Dodge for something concrete — a comp with no win condition, a teammate saying they'll feed — not a bad feeling five seconds into champion select. Below Master, a dodge costs 5-15 LP and a queue lock; at Master and above it's a guaranteed loss-equivalent, so the bar for leaving should be higher, not the same.

What a dodge actually costs

Ranked queue lets you leave champion select before the game starts, at a price. As of the 2026 ranked update, that price is tiered by how often you've done it recently:

  • First dodge: −5 LP, 6-minute queue lock.
  • Second dodge: −15 LP, 30-minute lock.
  • Third dodge and beyond: −15 LP, 720-minute (12-hour) lock.

That tier decays by one step every 24 hours since your last dodge — a clean day resets you a step closer to first-dodge pricing, and a new dodge resets the countdown. Dodging no longer resets your autofill status either, which used to be a separate cost people factored in.

Riot has changed these numbers before and will again, so treat the figures above as "as of the 2026 ranked update" and check the client if something looks off, or if the exact number matters enough to confirm. What doesn't change is the shape of the trade-off, which is the part worth actually reasoning about.

The trade you're making

A dodge costs a fixed, known amount: some LP and a queue lock you sit out. Staying costs an expected amount — the LP you'd lose if the lobby's real win rate is worse than a normal game's, multiplied by how much worse it actually is. Most games are close to a coinflip regardless of who's in your lobby, so most of the time staying is correct by a wide margin — the dodge fee is real money spent on a game you were roughly 50/50 to win anyway. The exact break-even point — the win probability where that trade actually flips — is quantified here, using the same asymmetric LP math the app's own dodge verdict runs on.

The trade only tips toward dodging when the lobby is genuinely, specifically bad — not "this feels off," but something concrete: a duo insisting on a comp with no win condition, someone locked into an off-role pick after being told it doesn't work, a teammate typing that they're going to feed. Should I Dodge's win estimate can help quantify "how bad," but it only ever reflects aggregate champion win rates — it has no idea whether your teammates are tilted, and neither does any tool.

Tilt is the thing to watch for on your own side of the trade. A dodge decided in the first ten seconds of champion select, before anyone has picked anything, is rarely about the lobby — it's about the last game. The penalty schedule above is designed to make repeated impulsive dodging expensive specifically so it isn't free to act on that impulse.

Signals that feel stronger than they are

Some things that trigger the urge to dodge are weaker evidence than they feel in the moment. A teammate on an unfamiliar role, zero mastery points on their locked champion, or a duo you've simply decided you don't like based on their name are all common triggers — and all of them are compatible with a completely normal game. Plenty of games are won by someone playing a champion for the first time, or by a duo that turns out to communicate fine. These are things to notice, not things to act on by themselves.

Stronger signals are the ones that describe an outcome, not just a vibe: a teammate stating they intend to feed or grief, a comp with no describable win condition at all, someone refusing to swap out of a role two other people also need. The difference is whether you can point to something specific that will make the game harder to win, versus a general sense of unease. The penalty schedule exists precisely because the vibe-based trigger is common and the specific one is rarer — Riot is pricing in that most dodges, historically, were the former.

Master and above: a different calculation

At Master, Grandmaster, and Challenger, a dodge costs −20 LP and an MMR penalty — Riot treats it as functionally equivalent to a loss. Below apex tiers, dodging leaves your MMR untouched; at apex tiers it doesn't. That changes the maths, not just the numbers: you're no longer comparing a small guaranteed cost against the expected cost of a bad game, you're comparing a guaranteed loss now against a maybe-loss later. A lobby has to be substantially worse than a normal apex-tier lobby — not just below 50% — before dodging beats staying on expected value. At that level, most lobbies that feel bad are simply normal variance in a queue where outcomes are naturally noisier, and reflexively dodging them trades a guaranteed loss for a maybe.

The short version

Dodge for concrete, specific reasons — a comp that structurally can't win, a teammate signalling intent to grief — not for a bad feeling five seconds into champion select. Below Master, the cost is LP and time. At Master and above, the cost is a guaranteed loss-equivalent, so the bar for dodging should be higher, not the same. See how the win estimate itself is built if you want to know what it can and can't tell you before you decide.