Leander sent me the spreadsheet at 7:14 on a Tuesday morning. I was still in bed. The subject line said "OCTOBER PARKING — PRELIMINARY FRAMEWORK (DRAFT 3)" and I want to be clear that I did not ask for a draft one or a draft two.
The spreadsheet was color-coded. Of course it was color-coded. There was a legend.
He called it the Foliage-Responsive Arrival Matrix, which he explained — unprompted, at length, while I was trying to put on my jacket — combines leaf-peep traffic patterns, day-of-week demand curves, and what he referred to as "canopy saturation indicators" to predict, within a fifteen-minute window, the optimal time to arrive downtown and secure parking without circling. He had mapped the City Center garage. He had mapped the side streets. He had assigned each a color corresponding to projected congestion severity, which ranged from Harvest Gold (fine) to what he had labeled, in a deeply concerning move, Crimson Critical.
"The leaves are the data," he said, handing me a coffee like that sentence was normal.
The outside temperature was 48 degrees and climbing and the light was that specific sharp October light that makes Saratoga look like someone commissioned a painting of itself. We were trying to get to breakfast. That was the whole plan. Breakfast.
I asked him where the canopy saturation data came from and he said he had been monitoring the tree situation on Phila Street "for about three weeks" and cross-referencing it with tourism event listings, specifically the Fall Festival on the 24th, which he described as a "high-pressure leaf-peep convergence node." He said this while merging onto 87. He had a printout.
The matrix accounted for the fact that the Tang Dumplings opening had added a new downtown draw. It accounted for weekend brunch spillover from the Washington Street corridor. It did not, I noticed, account for the fact that we were two people going to get eggs on a Wednesday at nine in the morning, which is arguably not a parking emergency under any reasonable framework.
I pointed this out.
He said, "That's exactly the kind of thinking that creates the problem."
We arrived at 9:08. We parked immediately. There were many spaces. It was fine.
He logged this in the spreadsheet on his phone before we got out of the car. He said the 9:08 arrival confirmed his Harvest Gold window hypothesis and that he was "cautiously optimistic" about the model's predictive validity. The sun was fully out by then, that mid-morning October warmth that makes you want to take your jacket off by the time you hit the second block, and downtown was lovely and easy and not gridlocked even a little bit.
Breakfast was great. We sat outside. There were leaves on the sidewalk — orange and yellow and that particular red that only shows up for about ten days a year and makes everyone feel briefly like they are living inside something worth feeling — and Leander was on his second coffee and explaining how he might build a precipitation layer into the matrix for the November iteration, and I was watching a golden retriever try to eat someone's takeout bag, and I thought: this is actually a really nice morning.
He asked if I wanted to see the November projections and I said maybe later and he said "smart, the data's not clean yet" like that was a satisfying answer.
On the way home he said he thought the matrix could be adapted for the Adirondacks. For "regional scalability." He was already pulling up a satellite view of the canopy coverage around Lake George.
I took a picture of a really good tree and sent it to my sister with no context.
She replied: "is this from him"
I said: "the leaves are the data"
She said: "I'm so sorry"
She was not sorry. Neither was I, honestly. It was a beautiful morning and we had found parking and he is going to spend the entire drive back from every single October outing for the rest of his life telling me whether the matrix held up. I have made my peace with this. I am part of the dataset now.