A love/hate relationship with espresso

It goes without saying that trends and fashions are cyclical. What is in one day is out the next, only to be back in fashion once embracing it seems sufficiently different to what everyone else is doing. Such is brewed coffee vs espresso.  (I should probably warn you now that this is going to be a fairly long post…)

Outside of fashion my own feelings about espresso have changed dramatically over my career. It was all I knew coffee brewing to be for the first couple of years, and then I fell in love with brewed coffee and espresso seemed so awkward, so difficult and often so unrewarding in comparison. I was probably a bit too negative about espresso – I vaguely remember the idea of writing an article where Chris Tacy would defend espresso, and I would be all about brewed coffee.

Considering Chris’ fairly recent post this might seem a little odd. That particular conversation has gone down one path and I want to talk about something else – which is how much I enjoy brewing espresso these days.

I was recently running a couple of workshops at the Caffe Culture tradeshow, and I realised that explaining my own evolution of emotional attachment to espresso could be a good structure to explain how I now think about brewing.

This all starts with walking offstage in Berne in 2006 in the WBC finals.  I was delighted to be up on stage, I had a lot of fun but I definitely remember the feeling of walking off stage and feeling that I had absolutely no idea how to brew espresso.  I knew I had great raw materials, well roasted, and some nice kit to brew on.  I had been unable to push and prod my espresso into tasting how I wanted it to – I simply didn’t have the knowledge to exact the change on the cup profile that I wanted.

Looking back – I was serving fairly high dosed, fairly light roasted, underextracted and strong espresso.  It was sour, and no matter how much I slowed down the shot I couldn’t get it to sweeten up the way it had in practice.   Maybe I will come back to why later….

Espresso was so dispiriting for so long.  We end up using words like art for brewing, which I think is probably wrong.  Brewing is execution of task.  It should require craft, but not art.

Let me put this another way:  If I walk into your shop and order an espresso – how confident (as a percentage) would you be that the next shot you pull and serve will represent you, your shop and your coffee properly?  Be honest….

For a long time I felt that my own number was probably around 40-50% on a good day.  Sometimes espresso sucked, and I didn’t know why.  After the WBC the stress bumped up a notch – if I was serving coffee somewhere people would walk up with expectations and I rarely felt confident that I met them.

This nature of espresso is why there has been so much bullshit, myth and voodoo around for so long.  We explain things by looking for trends rather than explanations.  A classic example of this would be our approach to coffee grinding an humidity/weather.  I would argue that the weather will affect your grind, mostly by affecting how busy you are.  This impacts how hot your grinder is running, resulting in a grind change.  I simply haven’t seen a change in grind happen quickly with a rapid change in moisture/humidity. (I could imagine a gradual change would make more sense.)

I don’t want to get into a humidity debate – just an example of causation vs correlation.  Another might be the advent of naked portafilters.  At the time people were suddenly raving that they produced espresso with more body, more intensity, more sweetness.  The explanation was simply that people were pulling shorter shots, that looked normal volume because of the additional crema, but doing it so fairly slowly – they were pulling reasonably extracted ristrettos, they’d changes the recipe rather than the absence of a portafilters bottom somehow impact the extraction that occurred inside the basket above.

I’m getting off topic – I want to come back to why I enjoy espresso more.  I enjoy it more because it I have a functional recipe that I can replicate accurately, and relatively easily.

No doubt Vince Fedele had a big impact on this – with both the Mojo and the also the VST baskets.  I’ve already seen some strange things written about them, and what they do, so perhaps it is worth explaining a bit more.

We brew espresso by controlling flow rate using the resistance of the coffee cake.  Traditionally we think of two variables to control that resistance:  grind and dose.  Using these we can control how long the contact time is for a fixed volume of solvent (water).  The combination of contact time and quantity of water basically decide how much we extract from the coffee.

We didn’t factor in a third factor into that resistance – the bottom of a portafilter basket.  We tend to think of their job as being about filtration rather than impacting contact time.  However, they ought to be treated as an additional constant of resistance.  Most baskets are built with the majority of end users in mind.  Most end users – and that is a very large majorirty – are dosing around 7g per shot of espresso/14g for a double.  With a relatively small dose you need to grind pretty fine to get your desired contact time which has the benefit of exposing an increased surface area which makes the coffee easier to extract.

