Tuesday, January 2, 2024


Music Week: Jazz Poll

Arts Fuse published the 18th Annual Francis Davis Jazz Critics Poll Tuesday. Specifically, two pieces:

The former offers an introduction to the Poll and a brief overview of the year in jazz, followed by a table listing the year's top 50 new jazz releases. The latter introduces tables of the top finishers in four more categories: Reissues/Historical (which Davis likes to call Rara Avis, but let's dispense with that here and just go with Old), Debuts, Latin Jazz, and Vocal Jazz. Both pieces were fairly minimal, at least compared to the long essays of previous years. For instance, in 2022 an ailing Francis Davis wrote My Poll Without Me, and I submitted five pieces, which ArtsFuse reduced to:

Most of the previous years' essays are available on the Jazz Poll website (but they need to be better organized; the index page only lists two early ones, but we have more). It looks like I started hosting ballots in 2009, and developed the programming for my modern system in 2011, but it was much later before Davis finally trusted me to do the counting. Finally, two years ago health crises led him to turn the whole Poll over to me (although he continues to monitor everything, and contribute some writing).

For several years now, I've been considering the possibility that I should radically cut back on my music writing. I have other book projects I've long wanted to pursue, and I felt like I was losing my touch, my patience, and my passion. Still, by October I realized that I once again had failed to quit, so I resolved to run another Poll. And with some lead time for once, I wanted most of all to reverse the decline in votership under my watch. (I know, 156 to 151 wasn't bad, but I felt we could do better.) For the first time ever, I started to make a systematic search for possible voters. I learned some things, which I should write up at some point, but didn't add anywhere near as many names as I had hoped for.

But I did manage to send out over 200 invites on or near Nov. 18, and I tried contacting another 30-40 up to (and in a couple cases even past) the Dec. 15 voting deadline. That last-minute hustle put us over the top, with 159 ballots finally counted. And I'm very proud of those additions -- probably because I've vetted them personally, but the more familiar I become with our veteran voters, the more impressed I became with them, too.

But then, results in hand, I got stuck in trying to write up one of those "state of the union" essays that seemed to be obligatory for polls of such ambition. Fortunately, Davis saved the day by writing his own, allowing me to duck the task. One reason I had so much trouble writing an essay this year is that I had become more fascinated with the voters than with the musicians, who throughout the year we are expected to serve (and to some extent revere). The big questions for me weren't along the lines of who's clever enough to win the next MacArthur, and whether that person will repay the acclamation with the genius promised? I'm more curious about how people searching for jazz -- music that is both pleasurable and intellectually stimulating -- find records and share understanding. In other words, I might as well admit it, people more like me. If you take money out of the equation -- and, let's face it, few arts have done so more completely than jazz -- the only difference between fans and critics is that the latter have crossed the boundary of fanaticism: we search so hard, so deep, and so far that we have started to build up our own information networks.

I wish I had a big pile of statistics on the voters -- not so much the usual demographic crap, which risks being generalized into stupidity, but stuff like how many promos do you get and listen to, how much streaming you do, how much radio you listen to, what the split is between jazz and other music, and what other kinds of music do you like or hate. I also have questions about money, but that's mostly to confirm my suspicion that everyone involved with jazz is giving up some amount of opportunity cost to do so.

This would turn in to a ridiculously long (and no doubt boring) essay if I stopped to explain why each of these considerations matter. But several things have changed in the electorate this year. For one thing, we have more independent bloggers (including a couple who mostly write about pop/rock/etc. but who cover enough jazz to impress me), while we have fewer people writing for the recently-defunct JazzTimes. We also have more non-Americans, including some from Asia and Latin America. One consequence is probably that the linkage between publicists and voters -- I wrote a fairly long and controversial piece about this back in 2021 -- has been weakened. One piece of evidence I noticed is that Blue Note fared very poorly this year -- although it turns out that 2019 was a similarly bad year for them, so it's not exactly q.e.d. Another curious turn is that a self-released album with no publicity that I'm aware of -- I've sunk beneath Blue Note's radar, but I'm generally pretty aware of who's hyping what -- came in second place.

Granted, it's by a previous Poll winner, Jason Moran, but it's an order of magnitude more than any of his other self-released albums have done. As for this year's top winner, James Brandon Lewis, Tao Forms may seem like a tiny label, but its publicity was handled by Fully Altered, one of the most effective independents anywhere. Plus, like Moran's album, it's really, really good -- number four on my ballot. Publicists may help and no publicity may hurt, but in a world where payola has little sway, quality is essential. (I wouldn't say that payola never works on jazz critics, but simply that there's never been enough of it to go around. Mosaic won the Reissues category four out of five years back (2008-12) when they were giving free copies of their expensive boxes to major critics, but hasn't come close as the electorate grew and their promo budget shrank. Legacy won four times with Miles Davis boxes 2007-15, but not since, and not really because they've scraped the bottom of that barrel.)

