.

skidder

Isn't 2017 about the time that 'adventure cycling' and 'bikepacking' style touring became popular? vs traditional touring bicycles that looked like more robust touring bicycles? (Trek 520, Surly LHT and similar).

I have no idea why France would come up, other than it seems to be a nice place to spend a few weeks riding through the towns/villages/ historical places that are there. Nice Mediterranian style climate in the south, wetter in the north, so you have that variation, too.

indyfabz

Must have been my 2017, two week adventure in MT and ID, including Glacier, then my week long ride across PA, plus some shorter trips.

But I also logged some miles in 2018, including northern VT home to Philly, with a weekend stop in MA for my 35th high school reunion. That September I had to be “rescued” from another cross-PA tour due to some crazy weather that moved in unexpectedly and the threat of hurricane remnants moving up the east coast.

I did get back to MT and ID for two weeks in 2019 and then had diagnostic surgery which kept me out of action until September, when I did an up and back on the GAP.

We all know what happened in 2020. My flight schedule for another MT and ID trip got so messed up that I had to cancel it. At least I got my airfare refunded, but I had to fight for it. Managed to get in another cross-PA tour that September.

saddlesores

2017? .............. i seem to have missed it completely!

spent the entire year in thailand, where they use the buddhist solar calendar.......so that was year 2560.

oops.

mev

Change from "US" to "Worldwide" and you will see the boom continue through 2018, 2019 and even 2020. Pick "Canada" and the peak also appears slightly later (except for an interesting spike in 2004). So my question is more - why did the bike touring term start to saturate in the US after 2017? I can think of some larger events having controversies in 2019 but that was slightly later Ragbrai (do a search for "Carson King") and Tour Divide. Also the 2018 wildfire season was significant along the west coast. It was later eclipsed by 2020, but relatively speaking 2018 was worse than the several years immediately proceeding.

Of course, I'd like to think I contributed (no - not to the fires or controversies ) - as I spent all of 2017 on the road while cycling between Tepic, Mexico (where I celebrated the New Year) and Ushuaia, Argentina (where I celebrated the southern hemisphere summer solstice)... That trip had started in June 2016.

Tourist in MSN

If this was a google search compendium, looks like all the new bike tourists have already figured out what bike touring is, no longer need to search for it.

2017. That would be my Florida tour. And a five day tour with a friend that wanted to test out his bike and new tent to see how the bike upgrades work before his cross country trip on it. And I had just built up my new Lynskey Backroad, so that bike needed a test ride too. And, my BWCA canoe trip from Gunflint lake to Seagull lake via Saganaga lake was that year. As was a two week kayaking trip on the shore of Isle Royale in Lake Superior. That was an outstanding year, I think I would have slept in a tent forty some nights that year.

staehpj1

I hesitate to say this for fear of veering into the political arena, but the 2016 election changed the US in a lot of ways. For example here in Florida (panhandle) I noticed an immediate downtick in civility of some drivers toward cyclists. Not all drivers, but the jerks among them suddenly seemed emboldened to yell slurs or maybe even throw things at cyclists, behavior that had seemed to have pretty much have dissappeared in the last couple decades was suddenly back.

I could well imagine this climate discouraging bike touring at least a bit.

fishboat

Ran the same setup and results look a bit different, not sure what the difference would be.. The query specifics are illustrated.

If I were to guess, the rise in bikepacking in general would be due to advertising that drives the herd to buy new stuff. Gravel bikes and bikepacking is a growth area for bike mfgs, at least compared to conventional road, mtb, and hybrid-fitness bikes(excluding e-bike data). With respect to 2017..it looks different, but I suspect a statistical test would show no real difference in the 2017 data vs the entire data population. The seasonality of the data complicates things a bit. The 2017 visual blip up.. it would be interesting to compare advertising budgets over 2014 to 2019 and see if more dollars were spent in 2017...assuming sensible data could be compiled.

The blip up in 2017 in "bike touring"..again a guess.. is confounded data with "bikepacking". Many people use the terms bike packing and bike touring interchangeably. Particularly non-BF members (the vast majority of bike riders in general, and specifically those with limited tech knowledge of bikes and the terminology surrounding bikes). If excess bikepacking advertising dollars were pumped into the market then people searching on bike touring, while actually interested in bike packing, would drag the bike touring plot up.

I'd need the raw data for better estimates(I used to do this for a living at one point).

fishboat

Passing thought..I just checked the sensitivity of the query to minor changes in the search topic..yikes.. The raw data used for this google analysis needs considerable pre-processing/cleanup. Example..We know "bikepacking" and "bike packing" are the same thing. The google analysis is sort of dumb as it doesn't recognize this.

The resulting plots are fun, very general eye-candy, but I wouldn't bet money of any of the results.

The plot below is the result of changing the "bikepacking" keyword to "bike packing"

Tourist in MSN

Those graphs that start in 2004, I ran fewer searches back then since I was using dialup. I might have still been using Alta Vista for searches, but I am not sure when I switched to Google.

But I find that the peak for bike touring in 2004 is interesting, as that is the year that I built up my first-year-of-production LHT.

At that time if you wanted an off the shelf touring complete bike in USA, you were looking at a Fuji or Cannondale or Trek 520. Some toured on Volpe but it was not sold for touring. (LHT was frame only, not complete bike.) There were some custom or semi-custom manufacturers too, but you would have to search hard to find them, Bike Friday would be in that group, as Bruce Gordon, Co-Motion, Independent Fabrications, etc.

