What Are the Real Odds of Going 82-0?
“82-0” is a dare precisely because it’s never happened. Here’s the actual arithmetic behind why.
No real NBA team has ever done it
The two best regular seasons in NBA history both fell short of perfect: the 2015-16 Golden State Warriors finished 73-9, breaking the previous record set by the 1995-96 Chicago Bulls at 72-10. Further back, the 1971-72 Los Angeles Lakers finished 69-13 but produced the most dominant single stretch in league history — a 33-game winning streak, still the longest in any of the four major North American pro sports leagues. Even that historic run included 13 losses across the full season.
Why win probability compounds against you
Say a team is good enough to be an 85% favorite in literally every game it plays — an extraordinarily high bar sustained across an entire season. The odds of winning 10 games in a row at that rate are only about 1 in 4 (0.85^10 ≈ 0.197). Stretch that to 82 straight games and the probability collapses to a number small enough to be functionally zero (0.85^82 is a fraction of a percent). Real teams don’t sustain a flat 85% win probability anyway — injuries, fatigue, a hot opposing shooting night, and simple regression all pull the real number down over a long season, which is exactly why even historically dominant teams like the ‘96 Bulls and ‘16 Warriors still lost close to 10 games.
What the math means for building a roster
A genuinely elite, well-balanced roster raises your per-game win probability — but even a very high per-game probability doesn’t get you close to a season-long “82-0” until it’s extreme enough that ordinary variance almost never intervenes. That’s the entire point of a chemistry and positional-balance system: it’s not decoration, it’s the lever that actually moves your simulated win probability meaningfully higher, the same way real roster construction does for actual NBA contenders.
Era length changes the math directly
Because the simulation runs an era-accurate schedule, the actual number of consecutive wins required for a flawless run ranges from 48 games (the league’s earliest seasons) up to the full modern 82. A shorter schedule is a mathematically easier target for the exact same reason a shorter win streak is easier to sustain than a longer one — fewer games means fewer chances for variance to end the run.
The honest takeaway
Treat a flawless result as what it actually represents: a genuinely rare outcome that requires both a strong, well-balanced roster and a real amount of good fortune along the way — the same combination it would take in real basketball. If your build keeps falling one or two losses short, that’s the format working as intended, not a sign something’s broken.
Build the strongest roster you can and find out
82-0 runs a fully seeded, transparent simulation — every game’s result depends on your actual roster ratings and era fit, and the chemistry breakdown shows you exactly what helped and what didn’t after every run. Free, no account required.
Frequently asked questions
Has any NBA team ever gone undefeated for a full season?+
No. The closest anyone has come are the 2015-16 Golden State Warriors (73-9) and the 1995-96 Chicago Bulls (72-10) — both extraordinary regular seasons, both with real losses.
What are the real odds of an 82-0 season?+
Effectively zero for any real team, and only slightly better than that for a simulated roster. Even a team that wins 90% of its games independently would have roughly a 0.02% chance of winning all 82 — and no real or simulated team's win probability is a flat 90% for every single game, since any given night carries injury, fatigue, and matchup variance that pulls the true odds down further.
Why is a long win streak so much harder to sustain than a single game?+
Because probabilities compound. A team that's a heavy favorite in any individual game — say, 85% to win — only has about a 1-in-4 chance of winning 10 games in a row at that same rate, and the odds keep shrinking with every additional game. Across 82 games, even a small week-to-week dip in form is usually enough to end the run.
Does the game's simulation ever actually produce an 82-0 result?+
It's designed to be rare by construction, the same way a real perfect season would be — every simulated game carries genuine variance (fatigue, in-game injury dips, opponent matchup, home/away, random noise), so even an elite, well-built roster loses games some percentage of the time. A flawless run (grade S) is meant to feel earned, not like the expected outcome of a strong build.
What roster qualities actually move the needle on going flawless?+
Real positional balance rather than overlapping strengths, genuine two-way play (offense and defense both above-average), a roster built for the era it's simulating in, and — since the schedule length is era-accurate — a shorter season length (48-70 games in earlier eras) mathematically requires fewer consecutive wins than a full 82-game modern-era run.