So where Francis Davis sees the top-tens forming some kind of jazz super-elite, shifting slightly but mostly steady over the years, I wonder whether they aren't just some relatively common center to an increasingly dispersed jazz universe. It's clearly not a random function: to get into the top ten, you first have to be heard, so you have to be someone critics want to hear. And you have to be very good, to stand out from all the other things people hear. But "good" varies a lot by critic, so what matters is how broad a segment of voters your record really appeals to. This year, the magic number to enter the top five is 30-32 votes, which is to say about 20% of the voters. Lewis won with 47 votes, so 29.5%, down a bit from his 33.9% in 2021, but about average for winners since 2013. The lowest voter share was 24.2% in 2018, for Wayne Shorter. Only in 2006-07, when we only had 30 voters, did we have majority winners: Ornette Coleman and Maria Schneider, two exemplars of the emerging postbop order.

This is not to say that many casual jazz fans recognize, or fully appreciate, this new postbop order. But most jazz critics do, and it's part of what makes us who we are -- the sense of a long and continuously evolving form of art. One thing that used to puzzle me about the Poll is that we while we collected ballots from at least half of the critics who voted in JazzTimes's annual poll, our much larger electorate voted significantly farther out than theirs. Normally, you'd expect larger samples to tend toward the median, but the exact opposite happened with us. Even if we had invited the other half, it wouldn't have made any difference. What I suspect now is that their polls -- indeed, virtually all polls from brand-name publications -- are tailored to their business plan, which like most businesses is a supply-side plot to push product.

Our Poll is very different: it mostly represents the demand side, that of ravenous but savvy consumers. That's why it is very important that we list every album that got voted for by anyone. Transparency, and accountability, hallmarks of free markets, features that M.B.A. programs are designed to eradicate. My biggest perk in compiling the Poll is that I get to look at lists as they come in, which inevitably send me scurrying off in search of unheard records.

The New Releases list reached 535 albums this year. As they came in, some 20-30% were records I hadn't heard, or in most cases, heard of. Another 128 records appeared in the Old list, which includes reissues and previously unreleased music recorded more than 10 years ago (2012 or earlier). We also have separate lists for the special categories: Debuts, Latin, and Vocal. Some records only appear in those lists, as they were created in the first place to recognize records that tended to get overlooked in the top-ten lists.

I do a fairly good job of tracking what I've managed to hear in any given year, as you can see from my jazz tracking file. At the moment, I have listened to, at least well enough to rate, 865 jazz albums this year (including reissues). That may seem like a lot, but it still leaves 182 new releases and 51 reissues/historical that got votes in the Poll that I haven't heard (34% and 39.8%). I don't know how many jazz albums others manage to hear, but I'm skeptical that many others are hearing as many, let alone more. (Adding in non-jazz, my total adding is 1458, so I listen to about 60% jazz, 40% non-jazz. More jazz focus is quite possible, but more hours are hard to come by.)

I keep two end-of-year files, one for jazz, the other for non-jazz. The former, at present, lists 80 new jazz albums graded A or A-, followed by 205 new jazz albums rated B+(***), and many more (454) with lesser grades. For quite possibly the first time ever, every one of my top-ten picks got at least one additional vote. The only album that exclusively appeared on my ballot was my vocal pick: Lisa Marie Simmons & Marco Cremaschini, NoteSpeak 12. I suppose this shift could be taken as proof that I've been rigging the electorate with friends and allies, but we're talking about the very margins of the Poll. And some, like Chris Monsen, I've been following for ages. (He was the first person I lobbied Davis to invite, and one of the first Europeans to join us.)

As an exercise, I thought I'd list my A-list jazz albums (out of 80) that didn't get votes in the Poll. Just as I failed to hear 30-40% of the records on their lists, I imagine that few other voters managed to hear many of these. To me, this just proves that the breadth of high-quality jazz far exceeds the grasp of even the most dedicated fans and critics (* indicates that one or more of the artists had one or more other albums that did receive votes):

