top corner

Expected ROI/Win Rate for HU Poker Players

A common question that many players ask is what a good winrate is or how high of a ROI can be achieved in the games they play.

These are good questions. Knowing what other successful players are able to accomplish gives you something to compare to. You can then determine whether you're ready to move up in limits, that you need to improve in some areas or that your game needs a complete overhaul.

But the answer isn't always clear cut. In fact, it often isn't. So, in this article I will outline some very rough ranges for what you can expect to achieve for a ROI in heads up sit and go's and win rate for heads up cash games. I will also briefly discuss the variance you should expect to see and sample sizes you need to make an accurate assumption as to whether your ROI or winrate is solid.  

ROI Ranges for HU SNGs

If you're a heads up SNG player, your ROI should be anywhere from 5 to 20%.

For limits up to $30 or so, 15 to 20% should be what you're aiming for give or take a few percentage points. This means you'll earn up to $6 per game or so.

As you move up in stakes, the competition will get tougher so you should expect your ROI to decline. In the mid stakes, players claim that 10% is a solid ROI with 15% being really good. At the highest stakes, you can expect to earn a ROI of 5 to 7 percent.

Winrate (WR) Ranges for HU Cash Players

For heads up cash game players, it looks like the "standard" winrate range would be between 5ptbb/100 to 10ptbb/100 or 10 to 20 big blinds per 100 hands.

Again, this range will fluctuate based on the stakes you play mostly. However, your winrate will also shift due to whom you're playing against.

These winrates have been reported by players playing as high as 400nl.

Variance and Sample Size in Heads Up Games

When discussing ROIs and winrates, it's also important to keep variance and sample size in the back of your mind.

Let's use heads up cash games for example. There are threads on 2+2 that discuss how many hands you truly need to accurately assess your true winrate. It's been said that players that have over 100k hands and are crushing their stakes are still 4-6 ptbb/100 off their true winrate. This is a huge difference. For a more accurate winrate, players will need to play over 1 million hands. This will result in a winrate +/- 1ptbb/100 of where they really are at. (This data was for a specific player. It may differ from player to player, but the principal still applies - the larger the sample size, the more accurate your statistics.)

The moral of this example is that when you're trying to figure out your ROI or winrate, you need to have a substantial sample size for the stats to remotely be accurate. For heads up cash, you need at the very least a sample size of 50k hands. For heads up SNGs, it is suggest that you have a minimum of 1-2k hands. Personally, I think that's a bit low. I think 5k should be the minimum. And even then, your ROI will be off a few percentage points.

Summary of Expected ROI/Win Rate in Poker

Knowing what a solid ROI or winrate is gives you the opportunity to compare and see how you're doing. You'll see if you're on the right path to poker success or if you need to put in some more effort on the studying end of things. Keep in mind, however, that for in order for your stats to be accurate, the larger your sample size the better. That way the variance has leveled out some and you don't have stats primarily reflecting running good or bad.

More Heads Up Poker Strategy
bottom corner