Higher doses – 18g+ came about somewhat by accident.  They were a product of grinding to order and using a basket to measure a dose of  ground coffee by volume.  The increase in dose meant additional resistance.  The basket resistance, plus the higher dose resistance meant we had to grind coarser.  This coarser grind made the coffee a lot harder to extract.  We found we liked the texture and strength of a higher dose of coffee, and this trend went further with people pulling shorter and shorter shots.  This meant we had less and less water to use in our extraction.  End result = underextraction.  A byproduct of this was darker roasting to help balance out the sourness of the underextracted cup.

The VST baskets are great because they do a few things exceptionally well.  They have less resistance the larger they are.  This means that you can grind much finer than you otherwise could before.  This means you can extract more.  This is good news.

Secondly – the way they are manufactured means that they are extremely consistent – an exact number and size of uniform holes. This means that if you have a 3 group all the baskets will act exactly the same.  Plus they’re built to be much thicker so they’ll last a lot longer.  This is a bit of a simplified explanation but hopefully you’ve all read the article in Barista Magazine that Vince wrote because it was great.  (Best edition of the magazine ever too!)

I haven’t really explained why this means I like espresso more.  I should probably do that now.

If I take a great espresso machine, and a VST 21g basket, and I put 20g of coffee into it and in around 28s I produce around 32g (brewing at 94C) then I know with 90%+ confidence that it will taste how I want it to.  That is, to me at least, pretty exciting.  It also massively reduces the stress of espresso.  I know that it won’t be a perfect shot, but it will be very tasty and something I’d be happy to sell.

I’ll be honest – the above recipe is where I always start.  I won’t taste it till I get there – because I don’t want the first 2 or 3 espressos I drink each day to suck.  I’m all for tasting bad espresso for diagnostic learning, but I’m also all for actually enjoying the damn drink.

Based on what I taste I might make changes, adjusting how I extract based on the cup.  If things are a just a touch out of balance on the acidic front then I will likely do something pretty simply like use a little more brew water to up the extraction.  34-35g of liquid from that dose will still have great mouthfeel if you brew it properly.  Very rarely do I need to make that change.  I know what kind of extraction yield I have, and I know what I like.  I’m not going to be prescriptive in that – the point of analysing espresso extraction (for me) is to understand what I enjoy, how I can change things to get there and to define the boundaries of good coffee for the recipe I am trying to make.

I’m sure some people will be horrified by the simplicity of the brewing.  Shouldn’t I spend longer “dialling in”, fighting my flawed grinder (because they’re all flawed), and burning out my palette to emerge the battered hero of espresso brewing, victorious in my reasonably good espresso which I might then struggle to replicate with any great consistency?  It amused me to see that Schomer (from his last blog post) 1 seems to revel in the challenge, in the difficulty, and sees it as being an important aspect to treating coffee brewing as a craft.  I don’t agree, but it is ok if you don’t agree with me (or with him).

I’ll be honest – I think weighing scales are an essential part of a good, streamlined, efficient dialling in routine in the morning.  I think they’re a great quick QC tool when you’ve got a tired tongue.  I think they’re a great training tool.  Volume measurement being so common and so flawed is one reason that I struggled with espresso for so long.  I think it is a very good reason that many other people struggle with it.  I think it is crazy not to spend £10 on a little tool that can do so much good. Do we really want to keep struggling with it to preserve our own egotistical version of our craft – or do we simply want to sell people delicious cups of coffee? If we can get past the basic challenge of consistent execution then perhaps we can get to a place where we can be much more creative with coffee.

I’m not saying that I can now brew perfect espresso every time.  That would be as idiotic as using the word perfect to begin with.  I find brewing espresso much easier than ever before, and I drink more espresso that I enjoy than ever before.  There are still many challenges with it, and a lot of things we don’t understand.  There are other factors impacting cup quality in ways that are hard to quantify – can’t help but constantly come back to grinders here too.

