Aquascraper 2018 Review

Towards the end of 2017, I built the aquascraper, a Node and Phantom.js tool that takes the sales data of aquabid.com and stores them in a database. Everyday it goes to the site and uses the closed auctions link to look through each fish category and save the sales data from the past day. It basically works like this:

A visualization of what aquascaper is doing

That data is sorted by the categories that Aquabid provides and checks to see if the reserve price was met. I then take that data and display it in a series of graphs that update daily.

Graphing aquascraper data

While the technical implementations are very interesting (at least to me), after over a year’s worth of data, I thought it was finally time to sit down and do a bit of an analysis of the past year. If you’d like to join me, all of the data is publicly available here: all data(13.2MB), daily-averages (1.5MB), sales-by-month (11.7MB)

First thing’s first, I built a page similar to my daily aquadisplay, using only the 2018 data and examining the entire year. Warning, the site is not mobile friendly. Because rendering all the data points for every day would be both hard to read and hard on the browser, the sole initial category displayed is the average sales for that day, a sort of market average. Just by looking at this single line we can gain some valuable insights.

The single best day for Aquabid was January 30th, where the average of all sales was $37.3. This is due to a sale for Angelfish, where a pair of deep blue pearls were sold for $500 and a strong showing by the rest of the fish types. December was the only month that shows a significant slow down in sales. Makes sense, it’s the holidays.

The market average (Dark Blue)against two of the biggest performers, Plecos (Light Blue) and Angelfish (Orange)

The following dates had no sales and are not represented on the graph: Jan 6, Mar 11, May 30, June 9–11, 21, 24, July 3, 15, Sep 11, Nov 18, 24, 28, Dec 3.

Quick Tip: You can use the tab key or click to see each dates information displayed in a tool tip.

The highest sale was a Plakat Betta for $2500. However after closer inspection, that sale was likely a fat finger on someone’s part. Similar sales by the same seller and by were made that were $25, also for Ghost Plakats. That data point has been adjusted to be $25. I suspect that the actual sale was closer to that price.

The actual highest sale was a tie for $1500, both for a group of breeding L46 Zebra plecos. This follows with the trend of Pleco’s tending to be higher priced, and Zebras being a very beautiful species of an otherwise… acquired taste for the family. One of the auction pages is still public at this time.

But by volume, Plecos only rank 7th out of all the categories Aquabid provides. Perhaps unsurprisingly, Bettas, specifically Halfmoon Bettas, accounted for the majority of the sales on the site. 3556 Halfmoon Plakats and 3285 Halfmoons were sold in 2018. The next largest category by volume were Invertebrates with 2282, followed by Killifish Eggs with 2046. It’s interesting to note how many more Killifish Eggs are sold than live Killifish.

Sales by Volume

Interestingly, no Koi were sold in the last year on Aquabid. I went back to manually check if this was correct, and as far as I can tell, this is true. Aquabid just doesn’t seem to be the place for Koi sales. Being a specialty fish (larger, more expensive, and often in ponds instead of tanks), it’s likely that koi are just sold in a different marketplace.

I’m just scratching the surface here and if you find any interesting information or use the data, please let me know!

Edit: I’ve just added sales volume by month. Sales in January and February are almost triple what they are the rest of the year. This seems to be in line with what members of the community would expect.

Sales Volume By Month