Postseason Probabilities
Loading the season 0 seasons simulated — pp at 95% Starting
In Outno games picked
Checking the live feed
Loading the season 0 seasons simulated — pp at 95% Starting
In OutHow to read it. Every figure is washed by how likely it is: quiet grey towards nothing, deepening through blue towards certain. The heavier rule falls after the sixth club in each league—the current postseason field. Club colours and logos identify teams; a colour in the probability cells always means magnitude, never identity. A ruled 0 did not occur in the simulation, while <1 is a live chance too small to round up. Select any row for its full seed distribution.
Playoffs is the chance of earning any seed. Division is the chance of winning the division, #1 Seed of taking the league's top seed, LCS of reaching the League Championship Series, Pennant of reaching the World Series, and Title of winning it. Seed 1 through Seed 6 and Out are mutually exclusive and describe every simulated finish.
5k, 20k, and 50k select how many complete seasons are sampled. Larger samples reduce Monte Carlo noise; they do not make the underlying baseball model more accurate. This preview still uses approximate standings tiebreakers.
After you force games, a small corner arrow marks the cumulative percentage-point change from the no-pick forecast using the same model and sample count. Clearing every pick removes the arrows.
Picking games. Choose a winner for any game still to be played and the browser simulates the remaining season again from scratch. Baseball has no tie outcome. Your picks are a hypothetical scenario, not new evidence, so the pregame figures do not turn into 100–0.
Every club begins from the same neutral rating, then the preview replays completed 2026 games. It has no prior-season history, roster, injury, bullpen, or starting-pitcher information yet, so it is under-dispersed and not a finished predictive model.
This is not sportsbook consensus. It passes the same season-to-date ratings through a different curve to exercise the future market-model seam; it is not independent evidence.
The full schedule contains 2,430 games. Completed games stay hidden by default, the list is grouped by current date, and rows arrive in bounded chunks. Filtering to a club is the practical way to browse and pick a 162-game schedule. Doubleheaders and makeup games keep their own identities.
During an active game window, the page checks a same-origin MLB score snapshot once a minute. Live scores replace the pregame figures in that schedule row, but they are display-only: they never alter records, picks, ratings, or simulation. A game enters the forecast only after a separately validated final is accepted.
Pick the rest of the season and watch the odds move.
Monte Carlo preview
Background worker · progressive estimates · identity-stable scenario links
The browser simulates unpicked games off the main thread and refines the table progressively. October series probabilities are precomputed once per scenario, so a simulated postseason needs only thirteen draws. Picks use MLB's own game identifiers in the URL; a postponement cannot silently shift an old shared scenario onto different games.
The official 2026 regular-season schedule contains all 2,430 games and validated completed scores through the date shown above the table, including postponements, doubleheaders, and a suspended game. During active game windows, a display-only same-origin score snapshot refreshes once a minute.
Ratings currently replay only this season from a neutral prior, Market is a demonstration curve rather than betting prices, and tied standings are ordered by a random draw instead of MLB's full published procedure. The page remains unindexed while those gaps are open.
Schedule and completed results come from the MLB Stats API. This is an independent hobby rebuild, not affiliated with or endorsed by MLB or any club and not a source of official information. Club and league names, colours, and logos are trademarks of their respective owners and appear only for identification.