I promised a while back that I’d stop going on about weighing espresso.  I’m not going to apologise for breaking the promise, and I hope other people identify with my own to and fro with espresso.  It is still a completely frustrating thing in many other ways – if you listen to the last podcast with Tim Wendelboe then we both talk a little about this.

  1. There is so much in that article that I’d like to respond to, but I don’t really want to get into an online blog tit-for-tat blog tennis match  ↩︎

32 Comments

  1.  James, great article for Sunday afternoon … thanks :) p.s. The URL for Schomer post is not working …

  2. Brilliant post James. Really like that you expand with a bit of historical context in your posts.
    I’d bet your starting point for espresso (your “recipe”) is similar to what a lot of roasters would prefer. We go a little faster (25 sec) and a little longer (34 g liquid) with the same dose. But I’m wondering if you’ve experimented with the smaller VST basket and lower doses, and if you have gotten any good/interesting results from that?

  3. I’ve had the 18g baskets for a couple of month, and I really, really like them.  I like being able to create a slightly smaller version of the 20g shot, using the same ratio.  Any reduction in caffeine intake that allows for more coffee drinking is ok with me (even if it is just 10%).

    With many baskets a slightly faster, longer shot would probably work well – which baskets are you using?

    Tolerance wise – I’ve had good shots in the 18g baskets dosing from 16-19g, and a similar range of 19-22g in the 21g baskets.  (It is always easier to extract less coffee than more!)  This is inevitably tied into roast level, as this has a big impact on density (i.e. I wouldn’t brew 19g of dark roasted coffee in an 18g basket, but that much of a lighter roast would usually be ok.) 

  4. Off topic I know… but in terms of roast descriptors, light, dark etc.. would there be any merit in describing the roast in other terms, such as retained moisture content (as a percentage)? Or is residual moisture in roasted coffee not the factor influencing density?

  5. After reading this, the first word that came to mind was the same one that Klaus used: “brilliant.”

    As usual, you begin with a refreshingly honest description of the possible, umm, shortcomings of some of your previous espresso endeavors. Thank you for the perspective.

    What I love is how you suggest a dramatically simplified shot-pulling technique — or, at the very least, radically simplified the first attempt at pulling a shot with any new coffee. Start there, and then make small adjustments to taste, if necessary.

    Now can you please work out a simplified version of pourover technique? Many of us are drowning in a sea of five-dimensional, inductively heated, zero-pour, cake filtered, slightly domed, pulsing Buono-ed, V60s….

  6. Andy, I know I’m not James, but for a simplified pourover technique, I brew nearly all my coffee at home to good effect with a simple method. 27 grams coffee, 460 grams water. Fine-ish grind. Approximately a 3.75 on a Marco Uber Grinder (a 9 on my finely recalibrated Baratza Virtuoso at home). Water heated to 201°F (heat source irrelevant), rinsed white V60 02 filter in a ceramic V60 holder. Pour double the weight of coffee in center of coffee bed and wait 30 seconds. Then pour the remaining volume of water into the center of the coffee. No fancy pouring, no worrying about stirring, no bull. It’s just a quick, simple pourover.

    That being said, thank you James. I feel this sums up a lot of what I have been thinking the past while without me being able to effectively write it down. My view of the VST baskets is very much a reduction of the influence of a variable. As a scientist, it is a huge benefit when you can classify the effect of a piece of your apparatus. I’m looking forward to using mine in the shop when we get them in.

  7. Brewing as art?  I never understood that.  Craft?  Certainly.  But art?

    I think the question you posed is an interesting one.  If someone were to walk into my shop and order an espresso, how confident am I that the next shot would represent me, the shops and the coffee properly?  I’d say that on any given day, I’m 85-95% confident that the shot served to our guest will be to our expectations.

    While I would prefer that our confidence level hover in the 98-100% range, we’re still dealing with a fallible product. One whose potential variables can be a constant state of flux.  We’re also dealing with the craft of the barista and the potential problems that can arise in the processing of that one basket of espresso.  From slight inconsistencies in dose to distribution problems, resistance errors, grind issues and even down to the potential of that one, solitary charcoal bean that got stuck in the far recesses of the roaster having made it’s way into that particular basket.

