Best of Three: How Often a Series Ends 2-0 or 2-1

•0•0Comment iconComment iconComment iconComment icon

Explore the mathematics behind esports series lengths, sideboarding, Fearless Drafts, and final score probabilities in Bo3 and Bo5 formats.

Edit Article

Magic Arena runs queues of two lengths: a single game, and a best of three with sideboarding between games. Esports uses both of those and a longer one, from single games in a regular season to best-of-five finals.

The extra games change more than the time a match takes. They hand out a small set of final scores, and on paper every one of their shares follows from a single number: the chance of winning one game.

What a Best of Three Asks For

A best of three is a race to two game wins, so it lasts two games or three. Every series that reaches a third game got there through 1-1.

Paper Magic writes the rule with a small twist. A match runs until one player has won a set number of games, usually two. A drawn game does not count toward that total, so a long round can need a fourth game.

Arena's best-of-one format trades that structure for speed, and the reasons players pick itlink outside website come down mostly to time and fatigue.

What a lone game gives up is room to recover from one bad start. A hand with no lands usually loses whatever the deck's quality. That is why the mulligan exists at alllink outside website in every format, and a best of three goes further with up to two more games.

Four Ways a Best of Three Ends

Suppose one deck wins 64% of its games against another, and each game is a fresh draw that remembers nothing of the last. Every final score then has a share, set by the number of orders in which its wins and losses can arrive.

A 2-0 needs two straight wins, which happens 41.0% of the time. A 2-1 needs the first two games split in either order and then a win in the decider, which comes to 29.5%. The weaker side takes 16.6% at 1-2 and 13.0% at 0-2.

Together the two winning scores give the stronger side 70.5% of series. A decider is played in 46.1% of them, close to half, even with a clear favorite at the table. And for any side that wins more than half its games, a 2-0 is more likely than a 2-1.

Upsets lean the other way. Of the series the weaker side takes, 56.1% finish 1-2, so most of them come through a split of the first two games.

A series and a single game are two separate questions until the decider, where they become the same one. Opening the esports page at overdog.betlink outside website shows what that does to the markets. Its ladder of map and game winners usually stops one map short, because whoever wins the decider wins the series, so that game rarely gets a market of its own.

Best of Five: Why the 3-1 Comes Up Most

Stretch the series to first-to-three and the same 64% side has six possible final scores. It wins 3-0 in 26.2% of series, 3-1 in 28.3% and 3-2 in 20.4%, for 74.9% in total. The weaker side takes 11.5% at 2-3, 9.0% at 1-3 and 4.7% at 0-3.

The chart sets both formats side by side.

Every bar assumes a 64% chance of winning each game.

The 3-1 is the single most likely result here, ahead of the sweep. That holds for any side whose chance per game sits between one half and two thirds, and only past two in three does the 3-0 take over.

A deciding fifth game, at 3-2 either way, comes up in 31.9% of these series. That is less often than in a best of three, because the opening four games have to split 2-2 before a fifth is played.

Why Games in a Best of Three Are Not Coin Flips

All of those numbers assume that every game is drawn from the same 64%, and card games break that assumption by design.

Game one is played with the main deck, and before game two both players can bring in cards from the sideboard. Under the paper rules the loser of each game picks who plays first in the nextlink outside website and may hold that choice until sideboarding is done.

Take a deck that wins 64% of game ones and only 52% once the opponent's sideboard is in. Its sweeps fall from 41.0% to 33.3% of series and its deciders rise from 46.1% to 49.4%, while its series rate ends at 59.0%.

Games outside cards show the same dependence on earlier results once somebody measures it. In professional tennis over three sets, later sets leaned on the ones before themlink outside website instead of starting fresh.

League of Legends at Every Series Length

League of Legends uses all three lengths. Single games fill some regular seasons, best of three is the common middle, and best of five is kept for the late rounds of a playoff.

The draft is where its games stop being independent. Each team sees the other's champions in game one and can answer them in the next draft. Where a league runs Fearless Draft, a team cannot go back to a champion it has already playedlink outside website for the rest of the series.

Under that rule the late games of a long series are drafted from a thinner pool, so the depth of a roster counts for more over five games than over one.

A side's chance of winning one game and its chance of taking the series are therefore different numbers. Reading the match-winner and game-winner split in League of Legendslink outside website puts the format rule next to the markets. A best of three needs two game wins and a best of five needs three, so a favorite can drop a game and still take the series.

What the Venue Earns on a Series

Reading League of Legends drafts and deciders better than the market is an edge that pays only for as long as a venue keeps taking the bets.

Every market on both boards linked above is listed on Polymarket, a betting exchange whose traders post every price. OVERDOG's one charge there is a flat 1%, to which Polymarket adds its own fee wherever the order owes one.

A bet that a series reaches its decider is, in effect, a disagreement with a trader who expects a 2-0 either way, and the series settles it between the two of them. The series result, single games and series length trade as separate markets there. Correct calls on deciders do not lower what an account may stake on any of them, though the most a decider bet can take at a price is the 2-0 money already waiting there.

A sportsbook quoting the same series sits across the table from each of its customers. A regular at that table who keeps reading deciders right is playing against the house's own money, and the house sets the size of the next hand.

What a 2-1 Says About the Better Side

A 2-1 says about as much as one game does. The first two games cancel each other out, so of all series that finish 2-1 either way, the stronger side takes exactly 64%, its rate for a single game. A sweep says more: of all sweeps either way, the stronger side owns 76.0%.

Most upsets still arrive through a split of the first two games, and in a best of five the 3-1 is the score to expect from a side that is good without being dominant. Sideboards and drafts move every one of those shares, because the games stop being fresh draws.

Esports Match Outcome Probabilities FAQ

What is the most likely score in a Best of Five series for a favorite?

For a side with roughly a 64% win rate per game, a 3-1 finish is the most likely single outcome (28.3%), occurring more often than a 3-0 sweep (26.2%).

How does sideboarding alter game independence in Best of Three matches?

In card games like Magic: The Gathering, players adapt their decks with sideboard cards for games two and three, which lowers sweep percentages and increases the likelihood of a deciding third game.

What is a Fearless Draft in League of Legends?

A Fearless Draft prohibits teams from selecting champions they have already played in earlier games of the same series, making roster depth and champion pool crucial in longer Bo5 series.

Why do upsets happen more frequently in 2-1 series than in 2-0 sweeps?

When a weaker team wins a Best of Three series, 56.1% of those upsets occur via a 2-1 scoreline following a split of the first two games. Learn more about best-of-three match formatslink outside website.

What did you think of this article?

0%
User score (0)

Share this article!