More baseball is the easy answer
The FanGraphs ALCS thread opened with a familiar postseason wish: let the series go seven games. One Brewers fan nearby, a member wrote, didn’t care who won as long as it lasted. Another asked whether a longer series might drain Milwaukee’s momentum. The reply: the city’s “collective inertia” ought to carry the team for a while.
FanGraphs staffer Sean Dolinar was happy to have the extra games, though he pointed out the tradeoff. The Rays had just played a seven-game series with two extra-inning contests; a Game 7 would leave two off-days before the World Series, enough for relievers to recover, but perhaps not enough to line up the starting rotation. That’s the real cost of a long series: more baseball for fans, more questions for the team that advances.
The thread follows the weather—and its own detours
Rain in Cleveland put the game on hold, with the tarp back on the field. One member had settled onto the couch with a sleeping cat and a book stranded in another room. Keaton Arneson, another FanGraphs staffer, supplied the other version of the delay: arriving home with delivered pizza only to find a tarp. The member with the cat had the obvious rejoinder: at least Keaton had pizza.
The postponement also sent Dolinar checking Cleveland’s weather more than usual. A member joked that Mike Trout would be disappointed; Dolinar clarified that he checks weather often, just not Cleveland’s—and added that he’d been looking up sunrise and sunset, too. A fan across Lake Erie explained why Cleveland storms were on their mind: storms in their area generally arrive from that direction.
Baseball talk turns into career talk
One member shared that a FanGraphs job posting had led to a baseball interview. The congratulations came quickly. Dolinar asked whether the application involved AI-written résumés and automated HR screening; the applicant said the process had been lightweight so far. The exchange landed on a familiar anxiety: Dolinar said he’d be scared to apply for jobs again.
That prompted a side conversation about learning to code. One student was hearing classmates complain about R, and resented the pressure to “upskill” when they already found computers frustrating. A longtime R user countered that the language had been part of their college coursework more than 15 years ago, and its continued use might make it a better bet than learning something newer. Another member said R had helped keep them steadily employed since 2021, alongside non-coding skills. The original poster conceded that statistics might be useful, but said they simply didn’t enjoy it; taking quantum mechanics the previous year with a similar practical aim hadn’t been pleasant either.
Python came up as an alternative: the baseball job applicant had chosen it over R, while another member was building Python skills for ArcGIS. Dolinar was ready to turn the delay into a coding chat. The group wondered whether FanGraphs’ chat ran mostly on JavaScript, whether a coding discussion would teach anyone anything, and whether Dolinar had invented a programming language just for the site. The joke was met with a practical limit: the coding talk might be hard to follow, dull, and occasionally interrupted by a bad joke.
Projections, bets, and the limits of a model
The thread had a betting subplot, too. Michael Minetti said FanGraphs projections had helped him back the Rays at 25-to-1 to win the AL East. He liked Tampa Bay at +155 to win the World Series, but the FanGraphs simulation called for 2-to-1, so he passed. Dolinar noted that betting against the Brewers because of the projections would have cost money. Minetti said he’d lost money betting against Milwaukee for three straight years, but added that FanGraphs had helped him pick two long-shot AL East winners, Toronto and Tampa Bay—enough to make up for many over-under misses.
The exchange captures what projections can and can’t do. They can help identify a price that looks appealing, but they don’t make every bet a winner, and Minetti’s own example is a reminder that even a model-backed pick comes with uncertainty.
A patchwork stack behind the chat
The chat poll drew a more believable picture of the game than it had in Game 1, when it gave the Rays a 91% chance to win. That small improvement came amid a revealing discussion of FanGraphs’ technology: the site’s front end runs on Next.js, with the chat interface built in Next, React and JavaScript. Its backend uses .NET and C#. A handful of older pages still run on .NET Web Forms.
That mix will sound familiar to anyone who’s worked on a long-lived product. Sean Dolinar said plenty of corporate teams have used similar stacks, though some may have moved on in recent years. He also wished FanGraphs used more Python, while adding that the language choice matters less to him because he doesn’t work much on the site’s ETL process.
The chat offered a quick tour of the language loyalties around the table: one reader declared for R, another for Python, and a third said they could manage Python documentation well enough to get by. The exchange turned into a useful reminder for people learning to code: you don’t need to know every tool in the stack. A field biologist in the chat said the basics, plus resources such as Stack Overflow, can help you work through harder problems.
What readers want from the product
The more practical ideas were about making FanGraphs easier to use. One reader wanted interactive visuals for leaderboards, with hover text and filters for splits. That’s a concrete request: tables can answer “who’s on top,” but filters and visual cues can make it easier to see how a player’s numbers change across situations.
Readers also asked for custom baseball emojis and notifications when someone replies to a comment—or likes a joke. Keaton Arneson said he often uses an Austin Hedges photo emoji in his baseball group chat, joking that Hedges’ league average wRC+ meant the image got plenty of use. Dolinar said FanGraphs’ chat already supports PNG images, though each one has to be hard-coded. On comment alerts, he said email would be the simplest option and app notifications might also be possible. He even floated replacing WordPress’s comment system with a modified version of chat.
That’s the most consequential product idea in the discussion. Reply alerts and reactions could make comment threads feel more like an active conversation, while a chat-based system would represent a larger change to how the site handles reader discussion. The chat doesn’t establish whether that redesign is planned.
The rest of the exchange wandered through baseball jokes, weather reports and cats. One reader wondered whether a run was basically a point; another answered with a reference to Fred Merkle. Cleveland’s weather was uncertain too: a reader west of the city reported little more than drizzle, while another saw the storm stretching across the lake. Dolinar said it was moving south and east, and he thought conditions were clearing.
And, naturally, there was a lot of cat talk. Arneson’s 15-year-old cat caught a fly; another cat was apparently trying to convince its household it had never been fed. A sleeping tabby had one reader pinned to the couch, while Arneson’s orange tabby got up to wait for food due 90 minutes later. His household feeds the cat at 7 a.m., noon, 6 p.m. and 9 p.m.
A rain delay turns into its own game thread
While the weather stalled the baseball, the chat wandered from a supposed baseball team hiring questionnaire to coding, 1908 references, Cleveland weather, beer, and cats that demand meals. One reader summed up the mood with a .000/.000/.000 slash line for understanding the conversation; host Sean Dolinar joked that this was why he wasn’t often allowed to host.
The digressions also surfaced two feature requests for the chat: image uploads and the ability to delete your own comment. And the cat advice was emphatic: don’t start feeding one three times a day unless you’re ready for a hard-to-break routine.
By the end, Dolinar called it the best “non-game game thread” they’d had, then set the next check-ins: the same time the following day for Game 4, with Game 5 scheduled for Sunday at 5 p.m. ET. The rainout question—when a playoff game last got rained out—was still hanging in the chat.
The questionnaire linkSource