🍩 Phish Setlist Tracker
← 2026-08-01 · FENWAY PARK 2026-09-05 · DICK'S SPORTING GOODS PARK →

Setlist prediction — 2026-09-04

Dick's Sporting Goods Park · Commerce City · Summer Tour 2026 · based on 2002 prior shows · 0 recently-played songs rotation-suppressed · model v3.4.1 · actual setlist known

Predicted vs. actual

Predicted

Set 1: 46 Days [8:54] ★31% · The Moma Dance [8:36] ★20% · Runaway Jim [7:38] ★19% · Cities [16:17] ★44% · Bathtub Gin [15:44] ★48% · 555 [7:16] ★18% · Reba [12:39] ★20% · Dirt [5:59] · Most Events Aren't Planned [8:49]

Set 2: Harry Hood [13:42] ★25% · Fuego [16:52] ★49% · No Men In No Man's Land [10:52] ★29% · Blaze On [10:27] · Tweezer [10:44] ★53% · Piper [12:20] ★43% · Everything's Right [12:34] ★25%

Encore: First Tube [7:17] · Tweezer Reprise [3:22]

Actual — DICK'S SPORTING GOODS PARK

Set 1: No Men In No Man's Land [9:38], Oblivion [16:01], Prince Caspian [9:20] -> Plasma [14:41], Tube [8:16], Ya Mar [8:21], Wolfman's Brother [13:20]

Set 2: 46 Days [6:40] > Down with Disease [35:18] > Life Saving Gun [18:26], Lonely Trip [6:46], Run Like an Antelope [10:59] > Character Zero [5:59]

Encore: Wading in the Velvet Sea [6:59], Harry Hood [16:34]

Score vs. what they actually played

30.0%
Better than this share of nights
11.46/100
Raw score · median 19.3
-6.2%
Skill vs. baseline
16.65
Baseline score
3.02 bits
Difficulty

The raw score is graded against a perfect setlist call, which never happens. Phish rotate deliberately and go out of their way to be unpredictable, so 100 is unreachable — over the last 60 shows the score runs from about 2 to 42.09, with a median near 19.3. That is why the headline ranks this night against the ones actually achieved rather than against an imaginary ceiling. The number that means something is Skill: how much of the headroom above a baseline that just guesses the most-played songs this prediction captured. 0% is “no better than guessing the hits”; anything above it is real signal. On this night it came out -6.2%: the model did worse than the baseline, i.e. its cleverness actively cost it against simply guessing the hits. Difficulty is the average rarity of the songs they actually played (3.02 bits) — a night of deep cuts scores lower for the same skill, which is why the headline is normalised against a perfect prediction of this show.

Sub-scores

17.31
selection
Right songs, ignoring where they landed
0
set match
Right song in the right set
0
proximity
Right song, right set, near the right slot

Each is out of 100 and rarity-weighted: picking the song at all earns the most, putting it in the right set earns more, landing it near the right slot earns the rest.

Precision & recall

MetricValueWhat it means
precision 16.7%
Of what it predicted, how much got played
recall 20.0%
Of what got played, how much it predicted
f1 18.2%
Harmonic mean of the two
precision (rarity-weighted) 15.0%
Precision, weighted by how rare each song is
recall (rarity-weighted) 17.3%
Recall, weighted by how rare each song is
top5 hit rate 0.0%
Of the top-5 shortlist, how much got played

18 songs predicted, 15 played. The rarity-weighted variants are the interesting ones: calling a warhorse is easy, calling a bustout is not, and the weighted numbers pay out accordingly.

Sequence

Adjacent pairs called right 0 of 15 predicted
Adjacent pairs actually played12
Longest correct run 1 song

Getting two songs adjacent in the right order is much harder than getting both somewhere in the show.

Structure

SetPredictedActual
Set 197
Set 276
Encore22
Total18 15

Biggest win

46 Days

The rarest song it called correctly — 3.061 bits (common).

Biggest miss

Ya Mar

The rarest song it didn't see coming — 4.368 bits (uncommon).

By rarity

How it did across rarity bands. The warhorses are supposed to be easy; the deep cuts are where a model earns its keep.

TierPlayedCalledHit rate
staple41 25.0%
common72 28.6%
uncommon40 0.0%
rare00 —
bustout00 —

Song by song

Everything they played, rarest first.

