Matt Perger on the definition of “baked” coffee defect

I’m reading Matt’s article in the Barista Hustle about coffee being baked in the final stages of a roast. I always thought that stalling was supposed to lead to baked flavors but he is saying that the RoR increasing towards the end of the roast is what causes this to happen.

End Speed

At the end of the roast, the coffee beans are quite dry and fragile. Small, brief changes in temperature can make or break the whole roast. Here is an excerpt from his blog, “Let’s Talk About Roasting.”


Coffee Roast Baked

If the coffee experiences an increase in speed somewhere after first crack, that will “bake” the coffee. This is characterised by the coffee exhibiting a lack of sweetness and/or dark roast flavours, even though it might not be a dark roast. You can finish a light roast with a bake that will then taste a little bit bland, ashy or dry. Not ideal. Try not to fall into the trap of calling a baked roast “dark” (it happens all the time and you’ll look like a dingus). Look out for dryness, a lack of sweetness, dull acidity, and in worst cases; ash. Sang Ho from Square Mile Coffee Roasters introduced me to calling this the “Flick of Death”. Precisely. It kills the party.


Coffee Roast Stalled

If the coffee temperature stops increasing for a significant amount of time, it’s called stalling. Sometimes, the coffee might stall so hard for so long that the temperature starts to drop a few degrees. The worse the stall, the less developed the coffee will be. It’s super hard to pick this one out without seeing a roast profile. If you start calling out roasts for stalling without seeing the profiles be prepared for a slap. Stalling can create weird thinness, waxy cardboard flavours, sharp acidity and sweetness that’s frail and lacklustre.


Coffee Roast Ideal

The coffee can’t experience an increase in speed, and it can’t stop, but it should still be rising in temperature after first crack. At first glance this might seem impossible, but it’s not that hard. Just think of a car constantly slowing down until it reaches a stop sign. Before and during first crack, the coffee should be increasing in temperature quite quickly. This momentum carries it further and hotter after first crack, and allows for a constant slowing until the end. The ideal end of roast is constantly slowing down after first crack until the speed approaches or reaches zero right at the end. This results in no baking and no stalling. If you don’t notice any problems with the coffee, it was likely ideal.

This comment is interesting:

I’m loving it too, but also don’t get the comment “What I try to do is achieve the highest possible extraction that doesn’t taste dry with the coarsest possible grind. That is almost always the best combination.” at other times it seems like a very fine grind is preferred by Matt…because with coarser grinds you are going to have to brew longer and then you overextract the fines

Matt replies a few comments later:


Gents! You’ve found me out! (kind of)

I tend to recommend “as fine as possible” when people are sifting, because they have total control over particle spread.
When using a normal grinder/brewer, the coarsest point that extracts well almost always tastes better.

Re solubility – I’ve been trying to max solubility at lighter roasts for 3 years now. Getting there..

This comment helps too:

I’m basically using an espresso grind, sifting fines with 100um, boulders with 475um (this gets rid of 40% of the initial coffee weight because I’m using a conical grinder), fantastic sweet, clear, transparent result! (lots of stirring too) Will something like an EK give me less wastage from sifting AND “better” morphology? Is there such as thing as “better” morphology? ie a sphere is perfection?

I’ve been getting recommendations for a grinder for pour over but I think most of the advice assumes I’m using a traditional pour over filter grind, and so they say, “if you’re not using it for espresso, the Ditting 804 or Bunzilla is as good as the EK43 but cheaper, the real benefit of the EK is it can do espresso.” I might post my pour over routine on YouTube and see what you think!

OK. Here’s another comment from Matt that starts to confuse me but also clarifies somewhat:

I’m not disagreeing with you on stall.

I am with bake, though. The Q course sits a roast at the same temperature for 5 minutes and calls that baked. You _have_ to apply extra energy to the roast to make that happen. Ergo, bake = too much energy.

I did say a lot would disagree, no? Read my replies below. If the differential of the curve sits above ideal (flat lining) it’s baking. If the differential sits below (short flat line followed by decrease) it’s stalling. They’re different things and create different flavours.


Author: korkiley

Systems Administrator at University of Vermont (retired as of 7/1/2012) Married Favorite Activities: Condor Glider Online Competition, Developing web sites, making espresso, and keeping a blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: