1 Reply Latest reply on Dec 2, 2018 10:27 AM by Paul Wachtler

    Need quick feedback for my Udacity Baseball Project

    Mark Trapani

      Hi there,


      Just finalized my udacity project using a data set containing 1,157 baseball players including their handedness (right or left handed), height (in inches), weight (in pounds), batting average, and home runs. Would appreciate some advice on how to improve my visualizations and whether findings are communicated clearly.


      Here's the link to my project story: https://public.tableau.com/views/UdacityBaseballPlayers/Story1?:embed=y&:display_count=yes&publish=yes


      Thanks in advance

        • 1. Re: Need quick feedback for my Udacity Baseball Project
          Paul Wachtler

          Hi Mark,


          It's a very comprehensive report.  As I'm not sure of the intent of the project or how it will be graded, I'll focus my feedback on the visualization aspect.


          Page 2 (first one with charts):

          - It's hard to decipher how large percentages are in a pie chart - are there twice as many right handers as left, three times as many?  I'd recommend turning this into a bar chart, just like your home run chart.  Make sure to label your marks to so it's easier to tell what the % is for each bar.

          - If you want to color code the three categories (you don't have to), then you should do it consistently across your charts.  Don't have one chart with color for each category and one that uses the same color for each category.

          - Sort your bar charts - I recommend using a descending sort order (highest values at the left)


          Page 3

          - does your audience understand how to read a box plot?  If not, you may want to include descriptive text to explain them (explain the median, quartiles, what the outlier values outside of the box plot mean, etc)

          - recommend sorting all of these in descending order too

          - I'm not sure your home run vs batting average chart really shows any differences between the three.  If there is something specific you want to point out here, do it with an annotation or with text.

          - Your batting average axis needs an extra 0 at the end of each number.  A .10 average should read .100 (that's how batting averages are written - out of .1000)


          Page 4

          - these two charts are confusing.  You're putting three different variables on each chart.  It's difficult to try to read the relation between weight, batting average, and home runs all at the same time - especially when the axis for home runs and batting average are completely different.

          - you should break this into two dashboards, or at least two charts for each one.  You should have weight affect on batting average, weight affect on home runs, height affect on batting average, and height affect on home runs as 4 separate charts


          Page 5 (last page)

          - don't encode handedness with both color and shape - choose one.  Using two makes the viewer have to think twice about what the one single point is showing.

          - suggest making this much wider so it spreads out some of the cluttered shapes.  If you use a circle shape, you can set a border (on the color menu) around each one which will distinguish each circle a little.  You could also lower the opacity (also on the color menu) a little so you can see more of the shapes.

          - your summary point of batting average correlating with home runs has nothing to do with handedness.  You don't need to split this out by handedness if you only want to make that point on this page.



          Overall you've clearly done a lot of work on this and put a lot of time into it.  My recommendations above are purely that, recommendations.  This is your project so take or leave any of my recommendations that you would like.  I'd highly recommend sharing work that you'd like feedback on on Twitter in the future as there is a vibrant Tableau and Data Viz community there that are more than happy to help as well.


          All the best,