  1. Mark Feldman/Dave Rempis/Tim Daisy: Sirocco (Aerophonic) *
  2. Ivo Perelman/Ray Anderson/Joe Morris/Reggie Nicholson: Molten Gold (Fundacja Sluchaj) *
  3. Floy Krouchi/James Brandon Lewis/Benjamin Sanz: Cliffs (Off '22) *
  4. Ivo Perelman/Dave Burrell/Bobby Kapp: Trichotomy (Mahakala Music) *
  5. Das Kondensat: Andere Planeten (WhyPlayJazz)
  6. Izumi Kimura/Gerry Hemingway: Kairos (Fundacja Sluchaj) *
  7. Margherita Fava: Tatatu (self-released)
  8. Elijah Shiffer: Star Jelly (self-released)
  9. Kaze & Ikue Mori: Crustal Movement (Libra) *
  10. Daniel Bingert: Ariba (Moserobie)
  11. Andrea Veneziani: The Lighthouse (self-released)
  12. Dave Bayles Trio: Live at the Uptowner (Calligram)
  13. Michael Jefry Stevens Quartet: Precipice (ARC)
  14. Don Fiorino/Andy Haas: Accidentals (Resonant Music)
  15. Chad McCullough: The Charm of Impossibilities (Calligram)
  16. Benjamin Herman: Nostalgia Blitz (Dox)

One thing that happens a lot on the free jazz end of the spectrum is that votes get scattered, especially artists who release a lot of albums. Ivo Perelman received 42.5 points, but they were scattered over six releases, while missing two from my list. Similarly, Satoko Fujii (Kaze above) netted 43.5 points from three albums, Joe McPhee 26.7 from four, Dave Rempis 17 from three, Nate Wooley 15 from three, Paul Dunmall 11.5 from three. Most prolific of all is John Zorn, who got 67.5 points from seven albums. Adding more free jazz partisans to the voter rolls increases this spread of albums, without having much (if any) effect on the top 10, 20, 30, etc.

My list also includes 22 A/A- albums of older jazz, 34 at B+(***), and 41 graded lower. The main thing to note here is that my list almost doubled in length after the ballots started coming in, with many records -- starting with the Roy Hargrove on top -- that I wasn't aware of. That left only a few in my A-list that no one voted for in the Poll:

  1. Alon Nechushtan: For Those Who Cross the Seas (2006, ESP-Disk)
  2. Albert Ayler: Summertime to Spiritual Unity Revisited (1964, Ezz-Thetics)
  3. JuJu: A Message From Mozambique (1972, Strut)

Since I'm down into the Old here, one thing I want to note (well, get off my chest) is that one long-time voter refused to participate because we only allow three picks in the category. He had many more he wanted to mention -- obviously I, and most likely many others, were of that same mind -- but more importantly he felt that allowing 10 new music albums but only 3 old music implied that the new stuff was more important (better) than the old. He obviously believed otherwise, strongly enough to deny us (if I may pull a Trump, let's make that you) his expert opinion.

Unless you can identify the missing critic, his protest was in vain. I pointed out that he could have actually expressed it by voting for 3 old and 0 new albums -- that's been done before, but the opposite is much more common; it's also much more common to only vote for one or two old than it is to vote for less than ten new releases. That's actually one of several arguments I can make that cast doubt on his assertion (not so much about the superiority of old music as the need for more poll slots for it. It also matters that the total number of new records is much greater than the number of reissue/archival records. My tracking file divides them using the same rule we use in the Poll. There I have 210 old music releases, vs. 1257 new music, a ratio of 5.985 to 1 (or almost double the 3.3-to-1 ratio of poll slots).

Also, if you look at the Old results, and compare them to New, you should notice a couple peculiarities. One is that the Old winners get many more votes than the New winners: this year's Coltrane album got votes from 41.5% of the ballots, vs. 29.5% for Lewis in winning New. That's partly a fluke -- after all, we're talking Coltrane, and Cecil Taylor didn't run up nearly that kind of margin in 2022 (Taylor actually won on points, receiving only 38 votes to 44 for runner up Charles Mingus, or 29.1%). But if we look at the decline from third place (22 votes this year) to 10th (8), you've dropped 63.7%. In New this year, the decline is from 30 votes to 24, a much less steep drop of 20.0%. Some of this can be attributed to having fewer slots in Old, but not nearly all of it.

We all should thank Davis for having the vision of putting this Poll together, for relentlessly expanding its focus, and for sustaining it even through the collapse of any institutional support. It contributes significantly to what we know, and can know, about the world of jazz, a world that matters greatly to those of us fortunate enough to tune into it. I am simply thankful that he let me in to help out in whatever small ways I could, especially in these last couple extremely difficult years.

But once the counting is done, the results posted, and the notices given, here I am again, back to basics: deep down, I'm a sociologist, a philosopher, an engineer, and above all a critic. So let's dig through this data and learn something . . . and have some fun doing it.

Ask a question, or send a comment.