In 2023 I blew it. Wrong topic.

In 2004 I got it right. My Viz made the list of the ~200 finalists.

In 2025, how will I do?

My second, proper, Iron Viz entry is in. The title “What’s your Tipple”. All focused on the delicious world of cocktails and taking the viewer on a journey to find one, or more, for them.

What did I learn, too much to write here. I got a connected Beeswarm Chart to work, which means I got Beeswarm charts to work. I got the slick but truly tedious dashboard navigation to work – so one dashboard flips through charts, rather than having to change to different tabs. Tedious that is, hurt do my hands from all the mouse work.

I also learned a ton of charts as preparation and trials, the helpfulness and limits of ChatGPT in data. And I played with my vizology knowledge – the art side and the science side.

For the record, the viz. More learning to follow when my muscles recover.

It’s the start of the Formula One season. Lots of races for sports fans to enjoy, lots of potential data stories for those who like to viz out the F1 seasons. Such good data to play with.

So began the season and a comment I read from a driver about how “the cars are getting closer and closer in terms of time” stuck with me. After the qualifying session, I checked. And indeed, every car is inside a 2 second gap from pole position to the last. Pretty short times if you check out the link.

https://www.bbc.com/sport/formula1/2023/bahrain-grand-prix/results/qualifying

But what does that look like? So I though to plot it, use that length thing people are so good at to see if the gaps can been seen/understood better. Be fun, be easy since not much data and Tableau makes a Gantt chart well. Ohhhh was I wrong.

Seems that date functions in Google sheets, Tableau and others don’t handle microseconds well. Seconds yes, microseconds no. So some searching later, some other comments how bad its handled and I’m of with math. Ended up converting the time to numbers to handle the microseconds – so multiplied the seconds element by 1000 and worked with numbers, keeping in mind that the 1000, 2000 marks meant it was a new second. So math done, time to see it.

Got a reasonable view out in the end. The length shows the top 10 mostly grouped up together, some very very close drivers in terms of time, the inevitable large gap to Max Verstappen and clusters for the other two qualifying sessions. Added the previous year as well (same need to manually crunch, took less time since I know what to do) and you can see on the same track, for 20 drivers in 10 teams, they are indeed qualifying closer and closer to each other.

The remarkable thing, the whole top visual covers a span of under 2 seconds. The slowest driver in the slowest car is a less than 2 seconds slower over a lap (90 seconds plus) than the fastest. 2 seconds at nearly 200 km/hr goes by fast.

What next? Well, enjoy the F1 season. I might try to covert the time to distance to show in a different way just how close these cars are performance wise. And with lots of races, let’s hope for some interesting events in the season so each race generates some interesting data viz opportunities.

So, my tip for you. If you are going to work with short times, microseconds, be prepared to convert the minutes and seconds from time to integer numbers.

Chinese New Year is here. A new year, a new mythological character. Why not make it a year to learn and read a book. Or six Six you say? Well it’s only two per month, two per 60 days, that will help you learn and make you a better vizologist. You can do it, they can even be online digital books from the library after all.

The Big Book of Dashboards, by Steve Wexler, Jeffrey Schaffer and Andy Cotgreave

The best practical book on my list that will help make your dashboards better. With a great opening segment on the theory of visual information. That section alone you will come back to. Then some solid before and after dashboards and the why/how they work for reference when you are building or improving on your work. And at the end a well referenced 4 page guide to chart types. Love this book, it’s one of my mot shared.

Learning to see Data, by Ben Jones

I’ve built a few great samples to showcase the value of visuals from this book, to great effect. This for me is one of those fill in a few gaps, refine what we’ve worked with for a while book. Its well suited to someone beginning, and a good on to come back to every once in a in while for tips.

How Charts Lie, by Alberto Cairo

This book was an eye opener. With a start on the mechanics and theory of a good chart. It reads a bit dry but is one of those items that lets you see better where a good chart is used, and where a bad one is. Then into the details on perspective, and how commonly viewed charts can be so wrong – whether maps, lines or weather. A good read, and some great online viewing as well.

Practical Tableau: 100 Tips, Tutorials and Strategies by Ryan Sleeper

A lot of how tos. How to make a waterfall chart, a Pareto and more. For me, it’s how to get the most of Tableau, certainly in 2018 when this came out. If you use a different tool, look for a good set of how tos for that to up your visuals.

Storytelling with Data, by Cole Nussbaum Knaflic

Work or present charts and dashboards enough and you realize you need more than a cool, well constructed chart to make your point stick. You need the information to make a story that sticks with your audience, one they relay again or even better, take the action you recommend. How to do that quickly, check out this classic on Storytelling with Data. How to package that story onto just the space of a chart, with the right focus on data, the right use of text and annotation a truly valuable lesson from this book.

Effective Data Storytelling, Brent Dykes

Taking it up a level, this book if for you to get into a solid story across a whole dashboard, across a prevention deck or lie presentation. There are some solidly practical ways to improve charts. Two well used tips for me – the charts you use to find an insight are not the same you use to present the insight. How many times I’d I get that wrong in the past. The other, how to setup a story in 4 parts. I won’t give that away here, happy reading to gain that golden nugget.

Full disclosure, I have read all these books. I have them in physical book form. I refer to them often and I share the widely with people I work with who want to learn more about Data Visualization. They elegantly combine theory, practical and entertaining tips throughout that will help educate you, and make you a better vizologist.

Best give me a bit of time to the next post. I think I should go re read one or two of these and brush up my own skills.

It’s been said before, many times. It will be said again, many times (likely by me) that a Pie Chart is a poor way to represent data. As humans, we are not well evolved to easily and accurately compare the differences in area. This in contrast to how we handle length, where we have a great ability to discern small differences in the length of what we see.

So they start off weak. They then take a turn for the worse in the many ways our computers now let us manipulate them. It’s way too easy to add a bunch of data to a pie chart – exploding the number of slices. Add in colour thanks to a drag and drop and those explodes slices are now multi coloured rainbows (if you are lucky), a wheel of colour death if you are not.

We can use the chart type for evil even more (whether intentional or unintentional) by adding in values that exceed 100%. A pie chart really can only show you the portion of a whole, so when you add in values of percent that add up to 100s of percents, with all those colours and slices, it makes the head hurt.

So they can go wrong. They can though help tell a story. If you don’t need a kaleidoscope of colour or more slices than clouds in the sky to make your point, there is hope for a pie chart (and if you do need the slices and lines, you need another chart type, or more than one).

An example of one I might use for effect. For context, I was born in Canada but spent more than half my life in Asia Pacific How can I easily explain that to someone? Well, I do keep data and have some charts as above – one on all the places I’ve lived and about for how long. And one on the right on how much of my life has been spent in Asia. I would never subject anyone to the multi slide one except for educational purposes, yet the two colour one could fit a visual resume or presentation. It says, and only needs to say, I’ve spent more than half my life in Asia. I don’t need you to know if its 55.2% or 54.7% , nor where those years accumulated. Just that simple bit of a story.

And for that, while pie charts are less than great, I think the one on the right is a better chart than where it started.