Late September while walking along the river, I spotted a dog happily playing fetch in the water. I was wondering at the time whether the dog was cold or not.

Staying warm with bad data

Oct 28, 2025

I've lived in ancient leaky NY housing stock for the vast majority of my life. It's the sort of construction that dates from the early 1900s and the 1950s, and the one hallmark they have is that none of the building practices of those eras had any concept of insulation. The only thing we had going for us in the attached row houses of Brooklyn was that only the front/back/roof were typically exposed to the elements. Two of the structure's walls were "insulated" from the elements by your neighbors.

Anyways, heat leaks out of all the places I've lived in like you wouldn't believe. We do what we can with things like heavy curtains, upgraded door and window seals, rugs and carpeting, but ultimately warm air finds nooks and crannies to sneak out, while cold finds new ways to sneak in. Getting a thermal camera "for research purposes" has been a really fun education in observing just how bad things are in this ancient house. This was especially true when last winter we were dead in the middle of renovations and at one point in October we were missing an entire exterior wall. That tends to really screw up heating systems.

Thermal image of a wall of the room I'm writing from. you can see a hot raspberry pi at the window, the cold spot at the ceiling line, the heat rising up, and the wall is also cold.

With energy prices being what they are these days, keeping the temperature in a comfortable spot can sometimes feel like setting bags of money on fire. So, in the name of frugality and exploration I went around with thermal camera, infrared thermometer in hand trying to see if there's anything I could do to improve living conditions... After all, I can collect DATA! Right?

Well, not really.

Aside from a few random spots, there's not a whole lot of places I can correct a cold air leak in the house. I'm certainly not about to just cut out chunks of the wall and pump insulation in because that's been known to cause all sorts of horrible structural issues unless done correctly. So all the fancy, eye-catching, graphical representations of just how inefficient my living space was doesn't actually amount to much actionable insight. Yes, this thermal camera is nothing more than an expensive vanity metric dashboard =O. It makes you feel like you're learning something but is something you could do without. I'm lucky I got it to help detect leaks and observe electrical circuits and not energy efficiency.

Instead, the thing that is helping me on energy expenses is a simple Kill-A-Watt power meter, When we had that hole in the house and couldn't use the normal heating furnace, we relied on electric space heaters in our room to stay unfrozen. The meter allowed us to realize we only used a couple of dollars of energy to keep our little room warm during the time. Right now it's telling us that the nice electric blankets that we're using to keep our beds nice and toasty will use less than $5 of energy for an entire month.

What's really funny is that the Kill-a-watt device is like... one or three tech generations behind. It's really good at measuring energy usage (down to 0.1 watts), but it can only do two major things: tell you what the current instantaneous energy usage stats are (watts, voltage, frequency), or a running total of watts consumed and hours elapsed (so that you can calculate the important kilowatt-hours that appear on your energy bill). Mine doesn't even have a memory so it loses data when I unplug it. The whole interface is few buttons and a liquid crystal display.

More modern 'smarter' data loggers that do important things transmit records of energy use over time either don't exist, don't have the same accuracy, or cost many times more. It's bad enough that a measurement nerd like myself doesn't own a fancier one, because I can't find a decent alternative.

And yet! Despite it's supremely primitive list of functions, it is enough to make decisions with! I've learned that my previous window AC unit was so efficient that it costs tiny amounts to keep our bedroom cool in the winter guilt free. Same for when we had to space-heat our bedroom during renovations. Same for using heated blankets as much as we wanted. Same for figuring out how much energy my refrigerator dishwasher is using. Even when I have to do the manual division of kilowatts used by elapsed hours, it was enough to make decisions.

I find it very amusing that the most absolutely un-sexy device for monitoring my energy use offers the most utility. It's very similar to how fancy dashboards are ignored by millions of workers every day despite them being plastered on always-on TV screens in an office, while boring metrics like "revenue" and "customer retention" that have been measured for eternity continue to cause giant shifts when they move unfavorably.

The majority of my "big impact" analyses always eventually tied into similar boring things. But we're always drawn to new and different metrics because... well... they're new and interesting. At the least, they haven't been meaningfully done before unlike the boring accounting stats. But this is a reminder that novelty is completely divorced from utility. If anything, novel things tend to be useless things because only a small percent of new stuff stands the test of time.

Subscribe!

Standing offer: If you created something and would like me to review or share it w/ the data community — just email me by replying to the newsletter emails.

Guest posts: If you’re interested in writing something, a data-related post to either show off work, share an experience, or want help coming up with a topic, please contact me. You don’t need any special credentials or credibility to do so.

"Data People Writing Stuff" webring: Welcomes anyone with a personal site/blog/newsletter/book/etc that is relevant to the data community.

About this newsletter

I’m Randy Au, Quantitative UX researcher, former data analyst, and general-purpose data and tech nerd. Counting Stuff is a weekly newsletter about the less-than-sexy aspects of data science, UX research and tech. With some excursions into other fun topics.

All photos/drawings used are taken/created by Randy unless otherwise credited.

Supporting the newsletter

All Tuesday posts to Counting Stuff are always free. The newsletter is self hosted. Support from subscribers is what makes everything possible. If you love the content, consider doing any of the following ways to support the newsletter:

  • Consider a paid subscription – the self-hosted server/email infra is 100% funded via subscriptions, get access to the subscriber's area in the top nav of the site too
  • Send a one time tip (feel free to change the amount)
  • Share posts you like with other people!
  • Join the Approaching Significance Discord — where data folk hang out and can talk a bit about data, and a bit about everything else. Randy moderates the discord. We keep a chill vibe.
  • Get merch! If shirts and stickers are more your style — There’s a survivorship bias shirt!

Randy Au