Expected Loss Per Hour — Formula and Examples

If you know three numbers, you can estimate the long-run hourly cost of blackjack:

[ \text{Expected Loss per Hour} = \text{Average Initial Bet} \times \text{Hands per Hour} \times \text{House Edge} ]

For example, suppose you average $25 per initial hand, play 70 hands per hour, and the game plus your strategy produces a 0.50% house edge:

[ 25 \times 70 \times 0.005 = 8.75 ]

The theoretical loss is $8.75 per hour.

That does not mean you should expect to lose $8.75 after one hour. You might win $300, lose $400, or finish almost even. Expected loss is an average cost over repeated play; blackjack variance explains why one session can sit far away from that average.

What each part of the formula means

Average initial bet

This is the average amount committed before the cards are dealt for each hand. If you flat-bet $25, the number is easy: $25.

If your bets vary — $15, $25, $50, then back to $25 — use the average of the initial wagers over the period you are measuring.

Why “initial” matters: standard blackjack house-edge figures are usually expressed relative to the original wager even though correct play can later add money through doubles and splits. Those extra wagers are part of the game’s expected value calculation rather than simply being multiplied into the denominator again.

Hands per hour

This is the number of rounds you personally receive in an hour. It changes with:

• number of players at the table;
• dealer speed;
• shuffling method;
• side-bet settlement;
• buy-ins and fills;
• disputes and player decisions;
• whether you play one spot or more than one;
• live versus electronic format.

Do not treat one hands-per-hour figure as a law of blackjack. It is an operating assumption.

House edge

The house edge is the average loss divided by the initial wager under a specified set of rules and a specified strategy standard.

It is not “the blackjack house edge” as one permanent number. It changes with rules such as:

• 3:2 versus 6:5 blackjack payout;
• dealer hitting or standing on soft 17;
• double-after-split availability;
• surrender;
• number of decks;
• restrictions on splitting or resplitting;
• player strategy errors.

Use Blackjack House Edge by Rules before putting a percentage into the formula.

First calculate action, then apply the edge

The same calculation can be written in two steps.

Step 1: initial-bet action per hour

[ \text{Hourly Initial Action} = \text{Average Initial Bet} \times \text{Hands per Hour} ]

With a $25 average initial bet and 70 hands:

[ 25 \times 70 = 1{,}750 ]

You have put $1,750 of initial-bet action through the game during that hour.

Step 2: theoretical loss

[ \text{Expected Loss} = \text{Initial Action} \times \text{House Edge} ]

At a 0.50% edge:

[ 1{,}750 \times 0.005 = 8.75 ]

This two-step version is useful because it shows what can change the cost: more money per hand, more hands, or a worse edge.

A comparison table

The table below uses 70 hands per hour only as an illustration. The actual pace at your table may be different.

The numbers are linear. Double the average wager and you double expected loss. Double the number of hands and you double expected loss. Double the house edge and you double expected loss.

This is why a “small” rule disadvantage can become expensive at high action.

Session length converts hourly cost into session cost

If the assumptions remain roughly stable, multiply by hours:

[ \text{Expected Session Loss} = \text{Expected Loss per Hour} \times \text{Hours Played} ]

Using the earlier $8.75 hourly estimate:

Again, this is theoretical loss, not a forecast of your closing chip stack.

A player who wins $500 during a four-hour session did not “beat” a $35 expectation in a statistical sense that proves an edge. A player who loses $500 did not reveal a 14% house edge either. Both can be ordinary short-run results around a much smaller long-run expectation.

Rule quality can matter more than table minimum

Suppose two tables both run around 70 hands per hour and you plan to average $25.

• Table A plus your strategy produces a 0.50% edge.
• Table B produces a 1.50% edge.

Then:

[ 25 \times 70 \times 0.005 = 8.75 ]

versus

[ 25 \times 70 \times 0.015 = 26.25 ]

Same bet size. Same pace. Three times the expected hourly loss.

That is why the payout and rule placard deserve attention before you sit down. The difference between 3:2 and 6:5 blackjack can be more consequential than shaving a few hands from the session.

Slower play reduces cost if everything else stays equal

If your average bet and house edge stay fixed, fewer hands means less expected loss.

At $25 and a 0.50% edge:

• 40 hands/hour: $5.00 expected loss
• 60 hands/hour: $7.50
• 80 hands/hour: $10.00
• 100 hands/hour: $12.50

This is not a strategy to “beat” blackjack. It is simply less action. The same idea applies to breaks, shorter sessions, or choosing not to play every possible hand.

For players who think only in terms of percentage edge, this is one of the most useful corrections: house edge is a rate; total wagering determines how much money that rate is applied to.

Variable bets require a better average

A common mistake is to use the table minimum as the average bet.

Suppose you play 20 hands at $25, 20 at $50, and 20 at $100. Your initial action is:

[ 20(25) + 20(50) + 20(100) = 3{,}500 ]

Over 60 hands, the average initial bet is:

[ \frac{3{,}500}{60} \approx 58.33 ]

If you instead plug $25 into the hourly formula because “it was a $25 table,” you would dramatically understate the cost.

For changing stakes, the most accurate simple method is:

[ \text{Expected Loss} = \sum_i \left(\text{Initial Bet}_i \times \text{House Edge}_i\right) ]

If the rule set and strategy are constant, the edge term is the same for each hand and the sum reduces to total initial action multiplied by the edge.

