With our wild kinetic dreams

The correct number of bicycles one should own is determined through a mathematical formula well known to internet-savvy cyclists. Stephen Hawking resolved not to include any formulae in A Brief History of Time because he felt each instance might halve his sales, although he did eventually relent and quote one from Einstein. I however am not out for sales, and anyway, it's only a very little formula:

C = n + 1   [1]

where C is the correct number and n the current number of bikes.

So there is supposedly always another bicycle required to fill that little functional gap. It's about the need to transport a couple of bags of peat from the garden centre when you don't have a car and are too cheap for taxis; it's the need to travel with a bike on a train and take it into a hotel, and not worry about bookings or raised eyebrows; it's the need for something stable and unfalloffable for snow and black ice; it's the need for a really lightweight lowracer for smoking those roadies; it's the need for something with an all-enclosed multi-speed drivetrain and weatherproof brakes for hills and salt-encrusted winter weather; it's the need for a decent mountain bike with disc brakes and nicer front end geometry because you're tired of wearing out rims and getting a crick in your neck after 15 miles of trails. Own only one bicycle? That's crazy talk, man.

On the Team Estrogen forum there are various suggestions in the thread Do I need two bikes or one?: 'You need at least two bikes; three or four is even better.', or simply that one needs three or four. One contributor even owns the seemingly ideal 3.5 bikes (the 0.5 apparently being crash-assisted).

In the discussion Is the perfect number of bikes REALLY always N+1? on Singletrack's forum, one contributor stated that 'Four bikes is all I need, and I think that's indulgence. Couldn't go below three though.' Clearly, owning only two bikes would simply make life impossible.

During 2008-09 the Cyclechat forum hosted the discussion How many bikes should a cyclist own?. The first suggestion was the beautifully disarming 'as many as necessary'. Another view was that 'five bikes are quite sufficient, unless you like off road biking in which case you need a couple of MTBs too.' Despite six pages and 80 posts, there was little agreement reached over what the correct number of bikes was, except that the formula itself was correct, more or less. For some a more appropriate version was (n+2) or perhaps (n+3). Indeed, a contributor on the London Fixed-gear & Single-speed forum noted that multiple bicycle ownership, whether realised or aspirational, was 'worse than crack'.

The Blasphemous Biker, writing in December 2010, explained his rationale for needing three bikes. These would, in his opinion, comprise a road/touring/Audax bike, a mountain or mountain-type bike, and a bike to use in cities and in bad weather.

In fact, Peter Walker of The Guardian picked up on the (n+1) movement and the Cyclechat thread in July 2010, in his article Confessions of a serial bicycle buyer. Walker himself said 'I now own four bikes ... it makes perfect sense.' The 69 comments on that article quickly demonstrate the relatively few people who own only one bike, though one commenter (commentator?) came tantalisingly close to a magic single machine with his Rohloff-hubbed multiple-tyre optioned Thorn Enduro. At least one person picked up on the dichotomy of need vs. want, and considering the environmental impact of owning many bikes, but until it's proven that four, five, six bicycles has a greater environmental cost than one car -- especially a modern 5-seater petrol hybrid -- I remain on the side of the human power movement even with its ongoing materialism.

Fortunately there are already rules in place to help govern such desperate soul-searching and financial husbandry. Within The Rules: the simple truths of bicycle etiquette, and under Rule 12: Fundamental Principle of Bicycle Ownership, it is stated clearly that three is the minimum acceptable number, with the correct number calculated using the (n+1) equation. The three are explained in greater detail in the Badass By Association: Winter Riding discussion and may be summarised as i) a rain bike; ii) an inclement weather bike; and iii) a nice weather bike. However, one contributor helpfully pointed out that road surfaces, frame materials and frame types should be additional considerations, and '117 bikes per person' was therefore justified.

But as we saw from The Guardian, if we start to distinguish between want and need, our happy 'correctness' goes out of the window; that is to say, correctness tends towards optimality. We might, therefore, look for an empirical solution, N, to the equation

N = n + (F_d * F_e)   [2]

in which the absolute desirability factor, F_d, would include subjective things like price of a bicycle (or multiple bicycles) divided by expected mileage, peer group rarity, size, total inertia, and total rolling resistance; and the enabling factor, F_e, ranges from -1 to +1 and includes measures of household or relationship stability, price divided by expected salary, security of rarity, and available space.

This is all fine and good, but such empiricism requires hard numbers, not just statements and good ideas. While the correct or ideal number of bicycles is clearly 'one more', we can look at reported ownership to determine a real-world optimum. Obviously it will be at least one, and presuming it to be fewer than infinity, and on the available data being post-limited to personally owned bikes rather than including those in the garage/hall/shed but belonging to other people, we must consider not just the bog-standard average but all three statistical averages, that is, is there a normal or offset-normal distribution of the results, and where are the peaks? Also to be borne in mind is the possibility that all reported results are skewed by all the I'm-not-a-cyclist, I-only-ride-to-the-park single-bikists steering clear of internet discussion forums and little surveys, and indeed, pottering around completely oblivious to the value of their collective bike usage. In fact, the oblivious might make up a statistically significant proportion of bicycle owners, if not in the absolute numbers of bicycles.

