"Never Miss a Sunday Show" Is Half True.

“Never miss a Sunday show” is one of the most persistent pieces of lore in the Phish community. Zoom out far enough on the modern data, and the folklore looks entirely true: since 2009, Sunday shows lead every other day of the week in both community rating and composite show score. (Pre-2009, Sundays sat second-to-last of seven days by median composite, 55.6 vs Saturday’s 66.1.)

That’s the answer most fans would accept. It’s also the wrong test.

Still, let’s examine the claim at face value. Here’s every scored show since the 2009 return, grouped by day of week.

Here’s the problem: look at what a “Sunday show” is in the tour calendar. Since 2009, 27 tours have run 12 shows or more: long enough to cover several weeks and most days of the calendar. Here’s the day each of them ended on.

To find out whether Sundays themselves deserve the reputation, the scheduling effects have to come off one at a time.

Start with special-event weekends. Restrict to shows played on 12-or-more-show tours (the long summer, fall, and winter runs). That threshold does the exclusion work on its own: festivals, NYE runs, Mexico trips, Sphere residencies, reunion runs, MSG spring residencies, and the Vegas Halloween runs are all short enough to drop out automatically.

A cleaner test compares within tours. Many tours include multiple weekend runs, some ending on Sunday and some ending on other days. Filter to tours that had both types of run-closer, and compare inside each tour. Same band, same tour form, same era.

One control this test doesn’t apply: tour position. Sunday run-closers fall disproportionately late in tours, often as the tour’s final show (per Fig 2), while Mon-Sat run-closers land more evenly. Some of the residual could be a final-night effect rather than a Sunday effect.

One more cut sharpens the picture: run length.

Almost the entire premium lives inside 3-night runs.

The special night isn’t Sunday. It’s the Sunday closer of a Fri-Sat-Sun weekend run. That’s the slice carrying the folklore.

The composite score has six components: community rating (40% weight), average track score (15%), phish.in likes (15%), jam-chart flags (10%), segues (10%), depth (10%). Splitting the same-tour paired comparison into those components shows where the ~11.0-point gap lives.

The audience-response signal runs about five times the raw playing lift. Sunday-closer nights end up remembered more warmly than the playing on its own would explain.

One caveat: this measurement can’t distinguish between three explanations. Sundays might be experienced as better (real amplification of the on-stage difference); the rater pool on Sundays might be more invested (weekend travelers, run-of-a-lifetime attendees, people who paid more to be there), and rate accordingly; or the saying itself might be generating its own evidence, fans primed by the folklore rating the show more warmly because they were told to expect one. The data can’t tell the three apart. All three would produce the pattern we see.

The playing lift is real, though, and it has a specific driver. Sunday run-closers get measurably more long-jam minutes per show.

Split those long-jam minutes by set and the “Sunday Set II” folklore holds up. Sunday closers deliver +3.3 minutes of 10-minute-plus jam time in Set 2 over Mon-Sat closers (20.8 vs 17.5), and +1.3 minutes in Set 1 (5.0 vs 3.7). Both sets get a real lift; Set 2 gets about double the absolute minutes. Set 1 is proportionally more affected, but off a much smaller base. The classic “Sunday Set II” instinct is directionally right, even if it’s not the whole picture.

The extra jam time has two mechanisms. The first is song choice: some songs simply get picked more often on Sunday closers. YEM is the case in point — 19 Sunday-closer plays vs 8 Mon-Sat plays. Once chosen, YEM runs its full composed arc regardless of day.

Per-play jam length is more interesting, but requires a careful measure. Total track duration is a bad proxy because songs like YEM, Divided Sky, and Fluffhead are naturally long from their composed sections alone: a 20-minute YEM might contain zero minutes of unusual improv. So the table below uses the JAM portion: track duration minus the hand-tagged composed-section length. That reveals a real spread: some songs stretch further on Sundays, others do the reverse.

The premium is an average. Not every Sunday clears it. 66 Sunday run-closers on regular tours since 2009; the worst quartile medians 45.1 on composite score. Here are the twelve lowest, with the position each show holds inside its run.

Splitting all 66 Sunday run-closers into quartiles by composite score and comparing the worst 17 to the best 16 surfaces which features separate the great Sundays from the flat ones.

The saying survives, but narrower than fans mean it. What the data supports:

Don’t miss the Sunday closer of a 3-night run on a regular tour. That’s where the modern-era premium concentrates. Most of the composite gap comes through on the audience-response side rather than raw playing, but part of it is a real extra long jam or two. Sundays that aren’t part of a 3-night run, or that push a run to a fourth night, don’t clear the same bar.

The folklore had the right night. It was just measuring the crowd as much as the band.