If the edge itself changes — for example an advantage player varies bets with a count — one flat house-edge percentage is no longer an adequate model.

Side bets should be calculated separately

A $5 side wager added to a $25 blackjack hand is not “just part of the $25 game.” It has its own paytable, hit frequency, volatility, and house edge.

A clean estimate is:

[ \text{Total Expected Loss} = \text{Main-Game Expected Loss} + \text{Side-Bet Expected Loss} ]

Suppose the main game costs an estimated $8.75 per hour, and you also make a $5 side bet for 70 hands at an illustrative 8% edge:

[ 5 \times 70 \times 0.08 = 28 ]

The side bet would add $28 of theoretical hourly loss, taking the combined estimate to $36.75.

The example is not saying every blackjack side bet has an 8% edge. It shows why a small chip placed beside every main wager can dominate the economics if the side-bet edge is much larger. Use the actual paytable and the relevant analysis from Blackjack Side Bets Overview or House Edge When Side Bets Are Added.

Why casino “theo” may not match your own calculation

Casinos often estimate a rated player’s theoretical value using some version of:

[ \text{Theo} \approx \text{Average Bet} \times \text{Estimated Hands} \times \text{Assumed House Edge} ]

But the inputs may be operational estimates rather than a reconstruction of every exact hand. A rating system may use an assumed pace and an average edge for the game rather than crediting every basic-strategy decision perfectly.

The Wizard of Odds house-edge reference explicitly defines house edge as average loss relative to the initial bet and also publishes an illustrative casino comp table using 70 blackjack hands per hour and an assumed 0.75% edge. Those figures are useful examples of a rating model, not universal rules for every casino.

This distinction matters when comparing player expected loss with casino-rated theoretical loss. They are related concepts, but the casino may be using different assumptions.

Three sessions with the same player can have very different theoretical cost

A useful way to see the formula is to hold the player constant and change one or two operating conditions. These examples use illustrative edges, not claims about a specific casino.

Session A: lower bet, good rules, full table

Suppose the player averages $15, receives about 50 hands per hour, and the rule/strategy combination is estimated at a 0.50% edge.

[ 15 \times 50 \times 0.005 = 3.75 ]

Expected loss: $3.75 per hour. Over three hours the theoretical loss is $11.25.

Session B: same rules, faster game

The player keeps the $15 average bet and 0.50% edge but moves to a faster format producing 100 hands per hour.

[ 15 \times 100 \times 0.005 = 7.50 ]

Nothing became mathematically worse about each individual wager. The hourly cost doubled because the player bought twice as many decisions.

Session C: higher bet and weaker conditions

Now the player averages $50, receives 70 hands per hour, and the effective edge is 1.50%.

[ 50 \times 70 \times 0.015 = 52.50 ]

Expected loss: $52.50 per hour. Over three hours that becomes $157.50 in theoretical loss.

The contrast between $3.75 and $52.50 per hour is not a prediction that one player will visibly lose fourteen times faster in a short session. It shows how the long-run price changes when action and edge change together.

If you play two spots, count the action on both

Playing two hands at once changes the calculation because you are putting more initial wagers into action each round. If you wager $25 on each of two spots for 50 rounds per hour, the hourly initial action is:

[ 25 \times 2 \times 50 = 2{,}500 ]

At a 0.50% edge:

[ 2{,}500 \times 0.005 = 12.50 ]

A player who says “I only play $25 blackjack” can therefore understate exposure if the number of simultaneously active hands is ignored. The same principle applies when back-betting or multiple-position play is permitted: expected cost follows the amount of qualifying action, not the label printed on the table minimum sign.

Expected loss is not a bankroll requirement

If your calculation says $10 per hour, bringing $20 does not mean you have “two hours of bankroll.” Variance is much larger than expected loss over short periods.

A $25 blackjack player can easily experience multi-unit swings within a few hands because of:

• consecutive normal losses;
• doubles;
• splits;
• blackjacks;
• pushes;
• dealer naturals;
• multiple hands played at once.

Expected loss answers “What is the long-run average cost of this action?” It does not answer “How much can I lose tonight?” or “What bankroll guarantees I survive four hours?”

Those are variance and risk questions. Continue with Blackjack Variance Explained and the Bankroll Risk Calculator if that is what you need.

Five errors that make hourly-loss estimates useless

Using the wrong house edge

A 0.5% example is not permission to assume your table is 0.5%. Rules and strategy determine the input.

Using the minimum instead of the average bet

A $25 table can have a $60 average bet if you press frequently.

Ignoring side bets

Small repeated wagers can add large expected cost when their edge is high.

Assuming a fixed table speed

A heads-up electronic game and a full live table do not produce the same number of decisions per hour.

Treating expected loss as a promised result

The formula estimates a mean. It does not shrink short-term blackjack into a smooth subscription fee.

A practical way to use the number

Expected loss per hour is most useful for comparing choices before the session.

You can ask:

• What happens if I play $15 instead of $25?
• What does a worse payout do to cost?
• How much does a four-hour session add compared with two hours?
• What does a side bet add if I make it every hand?
• How much action am I creating by playing faster?

For a quick calculation, use the Expected Loss Calculator or Session Loss Calculator.

The important lesson is not that blackjack costs exactly a certain number of dollars each hour. It is that bet size, pace, rules, strategy, and time multiply together. A low percentage edge can still become expensive when the amount of action becomes large, while short-run wins and losses can remain far noisier than the expected cost.