Assessing a Cycle Path Scientifically
Ever since I’ve been riding on off-road paths I’ve noticed that some paths are good, some are bad, some are excellent and some are awful. Most riders generally agree on the quality of a path, but it’s hard to document and pass on to others. Their idea of slow or fast may be different to mine. Recording the average speed attained on a path can have little meaning to those that ride at a different speed. The science student in me has yearned to turn these subjective assessments into something objective — to ‘measure’ the quality of a cycle path or route.
Measuring interruptions to momentum
Momentum plays a huge role in the ‘amenity’ of a cycle route, or to be more specific conservation of momentum. Modern paths sometimes seem to have an annoying obstacle half-way down a slope causing the cyclist to brake hard, losing all that hard won potential energy from the previous climb. For example, we recently found the path alongside Hoxton Park Rd to have a huge number of frustrating dips and gutter crossings that made steady cruising impossible and reduced our overall speed quite a lot.
After riding paths of varying quality with a small group recently, the means of measuring a path became apparent when one of the group said of a perfect path, “You get back 100% of your effort in the trip time”. Stated simply, the perfect path where a rider can cruise at their normal cruising speed without any impediment gets a rating of 100%. As obstacles mount, so the path rating drops. Eventually there are enough obstacles to affect the mental outlook of the frustrated cyclist and the speed drops even further. We found our highest speeds were on the most attractive tracks, because we felt we would be rewarded for our effort.
Networking
Now rating a path from, say, Parramatta to Liverpool, is only useful for someone wanting to travel that entire route. If you only want to travel part of this route, and branch off somewhere else, only the part travelled is of use. So a large ‘system’ must be networked, and each leg of the network measured or rated; and then any rider can check the ratings on the parts they wish to travel.
As a commuting adult that’s not going to the next Olympics I feel that 20 kph is a decent cruising speed. I get to this conclusion by noticing that on the 60 km ride last week the group averaged 18 kph over a large variety of path qualities. It’s true that on a really wonderful path like the Lower Prospect Canal the group can average more than 20 kph but rough or narrow paths result in lower average speeds. Also, since 20 kph is 3 minutes per kilometre it makes the maths easy.
So rating a segment of path becomes as simple as riding the path with a cruising speed of 20 kph, no more and no less, and measuring the total time taken for the route leg as well as the distance in kilometres; and then simply start with 300 and multiply by the distance in km and divide by the trip time in minutes, to get the rating out of 100%. You’ll need a stop watch on your wrist; and a trip computer that measures to tenths of a km, properly calibrated of course, by riding a kilometre at 20 kph and see if takes exactly 3 minutes. Alternatively you can use bikely.com to measure the exact distance for you. Of course all road rules must be followed, and where there are road crossings a stop and wait should be made as road traffic variations could influence trip times. I call the result the trip rating of the path, ‘TR’ for short, and rating at 20 kph could be called TR20.
Here’s an example: it takes me 7 min 15 sec to ride a 2.2 km piece of path. So 300 times 2.2 equals 660. Divide by 7.25 and the result is 91%. It’s that easy!
So let’s go for a ride and try it out. I’m using a similar route to CAMWEST’s Bay to Mountains route to get a variety of paths including a decent leg of the T-Way, rail trail and Canal paths. We start from Canley Vale and head west to Bossley Park along Orphan School Creek (OSC), then north through Wetherill Park to Prospect, then across the Canal path to Guildford and down the rail trail back to our start point.
Mapping the data
Adding numeric data to a map is difficult, so the idea might be to colour the paths depending on the rating. Various colour ratings could be used, using the colours of the rainbow.
| TR20 rating | Mapping Colour |
|---|---|
| Less than 75% | Black (Infra Red = inferior to Red!) |
| 76%–80% | Red |
| 81%–85% | Orange |
| 86%–90% | Yellow |
| 91%–95% | Green |
| 96%–100% | Purple |
This gives new meaning to the phrase “Purple Patch.”