SongRarityTierCalled?Right set?Credit
Ya Mar 4.368 bits uncommon miss — 0.0
Lonely Trip 4.094 bits uncommon miss — 0.0
Wading in the Velvet Sea 4.047 bits uncommon miss — 0.0
Plasma 3.71 bits uncommon miss — 0.0
46 Days 3.061 bits common hit no 0.6
Life Saving Gun 3.039 bits common miss — 0.0
Oblivion 2.94 bits common miss — 0.0
Prince Caspian 2.731 bits common miss — 0.0
Wolfman's Brother 2.686 bits common miss — 0.0
No Men In No Man's Land 2.599 bits common hit no 0.6
Tube 2.575 bits common miss — 0.0
Down with Disease 2.475 bits staple miss — 0.0
Run Like an Antelope 2.406 bits staple miss — 0.0
Character Zero 2.325 bits staple miss — 0.0
Harry Hood 2.166 bits staple hit no 0.6

Predicted but not played

Dirt · 4.3 bits 555 · 3.865 bits Cities · 3.826 bits Most Events Aren't Planned · 3.648 bits Piper · 3.28 bits Reba · 3.007 bits Runaway Jim · 2.878 bits First Tube · 2.756 bits The Moma Dance · 2.752 bits Bathtub Gin · 2.734 bits Fuego · 2.537 bits Blaze On · 2.458 bits Tweezer Reprise · 2.439 bits Everything's Right · 2.436 bits Tweezer · 1.3 bits

Reported, but never subtracted from the headline: penalising bold calls by rarity would teach the model that rare songs are never worth predicting, which is the opposite of what this is for.

Predicted by model v3.4.1 · scored by v3.4.1 rules.

Top 5 most likely songs

  1. Tweezer + Tweezer Reprisein heavy recent rotation; due back by gap; gap 0 (usual ~5)
  2. First Tubedue back by gap; in heavy recent rotation; gap 8 (usual ~4)
  3. Blaze Onin heavy recent rotation; due back by gap; gap 7 (usual ~5)
  4. The Moma Dancedue back by gap; in heavy recent rotation; gap 8 (usual ~4)
  5. Bathtub Ginin heavy recent rotation; played on this date before; gap 5 (usual ~5)

Predicted full setlist

Set 1: 46 Days [8:54] · The Moma Dance [8:36] · Runaway Jim [7:38] · Cities [16:17] · Bathtub Gin [15:44] · 555 [7:16] · Reba [12:39] · Dirt [5:59] · Most Events Aren't Planned [8:49]

Set 2: Harry Hood [13:42] · Fuego [16:52] · No Men In No Man's Land [10:52] · Blaze On [10:27] · Tweezer [10:44] · Piper [12:20] · Everything's Right [12:34]

Encore: First Tube [7:17] · Tweezer Reprise [3:22]

Ordering is heuristic (openers/encore placed by role); treat it as a likely song pool with structure, not an exact sequence.

Predicted lengths & jam odds

SongSetLikely lengthRange (q10–q90)P(jam chart)
46 Days Set 1 8:54 7:21 – 13:13 31.1%
The Moma Dance Set 1 8:36 7:07 – 11:47 20.1%
Runaway Jim Set 1 7:38 6:56 – 12:18 19.3%
Cities Set 1 16:17 6:12 – 19:05 44.4%
Bathtub Gin Set 1 15:44 12:01 – 20:50 47.6%
555 Set 1 7:16 5:35 – 10:46 18.5%
Reba Set 1 12:39 11:40 – 13:44 20.0%
Dirt Set 1 5:59 2:11 – 12:18 6.2%
Most Events Aren't Planned Set 1 8:49 8:01 – 9:17 5.6%
Harry Hood Set 2 13:42 11:50 – 18:06 24.7%
Fuego Set 2 16:52 10:13 – 25:44 49.4%
No Men In No Man's Land Set 2 10:52 6:12 – 15:03 28.7%
Blaze On Set 2 10:27 7:17 – 14:15 15.0%
Tweezer Set 2 10:44 2:19 – 29:21 53.3%
Piper Set 2 12:20 7:13 – 16:14 42.8%
Everything's Right Set 2 12:34 9:44 – 16:49 25.2%
First Tube Encore 7:17 6:31 – 8:04 0.8%
Tweezer Reprise Encore 3:22 3:05 – 5:38 3.4%

Lengths are recency-weighted medians conditioned on the set — the q10–q90 band is where the honesty lives. P(jam chart) is the chance curators later flag this version; it is deliberately estimated from curated history only, since the last year of shows hasn't been written up yet.

Bustout watch

Train Song · gap 97 Cracklin' Rosie · gap 144 Brother · gap 229 Poor Heart · gap 81 My Sweet One · gap 77

Model backtest: top-5 hit rate 39.5%, setlist recall 31.4% over 40 shows. Full scoring history →