Exercise

# EDA: mean differences between odd and even splits

To investigate the differences between odd and even splits, you first need to define a difference metric. In previous exercises, you investigated the *improvement* of moving from a low-numbered lane to a high-numbered lane, defining *f* = (*t _{a}* -

*t*) /

_{b}*t*. There, the

_{a}*t*in the denominator served as our reference time for improvement. Here, you are considering both improvement and decline in performance depending on the direction of swimming, so you want the reference to be an average. So, we will define the

_{a}**fractional difference**as

*f*= 2(

*t*-

_{a}*t*) / (

_{b}*t*+

_{a}*t*).

_{b}Your task here is to plot the mean fractional difference between odd and even splits versus lane number. I have already calculated the mean fractional differences for the 2013 and 2015 Worlds for you, and they are stored in `f_13`

and `f_15`

. The corresponding lane numbers are in the array `lanes`

.

Instructions

**100 XP**

- Plot
`f_13`

versus`lanes`

using keyword arguments`marker='.'`

,`markersize=12`

, and`linestyle='none'`

. - Do the same for
`f_15`

versus`lanes`

. - Label the x-axis
`'lane'`

, y-axis`'frac. diff. (odd - even)'`

, and show it.