How does personalized ranking work on the home feed for a logged-in user?

When a Reddit user is logged in, their home feed is assembled from posts across every subreddit they have joined, ranked by a combination of factors that prioritize recency and upvote velocity. The fundamental mechanism is the Hot algorithm applied across the user's entire subscription list simultaneously: posts compete against each other based on their score relative to their age, with early upvotes carrying far more weight than late ones. A post from a small, niche subreddit with 80 upvotes received in its first hour may rank higher in a user's feed than a post from a massive subreddit with 5,000 upvotes accumulated slowly over an entire day. Reddit applies normalization to prevent large subreddits from completely dominating the feeds of users who subscribe to both small and large communities, though in practice the highest-traffic communities still generate disproportionate representation. Beyond the Hot algorithm, Reddit layers in additional personalization signals for logged-in users. The platform pays attention to which communities a user visits and engages with most frequently, and tends to weight posts from those communities more heavily. Posts from subreddits a user has recently joined may receive a slight boost while the subscription is new. Users can also influence their feeds by marking individual subreddits as favorites, which pins them to the sidebar and signals stronger interest. The alternative sort options — New, Rising, Top — are not personalized in the algorithmic sense; they are purely mechanical re-sorts of the same post pool using different criteria. Reddit has added a "Best" sort for the home feed that takes into account user engagement signals and is intended to personalize further than the standard Hot sort, though the precise mechanics of Best are not fully documented publicly.