Hi, my name is Ryan Annese and I am a process engineer who developed a model explaining five types of mistakes that disrupt all processes. You might find it useful – it works for me every time! I wrote a book about it called The 5 Types Of Mistakes You Are Making and it’s available on Amazon.
To help track these mistakes, I created a new kind of bar chart called ‘Moses Charts’. They work by parting information to either side of a graph using an algorithm. This way, it’s way easier to spot patterns and trends. I put some Excel versions of the charts up on Dropbox and they are completely free.
I wanted to find some other uses for the charts and I think the results came out pretty good. Here is an explanation of the monthly Moses Chart using Football statistics from Pro-Football-Reference.com.
The Usual Bar Charts
If we wanted to see a comparison of the Win/Loss% for NFL teams in 2016, we might make a chart like this:
Here we can easily spot the lowest and highest W/L%, but it would be hard to pick out the 5th highest or 5th lowest. To make it easier, we might filter the data like this:
So we get a nice comparison here. But what if we wanted to see how each team did over the past 4 years? We might make a different chart for each year:
But this a little ridiculous so we might look for a chart in Excel that shows all 4 months at the same time like this:
Or we could use this:
NFL Win/Loss Data Using A Moses Chart
So instead I like using these Moses charts. It looks confusing at first but once you figure it out it’s real easy. The mountain is actually a trend line showing each team’s average W/L% over the past four years. The Patriots are at the top with 78%. Sloping down the left side are teams that show a trend of improving. Some have had steady increases in W/L%, such as the Falcons.
Sloping down the right side from the Patriots are those that show a decline in W/L%. There are some teams with steady decreases, like the 49ers.
The Redskins are sort of an outlier on the right side. They actually had good improvement in 2013-2015, but since their overall average is lower the algorithm pushes them over there.
You can also filter the chart by division or team to really focus on comparisons.
You can also send me over an email (firstname.lastname@example.org) if there are any other data sets you think would be interesting to check out!