Assumptions and calculations explained.

This website is to show the average times for runners to complete a road race and here are some notes about how this information is collected and calculated, plus the assumptions I have made.

- Average Runner. This is important, there are only 2 people in the world who are Olympic champions in any race distance (one man and one woman), though there are quite a few who have been champions. There are millions of people in the world who enjoy going out for a quick run around their local area. I am one of these people, we are average runners, some are faster than others, some are slower but the one thing that makes us all equal is that we run because we enjoy it. World records are great because they tell you how fast those 2 people can run, but what about the rest of us? Anyway, for the purpose of this website the average runner is you or me, and the times I have looked at are for races that you or I could enter, pay our money, turn up and run. So these are the race results I have looked at

- Average Race. I should explain this one too. Average race is not any particular race, it is a mix of all the races, so hilly races, flat races, fast or slow races all combined to give an average race course. Of course, if you run on a course that is faster or slower than average (and they all fall either side of the average really) then your time won't be for a fastest race really. The things that are important is that the distance is the same (mostly, road races by their nature and geography are never exactly the stated distance), and the other thing is that they are a road race.

No one is excluded from the results. If someone has run in an average race, they can be counted here.
Big Races. Some races are excluded however and these are the reasons
- Each of these races with thousands of runners takes ages to analyse... I don't have the hours in my life to do that
- The results pages are long or are not easy to get a listing, making this harder again
- A large race has a huge effect on the results. 35,000 runners on one race and 35,000 runners in all the others. If the big race has unusual conditions then this affects the result greatly. A lot of smaller races average out the race routes and conditions to give overall better averages
- Big races are once a year or every other year event for average runners, not a race that we run every week, so are not really representing an average race for most people
- If I get time, I might make up separate pages for big race results.
Parkrun. Weekly 5km timed runs, not races as such. However these are a good place to analyse running times Off road races (trail, cross country, hills....). Generally. Each of these are unique courses and not directly comparable to road races or even each other Track races. Not a race for the average runner if only 8 or so people take part

The following information is counted from the race result:
Finishing minute (seconds are not counted) Where known the gender is counted where known if the runner is in a running club is counted

Times. For each race, the number of runners of each gender that finished in each minute is counted. Using minutes gives better results (the count is easier!). With a list of how many runners finished in each minute for the races being analysed, then I can work out things like the all important Average finishing time. Splitting the count up between men and women lets me work out the average for men and women, and adding the womens and mens results together gives me the overall average. Dead easy
Ratios. Having split the runners into men and women, it is easy to find out the proportions of who has run. Sum is simple, total number in each gender divided by the total number of runners and multiply by 100 to give a percent for each gender
Clubs. Same as for the gender ratio, club ratio is the same sum, however this isn't an exact count. Man or woman is easy to figure out, but it is hard to know if a runner is actually a member of a running club. This ratio is for indication only however I'll have the assumption that the proportion of people who declare to be a member of a running club remains constant with each distance (so if half of running club members say they are in a club, and half don't for a 5km race, then I assume that half of marathon runners will be the same) Number of runners. Do I need to say that I just add up how many runners I have analysed. Same for races Percents. Pretty much the same as a ratio really - add up all the bits to make into a percent, and decide by the total.

So they are the calculations - pretty simple maths, however the time consuming thing is counting the finishers!!

I mentioned above that Parkrun gives some insights into the world of running. This is because the same courses are run every Saturday by a similar number of people on each course and the geographical spread of these runners is all over the UK. So they are quite good to look at other tings to do with running. To do this I have made these assumptions:
No area of the country has a faster mix runners than another area Given enough runs and runners this eliminates variations such as weather and local events that would affect a single result Even if the individual runner change in a parkrun or any race, the profile of runners stays pretty constant - the same numbers of fast, middle and slow Club Runners tend to be faster runners and runs with large numbers of club runners will produce faster than average results Men tend to be faster runners and runs with large numbers of male runners will produce faster than average results A run with a large number of runners will have a greater spread of finishing times, I expect to find faster finishers but also slower finishers at these, just because of numbers running Up to a point, new runs will be slower while the word spreads about them. After a certain point the averages stabilise and only the extremes change really. I am stopping adding up the results after 200 runs Close areas of the country share a proportion of runners, so some runners from Strathclyde will run in Edinburgh and Glasgow for example. This makes some courses better to compare since the runners are the same. Other courses virtually never share the same runners - such As Edinburgh sharing with Exeter because of distances.