The term breaks down literally: bb stands for big blinds, and the 100 is a block of one hundred hands. So a win rate of four bb/100 means that, over the long run, the player gains four big blinds for every hundred hands they play. It is an average across all the hands, not a result from any single one - most individual hands are won or lost by far more than four big blinds, but they net out to that figure across a large sample.
The reason poker settled on big blinds rather than money is comparability. A rate quoted in currency would mean something different at every stake, but four bb/100 describes the same strength of edge whether the big blind is tiny or large. It isolates skill from stake size, which is exactly what a win rate should do.
01A worked example
Suppose a player has logged fifty thousand hands and finished up two thousand big blinds in total. Dividing the big blinds won by the number of hundred-hand blocks - that is, two thousand divided by five hundred - gives a win rate of four bb/100. The same arithmetic works in reverse: a four bb/100 player who puts in another twenty thousand hands would, on average, expect to win around eight hundred big blinds more.
What the example also shows is how slowly the money accumulates relative to the hand count. Four big blinds per hundred hands is a perfectly respectable edge, yet it takes a large volume of hands to turn into a large number of big blinds. That slow, steady accumulation is the normal texture of a winning cash-game graph.
Why count in big blinds at all, rather than just dollars? Because the unit is built to do things currency cannot:
- It removes stake-size distortion. The same edge reads identically whether the big blind is small change or a serious sum.
- It compares performance across limits. A rate at one stake lines up directly against a rate at another.
- It focuses on edge, not currency. The figure measures how well a player plays, not how high they sit.
- It suits cash games. Steady per-hand returns map cleanly onto bb/100, where tournament results need ROI instead.
- It keeps results in poker-native units. Players think and talk in big blinds, so the number is instantly meaningful.
02Turning bb/100 into an hourly rate
To translate a win rate into money per hour, three pieces are needed: the bb/100, the size of the big blind, and the number of hands played per hour. Multiply the bb/100 by the big-blind value to get the money won per hundred hands, then scale that by the hands-per-hour figure.
- Win rate. Say five bb/100.
- Big blind. Say one dollar, so five bb/100 is worth five dollars per hundred hands.
- Hands per hour. Around one hundred at a single online table, giving roughly five dollars an hour - and far more across several tables, since online play allows many hands at once.
The same five bb/100 produces wildly different hourly figures depending on hand volume, which is one of the clearest differences between online and live play. The conversion is also why a modest-looking win rate can still add up: hands per hour does a lot of the heavy lifting. Seeing how that rate sits against the swings around it is the work of poker variance.
Worth remembering: an hourly figure built from bb/100 is an average, not a guarantee for any given hour. The same five bb/100 player will have losing hours, losing days and losing weeks - the hourly number only describes the long-run trend once enough hands have smoothed out the noise.
| Win rate | BB value | Per 100 hands | Per 1,000 hands |
|---|---|---|---|
| 2 bb/100 | $1 | $2 | $20 |
| 5 bb/100 | $1 | $5 | $50 |
| 5 bb/100 | $2 | $10 | $100 |
| 8 bb/100 | $2 | $16 | $160 |
| 10 bb/100 | $5 | $50 | $500 |
The bb/100 stays the same; the money does not. Multiply the rate by the big-blind value to get cash per 100 hands, then scale by volume - a small per-hand edge becomes real money only across a large sample.
03Typical ranges by stake
There is no fixed scale, but the broad pattern holds across the cash-game world. At lower online stakes, winning regulars commonly post bb/100 figures in the low single digits up to the low teens, while a large share of recreational players sit below zero. As the stakes climb, those positive rates tend to compress, because the average opponent is stronger and a big edge is harder to maintain.
The takeaway is that a healthy bb/100 at the micros is not the same achievement as the identical number at high stakes, even though the figure is written the same way. How those ranges shift level by level - and why rake bites hardest at the bottom - is set out in the look at realistic cash-game win rates, while the broader idea of measuring edge across formats lives in the guide to the poker win rate itself.
| Stake level | Losing / break-even | Solid | Strong |
|---|---|---|---|
| Micro | 0 or below | 4-8 bb/100 | 8+ bb/100 |
| Low | 0 or below | 3-6 bb/100 | 6+ bb/100 |
| Mid | 0 or below | 2-5 bb/100 | 5+ bb/100 |
| Higher | 0 or below | 1-4 bb/100 | 4+ bb/100 |
04Why bb/100 still needs a large sample
A bb/100 figure is only as honest as the volume behind it. It is easy to log a few thousand hands, see a glowing rate and treat it as the truth - but at that sample the number is mostly noise. A handful of big all-in pots that happened to go the right way, a stretch of unusually soft tables, or a quiet patch of card-dead hands can each move the rate far more than skill does over the short run.
- All-in run-outs distort it. Winning or losing a few large flips swings a small-sample rate hard.
- Rake and table mix matter. The fee and the quality of games played both pull on the figure.
- 10k hands is still noisy. Cash rates need tens of thousands of hands, and real confidence often means 100k or more.
- Trust grows slowly. Only volume smooths the swings enough for bb/100 to settle near a player's true edge.
Read the rate next to the spread of likely outcomes in poker variance, and a short-sample bb/100 stops looking like a verdict and starts looking like the rough first draft it really is.