Red Brick and Freshness

How Does Freshness Really Affect Your Espresso?

Freshness is an interesting concept with coffee. It’s widely known that for the best brew, fresher is better, with one pretty major caveat: degassing. Here at Square Mile, like most other speciality roasters, we recommend resting beans for 5-10 days after roasting, before putting it into the hopper and then using that coffee within a month from the roast date to get the best results. But how accurate are these recommendations? What actually happens to the beans, and the coffee you brew from them, during that month? And, most importantly, when is your espresso going to taste at it’s best?

To put this to the test, I collected samples of our Red Brick blend from 5 separate roasts over the course of a month and devised a series of brewing experiments, to help us track just how the age of the beans really affects each facet of our espresso brewing, from grind size to extraction to the all-important flavour.

Given our existing knowledge, we would expect our freshest sample to be the most difficult to brew to our recommended recipe; it would likely need a much coarser grind size to achieve the same brew times as our rested samples and would be the most likely to under-extract. We would also assume that the freshest sample is likely to be the most unpleasant in terms of flavour.

As for our rested samples, there are some assumptions we can make about how they will differ with age. The older the coffee gets, the easier it should be to extract, making it more likely to potentially produce excessively bitter flavours; we would expect that coffee will taste at its best when rested for 1-2 weeks after roasting, and to begin to taste progressively less pleasant as time goes on.

Effect of Freshness on Grind Size & Brew Time

To test this, I took 19g (+/- 0.1g) of each sample using a single consistent grind size and recorded the time taken to produce a 38g (+/- 1g) espresso.

SAMPLE	SHOT TIME
1 day off roast	33 seconds
1 week off roast	30 seconds
2 weeks off roast	29 seconds
3 weeks off roast	29 seconds
4 weeks off roast	27 seconds

Shots brewed using a VA Black Eagle espresso machine and a calibrated and purged Niche ZERO grinder. Results shown are averages after the experiment was carried out 3 times.

As these results show, fresh espresso brews slower than rested espresso; this is due to the carbon dioxide inside the beans being rapidly expelled and creating additional resistance against the brewing water. As more of the carbon dioxide dissipates, this resistance is lost, allowing the shot to brew more easily and consistently.

Additionally, coffee loses density and moisture as it ages, so the longer the beans have been rested, the finer they will need to be ground to maintain a consistent brew time. So far, these results are in line with our expectations and our standard resting recommendation; coffee that is 1-3 weeks off its roast date behaves as we would expect and hope it to, while both coffee that is fresh and coffee that is beginning to stale won’t produce optimal results without intervention from the barista.

Effect of Freshness on Strength & Extraction

To test this, I produced an espresso from each sample to our recommended recipe (19g dose, 38g yield, 28-30s shot time) and recorded the strength and extraction using a refractometer.

COFFEE SAMPLE	STRENGTH (TDS)	EXTRACTION
1 day off roast	9.55%	19.1%
1 week off roast	9.32%	18.6%
2 weeks off roast	9.14%	18.3%
3 weeks off roast	7.91%	15.8%
4 weeks off roast	7.77%	15.5%

Shots brewed using a VA Black Eagle espresso machine and a calibrated and purged Niche ZERO grinder. Results shown are averages after the experiment was carried out 3 times.

Here, our results become more unexpected. The first surprise is that for our unrested sample, the measurable strength in the cup is still perfectly in line with our rested samples, despite the expected potential for under-extraction. Even more surprising is that, rather than over-extracting as expected, our ability to produce the strength and extraction desired actually drops for our progressively older samples. This result is the exact opposite of what we might have expected; one potential reason could be that our more recent samples have been roasted to a more developed profile and therefore are easier to brew than our older samples.

If however, we were to assume that this shift in strength and extraction is not simply the result of roasting changes, we would need to adjust our brewing parameters in order to get the best results for our espresso. One way of achieving this would be to extend our brew time or adjust our brew ratio with older coffee to increase extraction to the recommended levels; to test how this might be achievable, I brewed our oldest sample again to two different recipes and compared them to our initial findings.

ADJUSTED RECIPE	STRENGTH (TDS)	EXTRACTION
19g dose, 38g yield, 34-36s	9.81%	19.6%
18g dose, 40g yield, 28-30s	8.78%	19.5%

Shots brewed using a VA Black Eagle espresso machine and a calibrated and purged Niche ZERO grinder. Results shown are averages after the experiment was carried out 3 times.

Here we can see that these minor adjustments have allowed us to achieve the desired strength and extraction as expected when using coffee that is nearing a month old; however, it’s likely that a barista would soon hit a ceiling in how far they can push the extraction using this method, as older coffee will begin to degrade beyond being saved.

Effect of Freshness on Flavour

To test this, I produced an espresso from each sample to an optimum recipe (19g dose, 38g yield, 28-36s as required to produce the desired extraction) and had a group of 15 volunteers blind taste-test each sample. Each volunteer was then asked to vote for their favourite on the table, guess which of the samples was the freshest, and record any comments.

COFFEE SAMPLE	FAVOURITE VOTES	FRESHEST VOTES	COMMENTS
1 day off roast	0	12	”˜sharp’ ”˜sour’ ”˜blows your head off’
1 week off roast	2	1	”˜weird aftertaste’ ”˜bright’
2 weeks off roast	1	2	”˜familiar’
3 weeks off roast	3	0	”˜sweetest’
4 weeks off roast	9	0	”˜fruitiest’ ”˜good acidity’ ”˜best balance’

Shots brewed using a VA Black Eagle espresso machine and a calibrated and purged Niche ZERO grinder. Results shown are collated from blind taste-testing of 2 brews per sample.

Here, the results were almost unanimous. 80% of the voters were able to correctly identify which of the samples was the fresh coffee, and the comments were consistent with what we’d expect from brewing unrested espresso; under-extracted, sharp and sour flavours. More unexpected is the result that 80% of the voters preferred coffee that was more than 3 weeks old, with 60% preferring the oldest coffee on the table.

Conclusion:

From these experiments, we find that our recommendations regarding resting coffee are correct; coffee that is freshly roasted will be difficult to brew correctly due to the escaping carbon dioxide and will produce espresso that is generally unpleasant in flavour with a dominant sour taste. Regarding older coffee, however, we find that our recommendations do not take into account the changes that occur in coffee over its first month after roasting, and therefore are not ideal for allowing our customers to get the best results.

Firstly, we see now that coffee requires different ideal settings for brewing at different stages throughout that first month; for our wholesale customers, this means that consistent monitoring of espresso strength and flavour as it ages is imperative to ensure that your customers are getting the best brew you can give them. Secondly, we see that our recommendation for using coffee within one month of roasting is maybe too restrictive; if coffee that is 4 weeks off roast can taste as good, or even better, than coffee one week off roast, then perhaps we should be more charitable towards our use-by dates. This may not make a difference to our wholesale customers, who will typically be finishing their stock within two weeks of its delivery, but for home baristas, this is great news; go ahead and brew that bag of coffee you worry could be a bit past its best, and you might be pleasantly surprised by the result.

Since some of our results were so unexpected, we’ll be checking back with these experiments at a later date, so be sure to check back for updates then!