NFL Player IQ by Position Played
I'm not a North American. I understand close to nothing about that funny type of football they play there. Yet I understood the graphic shown at the linked page with almost no conscious thinking.
Now to invent such clever visualisations for Poker Copilot. Suggestions appreciated.
The Poker Copilot Blog
Tracking the development of Poker Copilot, Mac OS X software for poker analysis and statistics.
Saturday, 19 July 2008
Great Example of Data Visualisation
Subscribe to:
Post Comments (Atom)
Poker Copilot Translation Project
Want to use Poker Copilot in your language? Then help with the Poker Copilot Translation Project.
About This Blog
In May 2008 I started working on Poker Copilot, initially as a product to help me with my own poker playing. Soon I joined a "30-day Challenge", where the participants each aimed to launch a software product in 30 days. As a result of this challenge, Poker Copilot version 1.0 was launched in July 2008.
This blog tracks the ongoing development of Poker Copilot. Who would find this blog interesting? People interested in 1-person software development, in Poker, or in both.
Contact me via email at steve at pokercopilot dot com.
This blog tracks the ongoing development of Poker Copilot. Who would find this blog interesting? People interested in 1-person software development, in Poker, or in both.
Contact me via email at steve at pokercopilot dot com.
Blog Archive
-
►
2012
(96)
-
►
May
(13)
- Lock Poker News
- What do MP, EP, CO mean in the positional stats?
- Adding a new Ongame Network poker room to Poker Co...
- Adding a new Merge Network poker room to Poker Cop...
- PokerStars on iPad and Poker Copilot
- Lock Poker and Poker Copilot
- Phil Gordon's Little Gold Book and Poker Copilot's...
- Poker Copilot 3.10 Now Available
- Crushing the Microstakes and Poker Copilot's HUD
- Early Access Version of the next Poker Copilot Upd...
- How many hands do I need for Poker Copilot's HUD t...
- Coming soon: PFR by position, Aggression by street...
- Comments on this Blog
-
►
April
(18)
- Important information from 2+2 : Two Plus Two Foru...
- Coming in the next update: VPIP by position
- A Nice Email…in French
- Poker Copilot Tip: Move HUD Panels without the Ful...
- Update for Winamax Users
- Poker Copilot 3.08 Now Available
- Poker Copilot Blog: 1000+ blog posts
- Poker Copilot and Zoom Poker: Update
- Poker Copilot 3.07 Now Available
- A poker player uses Poker Copilot to improve
- Coming in the next update: FALSE filters
- Mockup of Potential Poker Copilot Dashboard Concep...
- A Poker Copilot work-around for Merge Network idio...
- Poker Copilot App Updating
- Carbon Poker has its own HUD! (but only on Windows...
- Improving Poker Copilot's App Updating
- Poker Copilot 3.06 Now Available
- False Assumptions
-
►
March
(18)
- How NOT to make money as a software company
- Zoom Poker HUD Early Access Edition
- Changes to the Poker Copilot Hand Formatter
- Coming in the Next Update: Zoom Stats can be separ...
- Poker Copilot and Zoom Poker: played my first sess...
- The Absurdity of Modern Life: Facebook claims trad...
- Poker Copilot and Zoom Poker: Progress Update
- What character is that?
- PokerStars screen names
- Poker Copilot and Zoom Poker
- Poker Copilot 3.05 Now Available
- Ongame and Table Size
- Updated Poker Copilot User Guide
- Coming in the next Update: Unopened PFR Statistic
- Poker Copilot 3.04 Now Available
- Google Analytics Update: Almost a Good Graph
- Poker Copilot Website in Russian
- Downloading Older Poker Copilot Updates
-
►
May
(13)
-
▼
2008
(184)
-
▼
July
(14)
- Poker Copilot 1.1 - with HUD - Almost Ready
- First Tech Support Call
- Question for a Unix Command Line Tool Expert
- Big Download Boost
- Using Java for a Mac OS X Desktop App
- Mockup for Analysing Full Tilt Poker Tournaments
- Great Example of Data Visualisation
- Impressive Tool for Creating User Interface Mockup...
- Adding PokerStars Support
- My Definition of a Good Year...
- Poker Copilot: The Costs so far
- Running Poker Copilot While on Holiday
- Poker Copilot indexed in Google
- Changes Already
-
▼
July
(14)


1 comments:
Hmm. I personally think the visualisation is pretty poor. If you click through to the original article from the one you linked to, you'll see a couple of different versions, including one with the numeric scores. This diagram is actually far easier to read and understand than the supposed final diagram that's in the linked post.
The reason I think it's pretty crap is that values are represented by the size of the circles. People are hopeless at perceiving sizes this way, in part because our brains don't really follow the mathematical rule that says area of a circle doesn't grow linearly as the radius increases. That is, a circle that looks twice as big as another circle is, in fact, a lot bigger in area than a factor of two. In a chart trying to depict, in part, magnitude, choosing circles means that the physical representation of differences in magnitude is actually x:sqrt(x) rather than x:x.
The other is the problem of trying to compare one shape to another, surrounded by a whole lot of other shapes. Compare, for example the 'OG' to the 'TE': which one is bigger? To my eye, I wouldn't be able to tell them apart, and yet one is smaller than the other.
These are the same problems that pie charts have, and why they generally should not be used. Presumably, the 'story' of the chart is that there is a negative correlation between distance from the ball and test scores. In terms of readability, a table of numbers (with distance from the ball and test score as two columns) sorted on one of the columns would tell this story much more effectively. Alternatively a scatter plot showing clustering along an axis also would be better. Even the same diagram with numbers rather than position names works far more effectively, since most of the readers of the diagram would know the positions anyway.
Post a Comment