Here’s the data that resulted from my ‘test ride’.
| Trip Leg | Elapsed Time (min:sec) | Distance (km) | Rating | Group | Comments |
|---|---|---|---|---|---|
| Canley Vale to Avoca Rd | 11:33 | 3.5 km | 91% | Green | Two Xings, narrow underpasses |
| Avoca Rd to OSC/T-Way | 8:49 | 2.7 km | 92% | Green | Three road Xings, one VV busy |
| OSC to Horsely Dve | 6:16 | 1.9 km | 91% | Green | Three traffic lights but responsive |
| Horsely Dve to Prospect Ck | 14:32 | 3.7 km | 81% | Orange | Six traffic lights and a ped Xing |
| Prospect Ck to Canal path join | 1:47 / 2:05 | 0.55 km | 84% / 79% | Orange/Red | Hill and tight turns Stop Sign at busway |
| Canal path join to Pemulway exit | 2:32 | 0.85 km | 100% | Purple | Perfect |
| Pemulway exit to Cumberland Hwy | 10:37 | 3.5 km | 99% | Purple | Bollard |
| Cumberland Hwy to Railway line, Guildford | 11:10 | 3.5 km | 94% | Green | Multiple road Xings |
| Railway bridge, Guildford to Prospect Ck, Fairfield | 7:15 | 2.3 km | 95% | Green | Road Xing and Stop sign |
| Prospect Ck, Fairfield to Canley Vale | 8:49 | 2.8 km | 95% | Green | Some on road, 2 road Xings, ped Xing, narrow rough footpath |
| Total Trip | 82:30 | 25.25 km | 92% | Green | Variety of paths |
This is, of course, only a small part of the wonderful network of paths in Western Sydney. If you ride a segment of the network, put your stop watch on, cruise at 20 kph and let us know the trip time so we can put it on our map.
Western Sydney Cycling Network Map
I’ve redrawn the Western Sydney Cycling Network map with these and many other paths coloured in the appropriate colour. Here it is as a PDF file.
Many of these paths have been estimated from riding experiences (and google mapping) rather than strictly timed with a watch. If you ride these paths with a stopwatch and disagree with the rating please let us know.
As you can see, some paths score well, but others are clearly lacking. For example, at Avoca Rd I lost 6% on a single bad road crossing. This method provides an objective measure of the route’s attractiveness and amenity to commuter and recreational cyclists alike. Of course it’s fairly straightforward now to add up all of the segments of interest in a trip and estimate the total trip time with some accuracy. It’s interesting to note that the exact rating of a path may be different when travelling in the opposite direction due to slopes, sight lines and perception of obstacles, however except in unusual places the ratings in both directions should be within the same colour band. I found the section through Wetherill Park that rated 81% to be just about the minimum limit at which I would accept a path to be; at which point as an adult I would take to the road in frustration. The traffic light phasing in this area doesn’t do the cyclist any favours either. Unfortunately children and timid riders don’t have the option of taking to the road, and usually simply don’t ride a bike in those conditions. The section over Moonrise between Cowpasture Rd and the M7 is red due to the excessive slope, you need to be Cadel Evans to to 20 kph over that hill, and you can’t recoup the lost time on the downhill due to the speed humps and general lack of safety at such high speeds; and a normal person needs to recover across the top and can’t spin along very fast.
More questions
Some questions still remain however. Ascending at 20 kph can be difficult to maintain, so on long or steep slopes the rating could be different in the opposite direction. If there is an obstacle, but the path opens up later, should the test rider speed up to recoup the lost time? And up to what speed? My idea is that a reasonable attempt should be made to recoup any time lost while moving but only by using a slight extra effort; but that stationary time is lost forever. Bike trip computers make this adjustment simple, by aiming the average speed indicator to the desired speed (20 kph) rolling losses can be made up IF the path design allows. For design purposes, the rating should be calculated in both directions if the slope is significant enough. For the short section up the hill beside Hylands Road I simply couldn’t make the climb fast enough, so the uphill rating is lower than the downhill one. This slope is way above the recommended wheelchair grade; and the result reflects that objectively. As a guide, the carefully designed grades of the T-Way paths can be ascended by a middle-aged adult at 20 kph, just! Rating in both directions results in an objective indication of the grades involved.
It’s also quite instructive to rank the path based on riding it without a stopwatch, and then compare this subjective ranking with the timed and calculated result. Sometimes paths can ‘seem’ worse than they really are, and sometimes our ‘gut feel’ is spot on! The final part of this dream is to get engineers and designers to rate their paths on the drawing board, and then for Austroads to incorporate a minimum standard. Obviously paths of 100% rating are not always possible, but equally paths with rating less than somewhere near 80% should not be allowed as they are simply a waste of money as even the slowest and most timid riders would find them frustrating. My making the ‘transport rating’ calculation before construction, engineers, planners and politicians have the option to reassess their plans if the rating is lower than they find acceptable. It seems fairly easy to build a path rating just above 90% to get it into the Green band, but getting improvement up into the Purple band is increasingly difficult as this invariably means giving cyclists right of way over motorists or building expensive overpasses.
