Setlist prediction — 2026-10-07
Allianz Amphitheater at Riverfront · Richmond · 2026 Fall Tour · based on 2009 prior shows · 46 recently-played songs rotation-suppressed · model v3.4.1 · actual setlist known
Predicted vs. actual
Predicted
Set 1: Cars Trucks Buses [5:37] · AC/DC Bag [9:07] ★29% · Evolve [6:44] · Theme from the Bottom [10:25] ★24% · Tube [8:15] ★26% · Reba [12:39] ★24% · Hey Stranger [9:00] ★24% · Walls of the Cave [11:33] ★17%
Set 2: Sample in a Jar [5:06] ★25% · Twist [9:48] ★31% · Tweezer [14:43] ★56% · Golden Age [11:57] ★37% · What's the Use? [7:15] · Ruby Waves [16:16] ★48% · Backwards Down the Number Line [7:57]
Encore: A Life Beyond The Dream [7:58] · Miss You [6:56]
Actual — Allianz Amphitheater at Riverfront
Set 1: Gumbo [15:44], The Curtain With [13:27], Hey Stranger [8:11], Llama [5:01], Roggae [10:19], Tube [8:41] > Timber (Jerry the Mule) [8:59], Most Events Aren't Planned [9:37]
Set 2: Kill Devil Falls [15:32] > Ruby Waves [23:29] > Golden Age [26:01] > Meatstick [6:11] > Backwards Down the Number Line [6:41] > Loving Cup [7:01]
Encore: The Lizards [9:40], Julius [7:44]
Score vs. what they actually played
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 -1.3%: 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 (3.42 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
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
| Metric | Value | What it means | |
|---|---|---|---|
| precision | 29.4% | Of what it predicted, how much got played | |
| recall | 31.2% | Of what got played, how much it predicted | |
| f1 | 30.3% | Harmonic mean of the two | |
| precision (rarity-weighted) | 25.8% | Precision, weighted by how rare each song is | |
| recall (rarity-weighted) | 27.3% | Recall, weighted by how rare each song is | |
| top5 hit rate | 20.0% | Of the top-5 shortlist, how much got played |
17 songs predicted, 16 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 14 predicted |
| Adjacent pairs actually played | 13 |
| Longest correct run | 1 song |
Getting two songs adjacent in the right order is much harder than getting both somewhere in the show.
Structure
| Set | Predicted | Actual |
|---|---|---|
| Set 1 | 8 | 8 |
| Set 2 | 7 | 6 |
| Encore | 2 | 2 |
| Total | 17 | 16 |
Biggest win
The rarest song it called correctly — 3.636 bits (uncommon). Right set, too.
Biggest miss
The rarest song it didn't see coming — 4.11 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.
| Tier | Played | Called | Hit rate | |
|---|---|---|---|---|
| staple | 0 | 0 | — | |
| common | 6 | 4 | 66.7% |
|
| uncommon | 10 | 1 | 10.0% |
|
| rare | 0 | 0 | — | |
| bustout | 0 | 0 | — |
Song by song
Everything they played, rarest first.
| Song | Rarity | Tier | Called? | Right set? | Credit |
|---|---|---|---|---|---|
| The Curtain With | 4.11 bits | uncommon | miss | — | 0.0 |
| Meatstick | 4.041 bits | uncommon | miss | — | 0.0 |
| Loving Cup | 3.902 bits | uncommon | miss | — | 0.0 |
| Timber (Jerry the Mule) | 3.653 bits | uncommon | miss | — | 0.0 |
| The Lizards | 3.649 bits | uncommon | miss | — | 0.0 |
| Hey Stranger | 3.636 bits | uncommon | hit | yes | 0.75 |
| Julius | 3.632 bits | uncommon | miss | — | 0.0 |
| Gumbo | 3.575 bits | uncommon | miss | — | 0.0 |
| Llama | 3.574 bits | uncommon | miss | — | 0.0 |
| Most Events Aren't Planned | 3.542 bits | uncommon | miss | — | 0.0 |
| Roggae | 3.322 bits | common | miss | — | 0.0 |
| Golden Age | 3.029 bits | common | hit | yes | 0.821 |
| Kill Devil Falls | 2.844 bits | common | miss | — | 0.0 |
| Ruby Waves | 2.804 bits | common | hit | yes | 0.75 |
| Backwards Down the Number Line | 2.793 bits | common | hit | yes | 0.791 |
| Tube | 2.665 bits | common | hit | yes | 0.808 |
Predicted but not played
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
- Twistdue back by gap; in heavy recent rotation; gap 23 (usual ~4)
- A Life Beyond The Dreamin heavy recent rotation; due back by gap; gap 8 (usual ~5)
- Tubein heavy recent rotation; due back by gap; gap 6 (usual ~5)
- Evolvedue back by gap; in heavy recent rotation; gap 20 (usual ~4)
- Rebadue back by gap; in heavy recent rotation; gap 13 (usual ~7)
Predicted full setlist
Set 1: Cars Trucks Buses [5:37] · AC/DC Bag [9:07] · Evolve [6:44] · Theme from the Bottom [10:25] · Tube [8:15] · Reba [12:39] · Hey Stranger [9:00] · Walls of the Cave [11:33]
Set 2: Sample in a Jar [5:06] · Twist [9:48] · Tweezer [14:43] · Golden Age [11:57] · What's the Use? [7:15] · Ruby Waves [16:16] · Backwards Down the Number Line [7:57]
Encore: A Life Beyond The Dream [7:58] · Miss You [6:56]
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
| Song | Set | Likely length | Range (q10–q90) | P(jam chart) |
|---|---|---|---|---|
| Cars Trucks Buses | Set 1 | 5:37 | 5:29 – 6:09 |
5.4%
|
| AC/DC Bag | Set 1 | 9:07 | 5:44 – 11:57 |
29.4%
|
| Evolve | Set 1 | 6:44 | 5:50 – 9:21 |
2.5%
|
| Theme from the Bottom | Set 1 | 10:25 | 8:24 – 13:35 |
23.8%
|
| Tube | Set 1 | 8:15 | 6:09 – 11:51 |
25.9%
|
| Reba | Set 1 | 12:39 | 11:32 – 13:44 |
23.7%
|
| Hey Stranger | Set 1 | 9:00 | 6:56 – 10:29 |
24.0%
|
| Walls of the Cave | Set 1 | 11:33 | 10:46 – 12:28 |
16.6%
|
| Sample in a Jar | Set 2 | 5:06 | 3:47 – 14:05 |
25.4%
|
| Twist | Set 2 | 9:48 | 7:43 – 14:31 |
31.4%
|
| Tweezer | Set 2 | 14:43 | 2:25 – 29:41 |
55.9%
|
| Golden Age | Set 2 | 11:57 | 6:52 – 18:44 |
37.1%
|
| What's the Use? | Set 2 | 7:15 | 6:06 – 8:38 |
7.6%
|
| Ruby Waves | Set 2 | 16:16 | 11:44 – 29:25 |
47.6%
|
| Backwards Down the Number Line | Set 2 | 7:57 | 6:51 – 9:21 |
13.2%
|
| A Life Beyond The Dream | Encore | 7:58 | 7:10 – 8:24 |
0.9%
|
| Miss You | Encore | 6:56 | 5:47 – 7:50 |
3.3%
|
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
Model backtest: top-5 hit rate 39.5%, setlist recall 31.4% over 40 shows. Full scoring history →