There were more options for complete bikes in Europe, at that time there were several in UK alone, but it looks like the google search data is USA only.

fishboat

There's a drop down on the left side that let's you choose between several dozen countries as well as a composite selection of "Worldwide". Given their lack of understanding that "bikepacking" is equivalent to "bike packing", I'd have to wonder how they handle a term like "bikepacking" on a global chart as I doubt they are quantifying the, for example, Turkish or French terms for bikepacking as the English equivalent of the term bikepacking. If I were to guess..their worldwide chart simply quantifies people searching in, for example, France or Turkey using the English word, bikepacking.

The 2004 thing is rather mysterious. I just did a search for "backpacking" and 2004 had the most hits. I searched for "needlepoint"....and 2004 had the max number of hits. "Horseracing"...and 2004 had the greatest number of hits. "Culinary pig brains"..June 2020 was the max. From all this I can conclude the peak pandemic period created a huge serge in interest in bicycles as well as eating hog brains.

Hard to say if these charts mean much without understanding specifically how they are obtaining data, how they preprocess it, and how they are working it up. The (apparent) math they are using is straightforward. They normalize the data to the max number of hits. (Example.. the term "bikepacking" had the largest number of hits, say 5M, in 2004. In 2009 they had 2.6M hits on bikepacking. The plot point for 2004 is then 100%. The plot point for 2009 is then (2.6M/5M) * 100= 52%.

There's good potential for interesting data here in general. No doubt...Google has entire departments of analysts working up the data they generate and are selling it in various formats.

(..wandering off topic here..but everyone has heard of their "credit score". The credit score is just one of dozens (and dozens) of often obtuse "scores" that follow folks based on how they interact with the world. The credit score is just one of the scores that has bubbled to the surface and people are well aware of it. Not tin hat conspiracy theory stuff..this is the real world.)

gauvins

you use [bicycle touring] [ bikepacking]. I used [bike touring] [bikepacking].

Trends is an amazing tool. Try bike vs bicyle (both as a term and as a concept). Hours of intrigue in store

fishboat

Your results will be heavily influenced by motorcycle touring then..at least I'm thinking so by the number of motorcycles I see for sale in the craigslist bicycle listings.

In these parts (Great Lakes region) if you mention a bike to anyone and they'll start discussing Harleys, Hondas, Yamahas....

Comfort is King

I know exactly what happened. Just look at who won the POTUS the year before. Everyone began to realize our society had crumbled. The environment was dealt a final, deadly blow. Nothing will recover. Why not try to get away?

saddlesores

not the same.

bikepacking is bike touring with strapped on bags but no racks.
bike packing is disassembling your bike to put in a box to put on a plane.

jamawani 

Google searches - -
And people who actually do it - -
Are two entirely different things.

staehpj1

Yeah, folks try to get their for sale ads to hit all kinds of topics whether appropriate or not. That is true of just about any category on craigslist.

That said, I doubt if there is that much confusion with motorcycles when it comes to google searches for bike touring or bikepacking. If you do a google search for either bike touring or bikepacking pretty much all the hits seem to be for bicycle touring. I guess that doesn't prove what folks are searching for when the search with those terms, but what they find seems to be bicycles.

fishboat

Agreed..however we're working with tens to hundreds of millions of hits on a search and a short preview of hits isn't indicative of the entire dataset contents. One really needs to review the raw data to see what it contains and why. I used to work with logistics datasets in the range of 1-2 million transaction/transportation data points. Rather small in the world of "big data" that Google works with, though large enough to be really challenging to understand what the data actually contains. With Google sized data, one can only really get an estimate of what it contains.

Even with my moderate-sized datasets, it would take me 2-3 weeks (full time) of preprocessing to scrub the data of all the junk that prevents clean conclusions to be extracted with a proper workup. And that was with data that regulated input from order processing and distribution center software. With google searches there's no regulated input. I'm sure the raw data is a wild west show, from an analyst's reduction point of view. (I'd love to visit a Google analysts group to see what they do and how they do it)

If G-trends would allow boolean-constructed queries, that would help. I haven't tried it.

In the end..shrug..doesn't matter. Google trends is interesting and yields some eye-candy. Though for any use where there's anything at stake..it's worth what you pay for it.

staehpj1

For sure. That said, google searches can be at least a vague indication of level of interest in a topic.

saddlesores

maybe..........

i'm sure mr. googles is keeping tabs on what YOU search for in your search history and cooking crumbs, so the search results MAY be tailored to you specifically.

the retired dude in south florida with a goldwing and annual pilgrimage to sturgis might get completely different results from you with the same search terms. maybe if you do the same search on different computers you'll get different results.

prj71

I just rode my bike in 2017 and didn't care about trends.

staehpj1

True. I hadn't considered that. I do have some interest in motorcycles, but not so much in recent years. Still, I have done some recent searching for possible long trips on a dual purpose on/off road motorcycle though. I also have a more distant past racing motorcycles and using them as transportation. I have done some recent searches for motorcycle race topics to check on past friends who are mostly (not all) retired from racing now. Some have kids or grand kids who are still racing. I have eved done some searches for motorcycle models and prices. It may just be that I have way more interest in bicycles in my search history and not enough motorcycle interest to slant the search.

For the heck of it I tried doing a search in incognito mode and the result was pretty much the same, but I don't have any illusions that incognito mode is actually very incognito so that doesn't surprise me.

raceboy

2017? Everyone jumped on their bike to go checkout the Solar Eclipse.

thumpism 

It was me. I did my TransAm attempt in '16 and it took a while for the word to get out, but when it did...