COFFEE DATA SCIENCE

Espresso Baskets: Precision vs. Regular

Exploring an easy upgrade for most machines

Published in

Towards Data Science

7 min readSep 28, 2021

One of the easiest upgrades for many espresso machines is the filter basket. In the past two decades, precision filter have changed the field of espresso. In this piece, I look at the stock filter basket for my lever machine (Kim Express) and a Pesado precision basket.

Left: Kim Expres Basket, Right: Pesado Precision Basket

The test itself was tricky because the regular basket (16g) is a smaller basket than the Pesado (20g). So I made some adjustments:

Regular: 16g input

Pesado: 21g input

Then I aimed for an Output to Input Ratio of 1.25 for both as a fair comparison point.

Filter Holes

I have examined the filter basket holes in a previous work which showed precision baskets had much less hole variability than the standard basket. I have colorized the holes based on hole size where a darker blue is smaller and a lighter yellow is larger. These two images are colorized to the same scale with respect to the average hole size for each. So the color differences represent differences relative to the average, and the STD here is the Standard Deviation of the Kim Express because it was larger than the Pesado.

To start, the Kim Express basket has around 480 holes while the Pesado basket has 715 holes, so how much each hole on the Kim basket affects performance more than on the Pesado. The hole size and STD are much larger for the Kim basket than the Pesado.

The Kim Express has a large, uneven chunk of holes in the middle. For the Pesado filter, there is less local variation, but as you get to the edge, it seems the holes get a little larger on the edge. This could be caused by the slide curve of the bottom of the basket, but these changes are much less than the spatial changes of the Kim basket.

The Shots Compared

We can first look at the images from the videos. Each image is 5 seconds in time. Additionally, I did a 30 second pre-infusion and used pressure pulsing during infusion. The regular filter images are on the left, and the precision filter images are on the right.

Regular………………………………………..Precision

Infusion Starts:

The flow from the regular basket seemed to be slower, but it didn’t have as much of a donut affect as the precision. However, during infusion, the regular basket seems more uneven in flow.

We can look at the bottom of the pucks as well to see what the darker spots can say about channeling.

The regular basket has a lot of dark spots on the right side as well as an over all darkness except for the top left quadrant (just North West of center). The precision basket has a few odd spots, but it looks a bit cleaner. The holes are more distinct for the regular basket.

Metrics of Performance

I use two metrics for evaluating the differences between techniques: Final Score and Coffee Extraction.

Final score is the average of a scorecard of 7 metrics (Sharp, Rich, Syrup, Sweet, Sour, Bitter, and Aftertaste). These scores were subjective, of course, but they were calibrated to my tastes and helped me improve my shots. There is some variation in the scores. My aim was to be consistent for each metric, but some times the granularity was difficult.

Total Dissolved Solids (TDS) is measured using a refractometer, and this number combined with the output weight of the shot and the input weight of the coffee is used to determine the percentage of coffee extracted into the cup, called Extraction Yield (EY).

Performance

Looking at taste, the precision basket was a clear winner. Thinking back to when I made the switch a few years ago, there was a huge improvement in taste.

Surprisingly, the TDS and EY was lower for the precision basket. The final output to input ratio was 1.25 for the regular basket and 1.24 for the precision. One issue with this comparison is that the doses were very different due to the basket size.

In terms of time, it took the regular basket a lot longer to hit 10ml of liquid. That’s interesting considering the dose difference was so large.

Additional Data

Some time ago, Sprometheus posted a video about precision baskets, and he collected a little bit of data. He displayed it in a table, so I threw it into a chart to see what can be seen from the small sample size. I know it is hard to make a larger sample size for TDS/EY when taste already tells you that precision baskets are better.

The first graph is a scatter comparing the standard to the precision types. The second one plots TDS vs EY to help see strength vs extraction. IMS is clearly not doing well. In terms of TDS vs EY, the standard is still in the mix with the precision baskets. I’m curious how these measurements would be at different stages of the shot by salami shots.

The one part of this data that doesn’t sit right with me is the IMS performance. I’ve previously compared an IMS made basket (Pesado) with a VST basket over multiples shots and multiples roasts, and I didn’t see a difference in performance in either taste or EY.

Precision baskets completely changed my espresso experience. I would recommend them to anyone looking for a big improvement in taste. Of course, the flip side is that they show you how good you are at making espresso. Precision baskets hide nothing so you have to level up in your craft if you want to make the best out of them.

If you like, follow me on Twitter and YouTube where I post videos of espresso shots on different machines and espresso related stuff. You can also find me on LinkedIn. You can also follow me on Medium and Subscribe.