Tuesday brought with it an unexpected November surprise in election calls as The New York Times published that The Needle might return, provided its tech supports work properly. Predictive tool, much maligned after its failure to correctly estimate Donald Trump’s chances in 2016, never really disappeared from our lives. As it evolved and was refined over time, this algorithm accurately predicted a contentious Senate election in Alabama in 2017 before being applied in primary and general election races since. Needle is unique among election winner-loser calls because it provides real-time probability estimates well before an official result becomes known. AP VoteCast and Edison Research — two primary election data services — wait until they can establish who won, particularly given this year’s contentious election cycle. Julie Pace of AP Executive Editor Julie Pace informed me in an interview several months ago that VoteCast does not speculate until their certainty threshold has been met; clients may however decide to allow more leeway for commentators at their events. Both The Times story Tuesday morning and an X thread by chief political analyst Nate Cohn at noontime provided a caution: With Times Tech Guild staffers striking, correcting glitches might prove challenging and potentially impossible; worse case, live versions of The Needle may no longer be published. “Should the Needle’s results not be available live, our journalists plan to run its statistical model periodically, examine its output and provide updates in our live blog about what they see – providing our readers a sense of where it really stands during a race’s progression,” according to an explainer published by The Times. Cohn addressed these risks in his post: “I do not yet know whether we will be able to publish the Needle; there may be reasons to bet against doing so; although perhaps there could be scenarios in which all goes smoothly; alternatively, bugs might arise early and we simply won’t make it in time.” Access to The Needle and related coverage will initially remain free. In response to its setback last year, The Times said the Needle wasn’t wrong but simply misinterpreted by readers. Tuesday’s story addresses one of these discrepancies by explaining that, when showing 75% chance a candidate will win, that also implies there is one in four chance she may fail – an easily understandable concept once explained. My guess is that in 2024 it is even simpler as legalized sports gambling has gained momentum and betting can now occur during games, changing odds constantly. One key characteristic of sports wagering is permitting bets during live action that allow bettors to wager as the odds change during gameplay. Even news consumers who do not engage in gambling may find The Needle an intriguing source, provided it overcomes technical hurdles and takes off successfully. It offers something extra beyond what can be found on national networks or papers adhering to AP or Edison Research rules. New York Times spokesperson Charlie Stadtlander strongly objected to my odds-making reference in an email: “[T]his isn’t correct characterization but rather placing election returns within their proper context as they come in – not creating false expectations with oddsmaking.”… The purpose of The Needle is simply providing context as elections results come in and remain undecided.” Early returns can often be misleading: initial counts often vary dramatically from those remaining. Here are a few other notes regarding decision calls before counts begin: AP’s Pace has informed me and other interviewers that one major change this cycle is increased transparency — offering longer explanations of methodology than before. Last week, The Washington Post and Wall Street Journal followed suit by publishing pieces on how their editors will make calls using VoteCast to quickly correct errors within an immense influx of data. I learned from The Post editors how important VoteCast can be at quickly recognizing errors amongst incoming information to quickly correcting it before passing along to voters. The Post also invests in additional data by subscribing to Edison’s product as well as that from AP, to ensure independence from data providers announcing decisions live or online. A key challenge of such systems lies within editorial oversight: editors play an essential intermediary role between data suppliers and those making announcements on air or via websites. Fox News in 2020 was an example of such an arrangement where an AP client might make calls earlier or later than AP itself; network data desks often work alone in separate rooms during election night reporting, making their decisions independently from others on election night reporting. Commentators on either left or right-leaning networks often spin election results in their favor; decision operations on such channels have an edge here – as was evidenced in Arizona where Fox News called Biden as victor well before left-leaning or middle-of-the-road outlets did so. Unfiltered AP: When it comes to political coverage of state and local races as well as presidential contests, AP has long been considered an indispensable source. An alternative way of keeping abreast of what AP is stating at any particular moment would be visiting their official site at apnews.com directly. An Alternative Decision Desk Provider: There is another provider, Decision Desk HQ. Although smaller in scale, their diverse client roster includes The Economist, The 19th, and even NewsNation network! Decision Desk offers another view into channel-hopping’s state of affairs for channel-hoppers, while number nerds like me who prefer more detail may want to explore these links early in the evening before state-by-state totals take center stage and make all of the noise. Poynter media business reporter Angela Fu and Poynter contributor Nicole Slaughter Graham contributed this report, while it has also been updated with quotes from New York Times spokesperson and to make clear that The Needle will remain free for readers.