I think it's a pretty impresive Viz, and you had to pass a lot of time trying to collect the data.
Not knowing anything about football it's quite complicated for me to enter in this topic. I think that depending on the public you target, you have to adapt it by adding some legend or title to help the navigation.
But if it targets a person knowing the subject, it can be easier.
I hope my comments, coming from a non football fan helps you.
You’re absolutely right – I’d initially planned to have the top and bottom halves of the dashboard on separate sheets but prior experience has shown me that only a small proportion of viewers look at anything but the first sheet so decided to merge them together.
The player selection parameters as they stand now have no effect on the scatter plot (other than they trigger labels to appear under that player’s mark if it exist in the scatter.
What would you like to see happen in the scatter plot based on player parameter selections?
Thanks for the feedback, Michael – much appreciated.
Yeah, I’m not a huge fan of the default legends for continuous fields in Tableau so I tend to design my own simple variations in Photoshop ☺
I think you’re absolutely right. I initially built the dashboard with FPL players in mind, who shouldn’t have any trouble understanding what everything in the viz is trying to show them. However I can easily see how someone not familiar with football/FPL may get lost trying to use the dashboard.
I’m going to add a few information hover sheets around different areas of the dashboard to explain what each section is and what it’s intention is.
Thanks for the feedback ☺
I think scatter plot should talk about players selected in parameters.
If not analysis become a little bit confuse.
Despite this little thing, it's very cool.
If the scatter plot were to only show the 2 players selected by the parameters it would defeat the entire purpose of the scatter plot.
The scatter plot is there to compare all players based on the stats selected so that the user can identify which players are the best as certain things – so they can then decide which players they want to compare in the tables at the bottom of the dashboard.
If I were to limit the scatter to only show the 2 players selected the scatter would only let the user know which is the better player of those 2 – ignoring the other possible 400 players that are better than both of them.
I became confused because parameters started with the same players in scatter plot.
When you see, you expect the scatter plot works with parameter selection.
Ahh right – I’ll include the definition for labels in the scatter plot in an information hover then.
Basically so the scatter isn’t over-run with labels for every player, I have a calculated field setup to display labels on certain players.
The player with the highest ownership (Ibrahimovich)
The most expensive player (Aguero)
The player with the highest x axis value
The player with the highest y axis value
Players selected by parameters.
I guess it was just a poor choice on my behalf to upload the dashboard with Ibrahimovich and Aguero selected by parameters when they have labels by default because of their value and ownership.
You’ll notice in your screenshot that when you changed the parameter one of the players (Kolarov) you selected did have a label show up in the scatter – it’s just hidden beneath lots of other marks. Barnes doesn’t get a label as he’s not in the scatter having not played any games and not meeting the at least 200 minutes filter.
I’ll definitely add some info hovers to clear some of this stuff up.
Hope that helps.
image003.png 334.4 KB
Could you introduce a second element in the scatter plot, one that would highlight the two selected players?
I think this is what Luciano Vasconcelos was getting at.
Your scatter plot would retain its functionality (in showing all players), yet still highlight the selected players.
You could do this through a couple of ways:
- Include labels only of the two highlighted players
- Create "cross hairs" (perhaps using reference lines) to target the two players. You could also use a dual axia and superimpose a cross hair image.
- Using a dual axis, change the outline color of the circle
Again, thank you for sharing your viz! --Michael
I suggest you let information only to selected players and maybe worst and best on selected axis.
Your scatter plot remains the same but viewers will be able to find players selected on it.
Yeah, now I understand what Luciano was getting at I’ll find a way to highlight the parameter selected players.
I’ll probably do it via labels but apply a blue label to the left player and an orange label to the right player.
Hopefully that should clear up any doubt ☺
And one more suggestion!
When reviewing the scatter plot, could you make this an interactive way of choosing players?
You could do this as a Dashboard Menu option:
Choose This Player As My #1 (or Make this my Blue Player)
Choose This Player As My #2 (or Make this my Orange Player)
I think that would make the hunt & peck for a user's name (400!!!) much easier.