Today was the 127th running of the Boston Marathon, one of the world’s most prestigious footraces. The race originally started out at a little over 24 miles in length, based on the distance the Greek runner Phidippides allegedly traveled between Marathon, a town in Greece, and Athens to report on a Greek victory at the Battle of Marathon. Legend has it that he died shortly after arrival, though I would take that with a grain of salt as couriers were commonplace in those days and there is no reason to expect he did not take any breaks.

The modern distance of 26.2 miles was instituted after the first few Boston Marathons had already been run. This forced the Boston Athletic Association to push the start of the race from Ashland into further west into Hopkinton, where it is today.

Poor Ashland. Not only does the starting line itself leave town at the first opportunity, Ashland tends to only make the news when people are running into it and are trying to get out as fast as they can without stopping. The number of athletes barreling down Route 135 is actually significantly greater than the population of the town as of 2023.

Interestingly, the Boston course is actually nonstandard and cannot be used for Olympic events. From what I recall, official Olympic races have a maximum elevation change along the course and are run in loops to prevent times from being skewed by topography and winds constantly heading into the runners’ faces. At one point the Boston race had the fastest time due to the steep descent out of Hopkinton. However, this time was not eligible for the record.

The prestige of the race lures runners from all over the world. Many of the recent winners are members of a tribe in Africa who have naturally been gifted with excellent endurance and stamina. Generally if someone from Kenya wins a race he or she belongs to this community.

The 2023 men’s race was won by the 2022 winner, Evans Chebet, in 2:05:54. Hellen Obiri won the women’s race in 2:21:38. Chebet’s time corresponds to an average speed of approximately 12.5 mph. Obiri’s is about 11.4 mph.

This suggested an interesting question, which I posted on the Brookline Townwide Discussion Facebook group. Suppose the following three individuals tackle the marathon route under the same weather conditions (for 2023, rainy and in the low 50’s).

1. Chebet, as he did today. 2 hours, 6 minutes.
2. An average American on a standard bicycle with no traffic lights or traffic to deal with.
3. Someone in a vehicle driving during rush hour. Traffic is unusually bad in this case, perhaps 95th percentile when it comes to drive time.

The bike and car turn down Gloucester Street instead of Hereford due to the fact that Hereford is one way. Otherwise, the courses and distances are identical.

The question is: who would win?

The post triggered a storm of debate on the Facebook forum. One would have naturally assumed that the cyclist or the driver would win. However, there are several caveats.

I used to bike to work from time to time when I worked in downtown Boston, and my route would follow Beacon Street from Coolidge Corner to the Public Garden via Kenmore Square. I would routinely hit 20 mph on the hill heading inbound from Coolidge Corner. On flat terrain, such as in the island in the middle of Commonwealth Avenue, I would average about 13 or 14 mph. This would imply that a cyclist has a good chance of beating the runner, who is averaging 12.5 mph.

However, read the challenge carefully. Contestant #2 is explicitly stated to be an average American. We are not dealing with Lance Armstrong here. In particular, the rider is not trained to do long distances, if he even bikes at all. Certainly, he can reach speeds of 14 mph or higher, especially at the beginning of the course. But can he make it through all 26.2 miles at that pace?

The course trends downhill, which will help the cyclist. Unfortunately, there is a minor snag in Newton known as Heartbreak. Having grown up in Newton, I would routinely get on my bike as a child, turn west onto Commonwealth Avenue from Hobart Road (near the top of Heartbreak Hill). I would head out to the Marriott near 128, find myself horrified at the possibility of fast moving cars trying to merge onto the highway, turn around, and come back. Trust me when I tell you that you do not want to bike up Heartbreak Hill after a 9 mile trip when you are a child. Now replace the child with a 40-year-old adult who hasn’t biked for a while, let him go for 20 miles, and then have him hit Heartbreak.

I don’t think so either. As it was, I thoroughly disliked the trip back up the hill into Coolidge Corner after work. And that was a 4 mile trip. I could make it, but my speed dropped to 6-7 mph. Suffice it to say that you are not beating Chebet if half of the course is done at 20 mph and half is at 6. The average speed in this case is a little over 9 mph.

Of course, a cyclist can draft behind a vehicle to let him coast a little (a tactic which was actually mentioned during the race today). Although this works in principle, in practice the vehicle will have to be traveling the same speed you are. In this case, the cyclist has no traffic he can coast behind.