Davis, in California, reputedly has 2.1 bikes per person. A census on the DavisWiki website, with returns from December 2005 through to November 2009. suggested that across 132 wikizens there were on average 1.9 bikes per person, which is a pretty good result. Ownership was perhaps slightly adrift of statistical rigour, because the numbers ranged from one bike to as many as 18 (though perhaps understandably, not all 18 were high quality mileage machines, and not every contributor gave a precise breakdown of bikes per household member.

Closer to home, the Cyclechat discussion referred to above saw returns from 39 people and counted a total of approximately 133.5 personally owned bikes (the 0.5 referred to a tandem, whose other rider seemingly didn't participate). The average number of bikes owned was 3.4 per person, with both a median number and modal number of 3 bikes per person.

A more detailed survey, How many bikes are there in your shed/garage/house?, conducted on yACF in 2008 covered 91 forum members (accounting for repeats and updates) and a total of 511 personally owned bikes. In this survey, the average number of bicycles owned was substantially higher, at 5.6 per person, with a median number of 4 bikes, but the modal values occurred at both 3 and 5.

And on .citycyclingedinburgh, the small but succinctly named question How many bikes do you have in your possession currently?, asked in late 2009, has had at the time of writing returns from 29 people, who together own 79 bikes. Including one forum member who reported being particularly well-wheeled, the average number of bikes owned was 3.9 per person, and again both the median and modal number was 3.

So perhaps there is some truth in the Velominati definition. And in fact, the eternal question isn't limited to only pedal bikes. A Kawasaki KLR owner goes with three motorbikes as the optimum: 'One to do this, one to do that, and one you are trying to sell so you can get another one you crave to do the other.' A member of the Suzuki SV650 forum said that his significant other 'failed to comprehend why on earth I might want more than one bike at a time. The optimum number of bikes to have is four, no less.' Honda VFR owners, on the other hand, are clearly the most financially comfortable of the group (or are the most adept at finding examples that are roadworthy but ancient, and therefore cheap), suggesting that the optimum number of motorbikes must be at least six, according to one owner, to as many as nine according to another.

While a small proportion of riders manage perfectly well with only one bicycle or motorbike, the evidence for multiplicity is compelling. It may all be summarised neatly using the definition of resilience.

"N+1 Redundancy ensures maximum uptime and continuous availability."

However, UPS units and RAID arrays and duplex cabling are not normally associated with familial harmony. An additional criterion is the number of bicycles, s, at which your significant other will announce an intention to leave you if you bring home another bike. This threshold limit is normally defined as:

C = s - 1   [3]

which of course is the simplistic form of equation [2], and appears as a qualifer in Rule 12. The threshold limit appears in a reworked form in another suggested modification of the original equation, again on the Singletrack forum. Known as the CHB Bike Theory of Everything we account not only for the number, n, of bikes owned currently but the number, a, of bikes made redundant by the additional two, and the number, b, representing the difference between n and the number of bikes one's significant other thinks you have. Thus:

C = n + 2 - a - b   [4]

If a is found to be greater than 2, there is generally an error in the types of bicycle which one might aspire to own. Clearly C must be greater than n for all n, otherwise one might reasonably assume that there is a significant risk of a rupture forming in the bicycle-time continuum.

Jill Homer, a cyclist in Montana, owns five bicycles and, worryingly, says she doesn't actually enjoy it despite each having 'a particular function that I Can Not Live Without.' But she redeems her alarming thought of selling one, in order to make space for another, by modifying it instead. She sounds a lot like me, actually.

Perhaps at the root of 'one more bike' is our desire to try new things and competitive spirit encourages us to go faster, further, longer using less energy. You can ride bumpy, wet offroad trails using skinny road tyres on a frame with no mud clearance, but you might slide and fall off. You can ride on tarmac using tractor tyres and full suspension, but it'll be hard work. You can carry planks of wood and tow bicycles on almost any upright two wheeler, but a load bike or a trailer lets you carry more, more safely. You might be able to take your long wheelbase recumbent bike with four panniers into a hotel room, but a folding bike could make it a lot easier. Indeed, three was the Magic Number which De La Soul referred to in the 1990 single, and it looks good for cyclists and motorbikers too.

Finally we come to Bob Gannon, a photographer in Pennsylvania, who in a comment to Jill, elegantly summed up the growing awareness of Bicycle Acquisition Syndrome: 'I think you should resolve the issue by getting a larger apartment.'