showing both median and mean is key because a single bust-out (one large number in the gap chart for that show) pulls the mean up ... but a low median indicates most of rest of songs were in heavy rotation (low gap number) ... so the separation between ⚫ and 🟥 is important

need to keep developing 🤔

I've been playing around with a new Phish plot, one that shows gap (# shows since last time played) info ... this has show on x-axis (this is first part of Fall '95 tour) with median (⚫), mean (🟥), and max gap (number) ... it's a lot of info, but it's not intuitive to glance at and learn something

me and my Phish buddies are about to start our Fall 1995 30th anniversary listening project, where we listen to each show on (or very close to) the day of a tour that was special for all of us

it's gonna be a lot of Phish ... 54 shows in 81 days 🤯 ... starting on 9/27

things have been very busy with life, work, etc. ... just now catching up on the recent batch of shows and, my lord, that Light from 9/12 (that comes out of the Chalk Dust) is sublime

yea, could pick a dozen or so 'heavy hitter' tunes and somehow plot the median durations by era for multiple songs together ... need to think about that 🤔 ... maybe each song as a unique color

that'll need to wait for another day

and visualized as 'box and whisker' plots ... bold horizontal line in middle of colored boxes is the median value, box extents are the interquartile range, and outliers shown as dots

here ya go ... I did this very quickly

Plot of duration over time with the HF Pod Mt. Rushmore candidate shows in orange. Blue line is a 20-point moving average. Data from phish.net (with a few manual fixes).

plot of duration/length (in minutes, vertical axis) over time (horizontal axis) with the HF Pod Mt. Rushmore candidate performances in orange

just for fun ... maybe it provides some context for the discussion 📈🤓

oh wow, glad these are available and accessible 🙌 ... hope to get into them soon

a man is making a funny face and says i don 't care ! ALT: a man is making a funny face and says i don 't care !

topic of discussion on my Phish friends text thread was Goose and I said that I don't have a strong opinion either way ... I hope they succeed, I'll probably listen to them more one day ... but very not interested in the forced 'where do you fall on Goose?' conversation ... it's okay to not care

this conversation with Charlie and Andrew was fantastic, was very interesting to hear about the evolution and approach to jamcharts

and, wow, Brinkman has an exciting 'internship' with the team coming up, he must be walking on cloud nine this morning

whenever I listen to the first few notes of this come out of Trey's guitar, I picture Page, Mike, and Fish grinning nervously and thinking to themselves "holy shit, here we go"

yep ... and I know everyone comes for the Gin but all of it is amazing ... including the Uncle Pen, which is both playful and tight as shit

playing around w/ the look of my song-duration plots on this Friday afternoon ... this iteration (using Disease as an example) has 'jamchart' versions in orange ... potentially useful (?)

perhaps unsurprisingly, almost all jamchart versions are longer than the moving average (blue line) #PhishPlots

to go with @hfpod.bsky.social's discussion of AC/DC Bag today, here's a plot of the duration of AC/DC Bags through time w/ some of the longer ones annotated (blue line is a 20-point moving average, just to help see any trends) #PhishPlots

check out the uptick in length in 4.0 (compared to 3.0)

and these textures of the current era aren't only Trey/Page/Mike but also Fishman's fundamental playing style ... in many of those late 90s jams he plays it pretty straight (in the pocket, solid backbeat groove) ... nowadays he can be all over the kit ... a percussion rinse, if you will

you know the Bittersweet Motel scene when Trey says fans don't care if they miss changes, that's not why they come, they come for the energy (something like that, I'm paraphrasing)

while I don't get bent out of shape with flubs, but I *love it* when a composed section is nailed ... so satisfying

me when I ask myself "Should I stop listening to the SPAC shows?"

yes! ... it is maybe the most spectacular segment of an all-time spectacular show ... the transition back into Reprise is breathtaking

I did not know this ... so, is not just you

I don't even know where to start ... when I have a moment of wanting to explore, I become overwhelmed with the catalog

and, finally ... I computed the median gap after removing outliers (defined as values outside of 2 standard deviations, a common approach) ... just a smidge better than above

none of this really matters, it's just nerdy fun ... if you all keep proposing these ideas on the pod I'll do more of this

I couldn't resist doing some more on this ... here is the median gap (middle value of distribution) instead of the average gap ... still a poor correlation but a little better; at least the blue trend line in the direction that supports the hypothesis

nice! ... I took your computed '# of 20/25 minute jams per show' stat and put into these simple bar plots ... just to help visualize

my gosh ... 2021, 2024, and 2025

Largest performance gaps for Summer 2025 tour (with at least a 50-show gap). Data from phish.net

Second, here's a similar bar chart but showing the largest performance gaps (for those with ≥50-show gap).

A bunch of bluegrass-y tunes up there near the top.

Most frequently played songs for Summer 2025 tour (with at least 4 performances). Data from phish.net

Alright, a couple of Summer 2025 tour data visualizations for those of you who are into that sort of thing. #PhishPlots

First, here's a chart of the most frequently played songs (with at least four performances). We all know why Tweezer Reprise is at the top 🤩

I have yet to explore relationships to jam chart designations ... is on my long-term to-do list ... if only I could somehow have a day job of analyzing Phish stats, lol

the dataset is rich to explore! ... however, I've been less inclined to analyze relationships to show rating because of the subjective nature of that (at least relative to 'hard' metrics like show gap, song duration, number of songs, and similar)

(of course, this all assumes that the phish.net show rating is an accurate measure of show 'quality', which is a whole can-of-worms as we all know ... but, it's what we got)