So we can basically write off the cyclist. To be fair, he will beat a typical jogger (6-7 mph) but he will not beat Chebet.

Now we come to the car. Assuming you are obeying the speed limit on course, much of the course can be driven at 40 mph or faster. This puts the car at an immediate advantage. The car also will not tire, like the cyclist will. (Chebet tires as well, but as a trained athlete he knows how to pace himself so it does not affect him as much). Cars can also maintain the same speed going up and down hills.

The problems start cropping up when we include traffic…and worse still, traffic lights. As anyone who has tried to drive into Boston during rush hour knows, things can get slow in a hurry. In theory the bike has to observe the traffic lights as well but more often than not they don’t.

We can determine the speed of a car in downtown using Google Maps. I live in Coolidge Corner, so I asked it to estimate how long it would take to get from the Newton Main Library (Walnut Street and Commonwealth Avenue, on the route) to my apartment. The app said 4.7 miles in 18 minutes as of 6:26 PM on Patriot’s Day (assuming the road is now open to traffic). That’s 15.7 mph.

Given that this section of the course has among the most traffic lights and traffic in general, this seems to hint that the car would beat Chebet. I am not sure what exactly the roads look like in Ashland and Framingham, but from what I saw from the coverage they were faster and had fewer lights. If the car beats the runner here, it would beat the runner everywhere else on the course.

So odds appear to be in your favor if you choose the car. But would you be willing to put money on it? You’ve driven these roads before and know the answer: probably not.

A trip which expected to last 18 minutes could luck out and be 15 minutes. However…all you need is one broken down truck and it can take 30. The range of possible outcomes is vast. And given that the question assumes worse than normal traffic…

Take for example one of the posts someone brought up in response to the Facebook question. An app was asked to estimate the amount of time it would take to get from Hopkinton to Boston. The app responded with two routes, the Pike (I-90) and Route 9. The estimates for the Pike ranged from 50 minutes to 1:30. Route 9 was worse, ranging from 55 minutes to 2:10. Although neither of these is the marathon route itself, it seems to hint that the worst case traffic will cause trips to take twice as long as the best case traffic.

For argument’s sake, assume the estimate provided for the Newton Main Library to Coolidge Corner is around average for this trip. Given that, I would suspect that the range of possible values for the trip will go from 15 minutes to 30 minutes (the average will likely be closer to the lower value than the higher value due to the long tail at the high end). 4.7 miles in 15 minutes will beat Chebet and the cyclist easily, as we have seen. But 4.7 miles in 30 minutes? That’s 9.8 mph. That is not going to be enough to beat Chebet or possibly even the cyclist on a good day.

Note that I once DID try to drive the marathon route. Although I do not remember the exact time, I think it took something like 1:30 or 1:40. Assuming that this time is typical, we are only one or two standard deviations away from losing to Chebet.

There is one transportation option however, by the way, which we have not considered. The T — the Boston subway system.

Most of the light rail (Green Line) lines paralleling the marathon route are street lines, which means that they have to deal with the same traffic light and traffic problems as the cars. Furthermore, their performance is hampered further by the fact that they have to stop from time to time to pick up passengers. Studies have reported that these lines average between 7 and 8 mph, losing to both Chebet and the cyclist.

However, there is one exception: the D line. This line, with no traffic to deal with and fewer stops, averages almost 19 mph. This brings up the possibility of beating Chebet by boarding the T at Riverside (near where the course goes over 128), taking the T to Reservoir, and rejoining the race at Cleveland Circle. This will almost certainly work, though clearly it doesn’t count as a viable option as the participant is cheating in this case.

The reason I considered it was because I had wondered whether Rosie Ruiz had won her infamous race by jumping ahead on the T and coming in first that way (as it was, everyone was very surprised when she appeared out of nowhere). However, this appears to not have been the case. What appears to have happened is that Ruiz jumped in less than 10 miles from the end in Boston and looked remarkably refreshed after completing such a grueling race. The subway incident did in fact occur, but it happened during the New York Marathon. Since she used her time during the New York race to qualify for Boston, her disqualification from New York would have disqualified her from running in Boston even if she had run Boston normally.

In short, it seems like the race will end in a close match between Chebet and the driver unless the cyclist is already a trained athlete and is not telling us. Only way to tell for sure is to try it for yourself.

The photo is the obligatory marathon photo. Look at all those people. Bonus points if you can tell where in the race it is.

Trust me. You won’t get it.