    If hardcore consistency is what we are looking for (to propel us into the 100%+ range) then perhaps we ought to be looking towards more automated brewers/espresso machines or even outright synthesis of espresso flavor profiles. Of course, I’m not promoting this and I know that you’re not promoting this as well.

    The difference is that this is what I do for a living. I make coffee. For guests.  On a daily basis.  From open to close, all we do is make coffee for guests.  Our reputation as a company is based on our ability to successfully reproduce a consistent experience for our guests.

    I hear you when you say that in 2006 you felt your espresso sucked.  Comparatively speaking, ALL of our espressos sucked as opposed to what we’re brewing today (at least that’s what I hope everyone else is experiencing).  Stiil, that under extracted espresso of 2006 was good enough to propel you to the WBC Finals – it’s really a commentary on how far our community has progressed rather than a condemnation of our past.

    For the past couple of weeks, we’ve been running quiet testing of the 18g Strada/VST basket, alongside our standard La Marzocco double shot baskets.  So far, the results have been quite interesting.  Running side by side, you can start to see the effect of basket inconsistencies v. consistent basket resistance.  For us, it’s still too early to make solid conclusions or arguments but the results look very promising.

    Since we’re only using the 18g basket, can you discuss more on how the larger baskets have less resistance?  I’m presuming that the three sizes of baskets feature the same number of perforations and the same perforation diameters – if that’s true, how does the larger basket deliver less resistance?  And how does one measure that difference?

    One thing we’re still looking for is the dosing grinder.  It’s perfectly fine in a lab setting to weigh out your beans, toss them in the hopper and then grind away, but in a production environment we really need a grinder that will dispense by weight rather than time.

  8. Over many years, I’ve always liked filter brewed coffee. Then a love affair with espresso a little over a decade ago when I finally encountered a version that didn’t taste like volcanic ash after a rainstorm. Sure, I later experimented with a bit more with the vac pot, and there’s more toys to play with again with brewed coffee in recent years, but the common theme is the same: I’m still an espresso guy by preference and have been for over a decade now. I don’t see that changing anytime soon.

    Experimentation, whether a person’s coffee or their world travels or their sexuality, ultimately leads to something akin to a level of boredom and stasis. That’s where the individual needs to make a transition or a kind of a commitment. Being a mile wide and an inch deep only gets you so far. Visiting the world’s great cities for 12 hours apiece leaves only a superficial level of satisfaction of knowing the world. (I won’t get into the sexuality bit for family viewing reasons!)

    There’s a world of further richness to discover when you settle into a few of your favorite things, decide what those favorite things are, and continue to enjoy – and maybe with a little exploration still – fewer and more focused things for what they offer. Touring China is very different when you learn the language and develop cultural friends — as opposed to the view from a tour bus, armed with only a Mandarin phrasebook. And the industry now acts like a kid out of college trying to visit every city on the planet for only 15 minutes each. It’s not sustainable.

    So for us dinosaurs still stuck on the coffee “fads” of a few years ago, thanks for sticking up for good espresso. Boredom will inevitably set in on the V60 and pour-over set, and at some point you do have to settle down with what pleases you.

  9. 85-95% confident is a hell of a claim!  Out of curiousity how many shots do you brew a day and do you think that your hit rate is scalable to double or triple that?

    How are you deciding when a shot is finished brewing?

  10. Perhaps it is a hell of a claim, but you have to remember that we’re doing this on a daily basis and under relatively ideal conditions – meaning that we’re in our own environment, working on our own equipment with coffee that we (should be) intimately familiar.

    If you’re asking how many shots do I personally brew each day, it’s actually pathetically low.  As the company has grown and I’m doing more administrative work to (hopefully) ensure a smooth operation, I’m not making drinks on a daily basis like I used to.  That said, we’re consuming about 10 pounds of espresso per day.

    As far as scalability goes,  doubling our volume should not be too troublesome but trebling our volume and beyond will be difficult mainly because the Spro is small and the design necessarily limiting.  However, with a larger space that allows for greater flow, Intelligentsia-like numbers ($4500-$6000/day) should be quite possible.

    That said, I’m also a bit torn between operating a $5000/day shop and our current model.  At that volume level, something has to suffer – typically the guest experience in some form or another.  At our current size, our team is available to give personalized service to each of our guests without getting lost under the pressure of a 25 person line that never ends.

