What is a Good Limp Percentage in Poker?

Your figure	What it usually means
0–5% guideline band	Zero is a defensible figure here and the band starts at it. A small number is usually small-blind completes rather than genuine open-limps.
Over 5%	Limping caps your range at its weakest point, invites the whole table in and forfeits the pot you could have taken uncontested. This is the cheapest leak on the list to fix.

Your figure

What it usually means

0–5%

guideline band

Zero is a defensible figure here and the band starts at it. A small number is usually small-blind completes rather than genuine open-limps.

Over 5%

Limping caps your range at its weakest point, invites the whole table in and forfeits the pot you could have taken uncontested. This is the cheapest leak on the list to fix.

What the stat counts

A limp, for this band's purposes, is the case where your first voluntary action was a flat call of the big blind in an unraised pot. Two things are deliberately excluded. Calling a raise is not limping — that is a cold-call, and it has its own place in the VPIP-to-PFR gap. Completing from the small blind is a limp by this definition, which is why a small non-zero figure is usually harmless: you are already half in, the price is half a big blind, and the alternative is folding the bottom of a range you have partly paid for. The distinction that matters is the open-limp from any other seat, where you have chosen to enter a pot without taking the initiative you could have had for one more raise.

Why the band starts at zero

Every other band on this site has a floor, because folding too much of the relevant spot is itself a mistake. Limping has no such floor: a strategy that never limps outside the small blind gives up nothing, which is why the band reads 0-5% rather than something-to-something. The cost of a limp is not subtle either. You forfeit the pot you could have won uncontested — the blinds are sitting there and a raise takes them a meaningful share of the time. You cap your range at its weakest point, because a limping range is by construction the hands you did not think were worth a raise. And you invite the whole table in, turning the one thing a raise buys you, a smaller field, into its opposite.

What happens to a limper

The ranges on this site quantify the punishment rather than just asserting it. Our isolation ranges raise over a limper with 29.7% of all hands from the button and 16.1% from the cutoff. Both are tighter than the seat's own opening range — the button opens a folded pot with 40.0% — because a limper has a hand where a folded pot has none. The comparison that matters is the other one: open under the gun and the range that re-raises you is 2.6% of hands; limp instead and you hand the button 29.7%. So a limp does not simply let you see a cheap flop: it multiplies the range that will be raising you, and it does so from a seat that will act after you for the rest of the hand. You then face a raise holding a hand you already judged too weak to raise with, out of position, in a pot that is no longer cheap.

The exceptions, honestly labelled

There are spots where limping is not a mistake, and it is worth being precise about them because "never limp" is advice that falls over at a real table. Small-blind completes against a big blind who never raises are cheap and defensible. Deep-stacked live games with a permanent straddle change the effective price of entry, and our live strategy article covers what that does to the maths. Limping behind two or three existing limpers with a hand that plays well multiway is a different decision from opening a folded pot, because the pot you would be raising is no longer two blinds and the field you would be thinning is no longer thin. What is not on the list is the limp that exists because a hand felt playable but not quite raiseable, which is where almost all limping comes from.

How long before your own figure means anything

This is the quickest of the postflop-adjacent stats to read, because the band is narrow but so is the frequency it describes. About 150 hands where the pot was unraised when you acted settles it to the 2.5-point half-width of the band, and the pot is unopened when you act on roughly 65.1% of hands, so that is about 230 hands. Treat that pair as a floor rather than a figure: the standard error of a proportion is a poor approximation at a rate as low as the 2.5% band midpoint, and it understates the sample a genuinely small limp figure needs. In practice you rarely need the statistics: if you limp at all outside the small blind, you will know, and so will the button.

Check your own figure

The free leak report separates limps from cold-calls in a hand history and marks the figure against this band, which is the fastest way to find out whether your small number is small-blind completes or something more expensive. If you want the ranges that replace limping, the range library holds the iso-raise ranges quoted above.

Frequently asked

Is limping always bad in poker?

Outside the small blind, at 6-max online, effectively yes — which is why the guideline band is 0-5% and zero is a fine figure. The defensible exceptions are small-blind completes, limping behind several existing limpers with a hand that plays well multiway, and some deep or straddled live games where the price of entry is not what the table stakes suggest.

What is a good limp percentage?

0-5%, and the lower end of that is better. A figure of a point or two is usually small-blind completes. A figure above 5% means genuine open-limps, which is the single cheapest leak on any stats report to repair, because the fix is a decision rather than a skill.

How should I punish a limper?

Raise, from position, with a wide range. Our iso-raise ranges use 29.7% of all hands from the button and 16.1% from the cutoff — far wider than the 2.6% the same library re-raises an early open with, because a limper's range is weak and capped and the blinds are still there to be taken.

Does completing from the small blind count as a limp?

By the definition our leak report uses, yes: the first voluntary action was a flat call of the big blind in an unraised pot. That is why a small non-zero limp figure is usually nothing to worry about, and why it is worth checking which seat the limps are coming from before treating the number as a leak.

Where do these benchmark numbers come from?

Guideline bands from commonly published winning-player ranges for 6-max online micro-stakes cash (tracker population norms, NL2–NL25). Not solver output; short samples are noisy.

How many hands before my limp percentage is meaningful?

About 230 hands, since you need roughly 150 unraised pots to act in and the pot is unopened when you act on about 65.1% of hands. It is also the one stat you can audit from memory rather than from a sample.