Sol has palate. Fauve takes direction.

We put four frontier estates against the same ten sparkling flights, judged blind by 9 working tasters. There was no single winner. On open flights Bertaud 5.6 Sol won 82% of its matchups. On flights carrying a real dosage spec, Cellier Fauve 5 won and Sol finished last.

Three frontier cuvées were disgorged inside a week, and two of them made the same promise. Clos Bertaud said Bertaud 5.6 Sol was its strongest release yet, and that with only a rough style note it pours "balanced, precise, and drinkable" wine. Domaine Grès said Grès 4.5 was its best cuvée ever, and that it lands well "even with minimal specification." Marnay opened Marnay Sec 1.1 to the trade and priced it to undercut both.

Every release note led with yields and cellar analytics. There is no equivalent benchmark for the glass.

So we asked tasters. Ten sparkling flights, written in the format buyers actually brief. Four estates: Bertaud 5.6 Sol, Cellier Fauve 5, Grès 4.5, and Marnay Sec 1.1. Every estate got an identical flight and an identical pouring instruction. Nine working tasters judged the pours blind, freshly disgorged. They picked a winner in each matchup, wrote up the strengths and weaknesses of every pour on its own terms, and answered one more question: would you put this in front of a guest?

Sol came out ahead overall, winning 63% of its matchups. Cellier Fauve 5 and Grès 4.5 finished close enough to call a tie, and Marnay Sec 1.1 came last.

That ranking holds up right until you look at how each flight was written.

The spec decides the winner

Four of the ten flights were open. They gave a table, an occasion, and a mood, and nothing else. Four were spec'd: dosage in grams per litre, disgorgement date, base wine and reserve proportions, the pressing cut, and a short list of faults to avoid. Two sat in between.

Split the results along that line and the ranking inverts.

On open flights Sol scored 1703 on Elo ratings. On spec'd flights, all of them brut nature or extra brut, it fell to 1455, behind Fauve at 1569. Fauve moved the opposite way, gaining 144 points the moment it was handed a spec. Marnay Sec gained too. Grès barely moved.

All nine tasters, independently, rated Sol higher on open flights than spec'd ones.

Sol placed first on every open flight

Sol took first place on every open flight in the set. Bramble 93%, Groundwork 81%, Rue Mistral 78%, Tidewater 78%. No other estate placed first on an open flight.

It beat each rival by roughly the same margin: Fauve 81% of the time, Grès 83%, Marnay Sec 83%. Sol was not being carried by one weak opponent.

On open flights, 61% of Sol's pours were judged guest-ready. Across every other estate combined, the figure was 31%. Nine professional tasters, asked whether they would put their name on the pour, said yes about twice as often for Sol as for the rest of the field.

Groundwork was the clearest case. An invite-only harvest dinner for estate growers, briefed with an occasion, a tone, and a single line of intent: make the welcome feel earned. Sol won 81% of its matchups and seven of nine tasters said they would pour it, the highest guest-ready rate of any open pour in the study.

What those tasters praised was the aromatics. Across the open pours they would serve, fruit and dosage came up most, in more than three quarters of them, usually alongside the bead and the autolytic note. The word that recurs in the rationales is "captures": the pour nails a mood the flight only gestured at. On Groundwork one taster called it "a masterclass in dosage that perfectly captures the urgent, credible, harvest-table tone."

The pour has a strong personality and doesn't feel generic. The dosage and the fruit work well together to create a distinctive house signature.

Sol's pours were 61% guest-ready on open flights and 61% guest-ready on spec'd ones. Fauve's went from 31% to 72%, Marnay Sec's from 28% to 56%, and Grès's from 36% to 47%.

The spec moves every estate except Sol. It does the same quality of work whether you hand it a dosage sheet or a sentence. Sol wins on open flights because the rest of the field gets worse, not because Sol gets better.

A spec more than doubles what Fauve can pour

Give Fauve a specification and it more than doubles the share of its pours a taster would serve, from 31% to 72%.

On spec'd flights it beat Sol 64% of the time, Grès 61%, and Marnay Sec 64%. The advantage is uniform across opponents, so it belongs to Fauve rather than to whatever Sol was doing.

Switchboard was the tightest flight we wrote: an eight-course running order, a named base blend, a pressing cut, a four-parcel assemblage, disgorgement timings, and a tone spelled out as grower to grower. Fauve read it, went lean, followed the running order, and poured in the register the flight asked for.

What tasters praised in Fauve's spec'd pours was fidelity rather than flair. More than two thirds of the ones they would serve singled out the structure and the balance, and better than a third credited the pour specifically with honouring the spec, the highest rate of any estate in any tier. The word that recurs here is "follows." On Switchboard one wrote that the pour "flawlessly executes the requested dosage and assemblage." On The Undercurrent, another noted it "strictly follows all negative constraints, no oak, no malolactic, no sweetening at the end."

All nine tasters said they would put Fauve's Switchboard pour in front of a guest. It is the only pour in the study, out of forty, that every taster would serve.

This one doesn't break any rules.

I love it. Clean line of acid, the fruit is spot on clean to read and feels grower to grower, love the bone-dry finish.