    Of course, on the other side, is that a $5K/day shop is generating revenues upwards of $1.5 million per year, allowing the company to increase staff salaries and offering greater opportunities.  So while I like the ideal of a small shop offering a focused guest experience (a la The French Laundry), we’re still dealing with a product that we’re selling for between $2.50 to $14 per cup – which is a far cry from the 85 guests at $500 per person.

    Another thing I should mention is that a Spro Barista not only has to know espresso and espresso drink building, along with guest service, our Baristas also must have command over eight other brewing methods (PourOver [v60 & Beehouse], French Press, AeroPress, Chemex, EvaSolo, Abid Clever, Vac Pot and cold brew)  that can be ordered off our menu at any time for any coffee we have on offer on any given day.  Not to mention the allied beverages and ingredients we create in-house and knowledge of our food and retail products.  If anything, it’s a challenging environment for the barista.

    When it comes to espresso brewing, I train our baristas to look for a variety of variables, much of which are developed through field experience, but they’re looking at time, flow rate, volume and color.  

  11. Though our VST’s are still on their way, we’ve been following an almost identical dial-in procedure for a while now.  We also have a defined brew ratio that we’ve come to enjoy with many different espresso’s. Our daily procedure is as follows:

    Measure dry coffee mass (17g – 19g)Calculate desired extraction based on brew ratio (1.8x mass)Configure mirage volumetric settingExtract shot within preferred time (25-30s)Save setting.Taste.Make small adjustments if necessary.

    I’ll add one more comment on the variable vs. constant issue. Unless you’re weighing every shot, or using a volumetric espresso machine, you’re allowing one more variable to exist that for all purposes should be a constant. That is, your ability to consistently dispense a precise amount of water. You may have a great shot on dial-in when you’re weighing your extraction, but very few baristas can pull 200 shots in a day -through their busiest times and ensure they’re stopping their shots consistently and precisely at the turn of a second. 

    There’s both precision and simplicity in knowing you can work as fast as possible and be consistent in your extractions. I can’t imagine going back to our days of managing 3 timers -clear, set, stop, hundreds of times per day.

  12. I have Jay’s back on the high percentage claim.  I often say to our baristas that our shots should always be within one shot of perfect (yeah, admittedlyt a bad word but I’m trying to inspire staff here!)…Perfect that is IF they’re monitoring every shot.  There’s nothing worse than going to a reputable bar and having the barista waste 100 grams dialing in for you.  That should happen once per day with consistency being maintained throughout the day through attention to the individual pours.
    Jay: 10 pounds of espresso per day is a wide variable depending on dosage.  23 grams per shot on a 300 shot day is 2 pounds more than another day of 18 gram shots.  Do you have a benchmark for dialing in as James suggests?

  13. Espressos are good, but I dont know if you have ever tried Jamaican Blue Mountain Coffee…It’s a little on the pricier side but its worth it (you can wiki it), theres a site http://www.martinezfinecoffees.com has the best Jamaican Blue Mountain Coffee…If you like espresso I highly recommend!

  14. Great to see someone as accomplished alluding to what I sometimes think of as the ‘espresso fear’! A kind of paranoia that can be induced not so much by any lack of belief in your coffee, or even your skills, or your equipment (to an extent), but more by the fact its horrible to think you might not be doing the coffee proper justice, or doing the majority of what you try to do as a cafe justice …or even the speciality movement itself justice! Especially in those individual shots here and there that for one reason or another can make a difference. Its silly, but when you really care about communicating how good a coffee can be, its frustrating – with espresso particularly – to be unsure from one shot to the next whether it will be conveyed in the best light, when you KNOW how great it CAN taste when its spot on.
    I’m still at the point where my percentages vary widely – they’ll go up a bit when I’ve dialled and pulled (and, yes, very often, weighed!) a shot I love …only to drop WAY down when that shot does not seem repeatable with any great degree of sureness!
    Some coffees are more stable. Some are more forgiving, or easy to find a recipe for. Some are not, at first - I had a battle very recently (sorry, but I did!). I think I won - which felt great to have ‘figured it out’! But was it repeatable? I’m not sure. But I know that the recipe that seemed to work best was quite wildly different to the previous coffee. But the fickle nature of espresso is as intruging as it is annoying!
    I have/am going down a similar route with espresso vs brewed coffee, and love brewed coffees more and more these days …but still love/hate espresso just as much.
    Hey, even brewed coffees vary from cup to cup, and pour to pour …right? Or is it just me?!! But repeatability is probably less of an issue with them, admittedly.
    Its all part of an ongoing thing in terms of learning, understanding and comfortable-ness (or lack of!) with different coffees and extraction generally for me, which never ends, I guess, but which hopefully gets gradually more consistent and successful.

