What Does "Units" Mean in Betting Picks?

What Is a Unit in Sports Betting?

A unit is your baseline bet size. One unit equals whatever dollar amount you've decided represents your standard wager relative to your total bankroll. That number is completely personal. For one bettor, a unit is 10 dollars. For another, it's 500 dollars. Neither is wrong because the point of units isn't the dollar amount. It's the consistency.

Most serious bettors set one unit at somewhere between 1% and 2% of their total bankroll. So if you're working with a 1,000 dollar bankroll, one unit is 10 to 20 dollars. That might sound conservative but it's intentional. Setting your unit size at a small fraction of your bankroll means a losing streak doesn't wipe you out before the edge has time to play out over a meaningful sample of bets.

The dollar figure is private. The unit number is how predictions get communicated publicly without making everything about specific dollar amounts that don't apply to everyone reading.

Read More: Sports Betting Predictions Explained: How They're Made

If you want data behind the picks, visit our Predictions page to see today's Shurzy AI prediction model and how it's performing right now.

Why Do Prediction Services Use Units Instead of Dollars?

Two reasons, and both matter.

The first is bankroll management. Betting in units forces you to assign a confidence level to each bet before you decide how much to stake. Instead of impulsively throwing 500 dollars on a game because your gut says it's a lock, a units-based system asks you to decide: is this a 1-unit play or a 3-unit play? That decision happens before the dollar amount comes into the picture, which keeps emotion out of the staking process.

The second reason is comparability. A prediction service saying they're up 47 units on the season gives you everything you need to project real-world results regardless of your bankroll size. Multiply 47 by your unit value and you have a dollar figure that's relevant to your situation. Without units as the common denominator, comparing a bettor who stakes 100 dollars per game to one who stakes 10 dollars per game is impossible. Units put everyone on the same scale.

Read More: How Betting Predictions Help You Make Smarter Picks

How Do Confidence Levels Translate Into Unit Sizes?

Most serious bettors and prediction services use a tiered unit scale to match stake size to confidence level. The bigger the edge, the bigger the bet. The more speculative the play, the smaller the stake. A typical scale looks like this:

• 0.5 to 1 unit: Low confidence, speculative, or high-variance play. Long shot underdogs, props with limited data, or games where the edge is thin.
• 1 to 2 units: Standard play. Enough edge to bet, not enough to go heavy.
• 2 to 3 units: High confidence. The model edge is clear, the line looks sharp, and the matchup strongly supports the pick.
• 3 to 5 units: Maximum play. Reserved for the best bets of the week or month, not used often.

The logic behind this scale is that your biggest bets should be your best-informed ones. Flat betting, which means always staking exactly one unit regardless of confidence, is also a completely valid approach and is often recommended for beginners because it removes the temptation to overweight gut feelings on specific games.

Read More: Betting Predictions for Beginners: What to Look For

Looking for a second opinion before you bet? Check out our Predictions page to review today's Shurzy AI model and its impressive success rate.

How Do Units Work on Moneyline Bets?

Units get slightly more complicated when you're betting moneylines rather than standard -110 spread bets, and this is where a lot of prediction services get sloppy in ways that matter.

On a standard spread bet at -110, betting one unit risks one unit to win roughly 0.91 units. Clean and simple. But on a moneyline bet, the risk-reward ratio changes based on the odds. If you bet two units on a -220 favourite, you're risking two units to win less than one unit. If you bet one unit on a +180 underdog, you win 1.8 units if it lands.

A reputable prediction service accounts for this when calculating units won. They don't just credit a full two-unit win for a -400 favourite that barely paid out. They calculate the actual return at those odds and report it accurately. If a service isn't doing this, their unit totals are inflated and their track record is misleading. Always check whether a service calculates units won based on actual odds or just win and loss counts.

How Do You Use Units to Evaluate Prediction Services?

Units are an accountability tool as much as a betting tool. When a prediction service publishes their record in units, they're creating a standardised way for you to evaluate performance that removes the noise of different dollar amounts.

A service that finishes a season at plus 85.5 units with a 7.2% ROI gives you everything you need to make an informed decision:

• Multiply the unit total by your own unit size to see real-world profit potential
• Check the ROI to understand how efficiently they're generating profit relative to total action
• Divide units won by total bets to get a sense of average bet size and confidence distribution

What you want to avoid is a service that reports units won selectively, only counting big winners and quietly skipping losses. Legitimate services log every bet including losses and pushes. The unit total is only meaningful if the tracking is honest and complete.

Don't rely on gut feel alone. Head over to our Predictions page to see today's Shurzy AI projections and how they stack up across the board.

FAQ

Should my unit size ever change?

Your unit size should change when your overall bankroll changes significantly, either up after a profitable run or down after losses. Recalibrate to keep your unit at the same percentage of your current bankroll rather than the original amount.

Is a 5-unit play always a better bet than a 1-unit play?

Not necessarily. A 5-unit play reflects high confidence from the tipster or model. Whether that translates to a better outcome on any individual bet is still subject to variance. Higher unit plays should hit more often over a large sample, but no single bet is guaranteed.

What if a prediction service doesn't specify unit sizes?

That's a problem. If a service just says "bet the Lakers tonight" without any indication of stake sizing or confidence level, you have no way to apply proper bankroll management to their picks. Look for services that are explicit about their unit scale.

Can I follow a prediction service if my unit size is much smaller than theirs?

Yes. Units are specifically designed to be scale-independent. Multiply their unit totals by your own unit dollar amount to see what their record would mean for your bankroll.