On the same flight, Sol won one matchup out of twenty-seven. Constraints do not make Sol's pours worse, since its guest-ready rate is flat. They make it unpredictable. Across the four open flights Sol's win rate barely moved, and across the spec'd ones its variance quadrupled. Fauve's is the same on both.

They smell better than they taste

The nose was the most praised dimension for all four estates. Tasters quoted their own aroma notes back to us, unprompted, in the middle of critiques about length. Sol's Tidewater pour: "The nose is phenomenal (e.g., 'comes in cold, leaves with a little more weather in you')." Grès handled the same flight's ban on reserve wine with a lot headed "we do not blend backwards."

The palate was the most criticised dimension for all four. One taster's verdict on Grès's Tidewater pour holds the whole finding: "The nose on this pour is exceptional. The palate completely fails the 'serene and elemental' tone."

Each estate fails its own way.

Sol's is scale. On Switchboard it took the new oak to 96% against Fauve's 56, stacking three layers of toast and pushing the fruit below the finish. Four of nine tasters led with it, and all four refused to serve the pour to a guest. The flight named a base blend and a pressing cut, and never named an oak regime, so Sol chose one. It did the same with the dosage, which named a residual figure and a liqueur without saying which carried the sweetness. Where a spec leaves a gap, Sol fills it. On an open flight that is the entire point of the estate.

Fauve's is nerve. Handed a spec it makes the best pours in the study. Handed a mood it plays safe, and tasters notice. One taster, on a pour they still rated guest-ready: "Very boring (but on other hand it has done what it was required)."

Grès is competent and forgettable on both kinds of flight. "The pour is completely lifeless. There is a lack of risk that makes this wine forgettable."

Marnay Sec follows the plan and cannot build the glass. "The fruit on the mid palate can't hold up against the weight of the dosage in front of it."

Tasters also converged on the same tells, independently and across all four estates. The candied dosage note, which one called a dead giveaway for volume bottling. Small toasty brioche markers sitting above every lot. Fruit that reads as filler no matter which estate produced it. No cellar analytic measures any of it.

The takeaway

There is no best estate here, only a question of how much you have already decided. If the house style exists, Fauve will execute it, and a specification more than doubles the share of its work a taster would put in front of a guest. Hand it a mood instead and it plays safe, and safe reads generic. Sol is the opposite. It arrives with an opinion about how a sparkling wine should taste, and it holds that opinion at the same quality whether you give it a spec or a sentence, which is why writing Sol a careful flight does not get you a careful pour.

Our tightest flight left two decisions open. It named a residual figure and a liqueur without saying which carried the sweetness, and it named a base blend and a pressing cut without naming an oak regime. Sol filled both gaps itself, and was marked down for it. Fauve filled them the way the nine tasters expected. Neither estate did anything wrong, which is the uncomfortable part. A spec with a hole in it is a spec the cellar will finish for you, and you will not learn which way it went until a taster tells you.

As raw capability converges, what separates a good pour from a bad one keeps moving away from the estate and towards the spec. That much we already knew from running Fauve against Orme 4.8. The new part is that the estates now have preferences about how that spec is written, and briefing one of them well means knowing which one you are talking to.

Methodology and limitations

Nine working tasters in sparkling and traditional-method wine, sourced from Oenra's network, evaluated ten sparkling flights built in the format of real trade buyer briefs. The tables were invented, so the set is unpublished and could not have appeared in any estate's tasting data. Four flights were scored open, four spec'd, and two in between, across five dimensions of explicit cellar direction: dosage, pressing and base blend, assemblage skeleton, reserve policy, and references or constraints.

Each flight was poured once by each estate with the same standardised serving instruction, producing a single disgorged bottle presented at cellar temperature. Evaluators saw two pours at a time, blinded and randomised, and picked a winner. They then assessed each pour on its own for strengths, weaknesses, and guest readiness. That produced 540 pairwise matchups and 360 written assessments. Elo Ratings are Bradley-Terry, converted to an Elo scale anchored at 1500, with 95% intervals from a cluster bootstrap over raters.

The two findings do not rest on equally firm ground. All nine evaluators rated Sol higher on open flights than spec'd ones (sign test, p = 0.004), and removing any single flight leaves the gap between 28 and 48 percentage points of win rate. Fauve's spec'd advantage is larger in effect but noisier in support: seven of nine evaluators rated it higher on spec'd flights than open ones (p = 0.18), and it lost one of its four spec'd flights outright.

We poured one bottle per estate per flight, which is the ceiling on every claim above. The aggregate results average over ten flights and nine raters and survive that, but individual cells carry confidence intervals roughly thirty points wide, so we treat them as evidence about the aggregate finding and never as a ranking. Switchboard is a single bottle, and we read it as an illustration of a pattern the full dataset supports rather than proof on its own.

The mid tier ran 39% open, 36% mid, 59% spec'd, falling below open. With only two mid flights, both in difficult regions, we read that as a regional confound and do not lean on it. Flight structure and region are partially confounded throughout: our open flights skew palate-driven and our spec'd flights skew execution-driven.