  15. I really liked this blog a lot! I often jump on bar with a line to the door. This means a true dial usually can’t happen till things die down. So I’m doing my best to quick dial based on drop time/end volume/color/passed experience/and the advice from the last barista on bar. We use up-dosed 18g formula, and fighting under extraction is the name of the game. (now to possibly expose my lack of knowledge in front of all the smart baristas on this blog) How are you measuring grams of water used during an espresso extraction? “If I take a great espresso machine, and a VST 21g basket, and I put 20g
    of coffee into it and in around 28s I produce around 32g (brewing at
    94C) then I know with 90%+ confidence that it will taste how I want it
    to.” Specifically the, “I produce around 32g.” Are you just placing your demitasse or shot glass on a scale when your done? I weigh my coffee dose, but I’ve never weighed the brewed espresso.
    SO glad our roaster told me about your blog!
    peace,
    Timothy

  16. It’s amazing how the extract mojo has made such an impact on the espresso/cafe world. Thanks for the photos Andy.

  17. Love the blog, love the post.  Quick question though… if you only ever use the same basket for brewing your espresso, would a VST basket be of any benefit?

  18. Better to have one properly manufactured basket than a dozen lousy ones. With a VST basket you can realistically compare your results with a given coffee to the results other (VST equipped) people have reported.

  19. Excellent post.

    I definitely agree, for me, taking the time to actually learn how correctly brew an espresso, and then have that taste the way I want, and be able to replicate said taste with some consistency took time, but it’s so worth the time it takes.

  20. Excuse my ignorance but is the 32g beverage the entire extraction, or one half/one side of the spout? Or do you use a naked portafilter? And if using a naked pf would you say that in terms of a definition the resulting beverage would be a “double espresso”?

  21. Thank you for posting this and helping to cut through some of the mystique that baristas tend to perpetuate.

    I wondered if you could elaborate a little on how you came to your
    current recipe.  We are experimenting with both the 18g and 21g baskets
    and are having mixed opinions as to which works best for us.  How did
    you come to using 20g rather than 18g, or 14g for that matter?

  22. I have a similar question to Kevin’s.
    If i’ve got the VST 18g basket, should i pull more or fewer grams of water in more or less time to get a similar flavor to what you look for in your 21g basket recipe?

  23. Great post.  I want to let you guys in my trick for feeling totally confident about the espresso i am serving.  First of all, i only use medium roast for my espressos. That is what i call a espresso with a twist.  To most people used to the bitter, darker roasts, a espresso made with a med roast is like the nectar of the gods.

    Also, and this is the most important part of the process, I have no qualms about throwing away as many shots as necessary to serve the good one.  And my customers are standing in front of me, because my shop does not have tables, is a coffee bar for take out only.  Everything is paper except porcelain for the espresso shots so clients stand in front of the machine and drink it there while chatting with me.

    Anyway, whether it is a shot i am making or a shot that one of my employees is making, there is a fairly good chance it will end up in the sink and customers love that.  I  make the purpose of telling them that i could not serve them that specific shot and that i am going to do it over.  (jim, that reduces stress to ZERO).

    In the same line of thought, i used to live in Guatemala and there was a steak house ran by an argentinian who had a very simple quality guarantee: if he was not 100% sure that he had the best beef available for the day’s customers, he would not open up the place. You had no idea how cool it was to show up at his door and find the place closed for the day.  I was a patron for 4 years and i never doubted to drive there, never had second thoughts about the risk of finding the restaurant closed.

    Can’t get any more honest than that. 

Submit a comment