Run or Walk in the Rain?

2010 November 30
by Daniel Lakeland

Mythbusters asked the question of whether it's better to run or walk in the rain, assuming you're trying to avoid getting soaked. They did it one way, and then again a different way, and I think they finally decided it was better to run. My friend Dennis asked me this question, he lives in Portland, where this is a nontrivial everyday problem. After thinking about it while out for a walk with the baby, I came up with the following basic model.

Suppose that the density of rain in the air is \rho and the rain velocity is v at an angle \theta from the vertical, with positive theta meaning the wind is blowing into your face. In a time dt the amount of rain that falls on your head is A_v \rho v \cos(\theta) dt where A_v is the cross sectional target area you present vertically. The amount of rain that falls on your front is A_h \rho (v \sin(\theta) + v_h) dt where v_h is your horizontal running velocity. Suppose everything is constant during your dash for the front door so that gusting and your acceleration and soforth are ignored, then rather than integrating with time we can simply multiply by the time taken. The time it takes you is L/v_h where L is the length to your front door. Then the total water falling on you during your dash is:

\rho v \frac{L}{v_h} (A_v \cos(\theta) + A_h(\sin(\theta)+v_h/v))

This expression has a term that is constant with v_h and independent of v, and two terms that go like 1/v_h. This reflects the fact that we need to cover the distance L regardless of how fast we do it (intercepting at least the rain that is in front of us right now). The term in \cos(\theta) is always positive since \cos is an even function and \theta is between -\pi/2 and \pi/2, and the term in \sin(\theta) can be positive or negative depending on whether we're running into the rain or with the rain.

Is there an optimum running velocity? Let's take the derivative with respect to v_h and look for where it goes to zero (a local minimum or maximum).

\frac{dW}{dv_h} = - \rho v \frac{L}{v_h^2}(A_h \sin(\theta) + A_v \cos(\theta))

This goes to zero at v_h \rightarrow \infty but does it go to zero anywhere else? If we multiply through by \frac{-v_h^2}{\rho v L} we get (A_h \sin(\theta) + A_v \cos(\theta)) = 0 which occurs at A_h \sin(\theta) = - A_v \cos(\theta). This is true when \theta = -\tan^{-1}(A_v/A_h). This suggests that at this certain angle, increasing the speed of running reduces the rain falling on your head but increases the rain falling on your front by an equal amount.

So unless you happen to be in this special wind condition, run hard Dennis!

3 Responses leave one →
  1. Adam permalink
    November 23, 2014

    Hi Daniel! I love the idea of modelling this and I think you've described it in a way that's easy to follow. I just wanted to ask about theta: why is theta between positive and negative pi/2?

    • Daniel Lakeland
      November 23, 2014

      theta is between +- pi/2 because it's measured relative to the vertical coordinate axis, and the rain is falling down. For brief periods rain can fall up because of wind gusts or whatever, but the average velocity has to have a downward component.

      • Adam permalink
        November 23, 2014

        Right, that makes sense! Thank you!

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS