How to Calculate What a Poker Game Really Takes Off the Table

You cannot reliably identify the lower-cost poker game from its advertised rake percentage and cap alone. Apply the complete fee schedule—including eligibility, rounding, occupancy-based caps, and separate jackpot drops—to the full distribution of pots and zero-fee hands. Then compare average total drop per qualifying pot, dealt hand, hour, or 100 hands.

The short answer: you need the schedule and the distribution

A game advertised as “10% capped at $4” is not automatically more expensive than one advertised as “5% capped at $6.” The first takes more from smaller pots but stops sooner. The second takes less initially but continues rising until it reaches the higher cap. Which schedule removes more money depends on how often pots occur at each size.

At minimum, collect:

• Base rake percentage
• Applicable per-pot cap
• Rounding or stepped-collection method
• Cap tier for the number of players dealt in
• Point at which a hand becomes eligible for rake
• Any minimum-pot requirement
• Jackpot or promotional-drop amount
• Trigger for each additional charge
• Frequency of pots in each size range
• Frequency of dealt hands that incur no fee

The percentage matters most when pots usually remain below the cap threshold. The cap matters more when pots frequently grow large enough to reach it. Occupancy can also change the maximum, so “5% capped at $6” may be incomplete if a different cap applies when the table is short-handed.

The Reddit comparison asking which game has the lower overall drop cannot support a winner. It supplies neither a verified fee schedule nor a pot-size distribution, and its jackpot assumptions are not established.

This guide calculates gross money removed from the table. That is not automatically the amount attributable to one player. A personal estimate requires an additional attribution method, such as assigning the fee to the pot winner, dividing it among contributors, or allocating it in proportion to each player’s contribution.

Decode every part of the fee sheet first

Do not treat “rake,” “drop,” and “time” as interchangeable terms. They describe different collection methods:

• Percentage pot rake: A percentage of an eligible pot, generally limited by a per-hand maximum.
• Rake cap: The maximum base rake that can be taken from one pot under the applicable schedule.
• Jackpot or promotional drop: A separately listed charge used to fund a jackpot or promotion. It may be added on top of base rake.
• Fixed drop: A set charge triggered by a specified event rather than calculated as a percentage of the pot.
• Dead drop: A fixed fee assigned before or at the start of a hand, sometimes according to button position.
• Timed collection: A charge assessed at regular intervals instead of being deducted as a percentage of each pot.

A base-rake cap does not automatically limit a separate promotional charge. If a fee sheet says “10% up to $5 plus $1,” establish whether the extra dollar is added to the $5 maximum or included within it.

You also need the exact trigger. A charge might apply when cards are dealt, when the pot reaches a minimum, when a flop is seen, or when the hand reaches the turn or river. Street-based charges require street-reach frequencies as well as pot sizes.

“No flop, no drop” means that qualifying hands ending before the flop incur no pot rake. It is not universal, and its exact application can vary by room, game, and stake. Caps may also change with the number of players dealt in, while some schedules treat particular preflop pots differently; examples of those variations appear in GTO Wizard’s overview of rake structures.

Record the complete schedule before doing any arithmetic:

Fee-sheet area	Details to record
Game	Stake, variant, and table format
Occupancy	Players dealt in and applicable tier
Base fee	Percentage, cap, and rounding increment
Eligibility and extras	Qualifying point, minimum pot, jackpot amount and trigger, collection order

Confirm whether multiple deductions are calculated sequentially or from the same original pot. Also determine how the room defines pot size and handles returned uncalled bets.

Keep these items outside the initial gross-drop calculation:

• Dealer tips
• Rakeback or rewards
• Promotional prizes received
• Jackpot payouts
• Food, hotel, or other comps

They can matter to a player’s net economics, but combining them with mandatory table deductions obscures how much the fee schedule itself removes.

Use a piecewise formula, not the advertised percentage alone

For one hand, define:

• P = eligible pot size
• r = rake rate as a decimal
• C = applicable base-rake cap
• R(x) = the room’s rounding or stepped-collection function
• J = total separately triggered fixed, jackpot, or promotional drops

The simplified base-rake formula is:

Base rake = \begin{cases} 0, & if the hand is ineligible \ \min(C, R(rP)), & if the hand is eligible \end{cases}

Total drop is:

Total drop = Base rake + J

Here, J=0 when no separate trigger is met. If the schedule contains several fixed charges, calculate each trigger independently and add the charges that apply.

Before rounding, the nominal cap threshold is:

Nominal cap threshold = C ÷ r

For 10% rake capped at $4:

$4 ÷ 0.10 = $40

Under this simplified schedule, the room takes $1 from a $10 pot, $2 from $20, $3 from $30, and $4 from $40 or more, matching the illustration in Red Chip Poker’s rake explainer.

The exact point at which the cap binds can shift if the room collects only in fixed increments, rounds up or down, changes the cap by occupancy, or uses a different eligible-pot definition. Treat C/r as a useful starting point, not a substitute for the complete rules.

The effective fee percentage for a hand is:

Effective fee percentage = Total drop ÷ P × 100

With 10% capped at $4 and no separate drop:

Pot	Base rake	Effective percentage
$10	$1	10%
$20	$2	10%
$40	$4	10%
$100	$4	4%

Once the cap binds, the base-rake amount remains constant while the pot grows. Its effective percentage therefore declines. A separate fixed drop must still be included when calculating the total effective fee.

Use one pot definition throughout the analysis. A practical default is the eligible pot before deductions, excluding returned uncalled bets, unless the room’s official rules specify otherwise.

Build a distribution that includes fee triggers and zero-drop hands

The most auditable method is to log hands individually and group them only after calculating each hand’s charge. A spreadsheet can use these columns:

Hand number | Pot size | Flop seen | Final street | Players dealt in
Base rake | Jackpot drop | Total drop | Cap reached

Pot size alone is insufficient. “Flop seen” and “final street” capture possible fee triggers, while “players dealt in” identifies the applicable occupancy tier. Recording the actual deductions also lets you test whether your formula reproduces what the room collected.

Keep preflop-ending hands in the dataset. If they qualify for no-flop-no-drop, enter zero rather than deleting them. Otherwise, a dollars-per-dealt-hand result will be inflated because its denominator excludes free hands.

For n dealt hands, the average total drop is:

Average drop per dealt hand = \sumi₌₁ⁿ (Base rake_i + Extra drop_i) ÷ n

After calculating charges at hand level, you can group pots into practical ranges such as $0–$19.99, $20–$39.99, $40–$79.99, and $80 or more. Preserve the observation count or weight for every bin.

For binned data:

E[Drop] = \sum_i w_i\,E[Drop\mid bin i]

Here, w_i is the frequency of bin i, and E[Drop\mid bin i] is the observed or modeled mean charge within that bin—not the fee on an arbitrarily selected midpoint.

Ideally, calculate that conditional mean from hand-level records. If only aggregate data are available, use exact pot values or bins narrow enough that every observation receives the same base charge. If a bin crosses a cap threshold, rounding boundary, stepped-collection point, or fee trigger, split it. If that is impossible, model the within-bin pot and trigger distribution explicitly and disclose the assumption.

The weights must sum to 1. If a $1 jackpot charge applies to 60% of observations in a bin, its expected contribution is $0.60 for that bin, not the full dollar.

Calculate cap-hit rate as:

Cap-hit rate = Capped qualifying pots ÷ All qualifying pots

Always name the denominator. A 30% cap-hit rate among qualifying pots is not the same as 30% of all dealt hands.

Do not apply the formula only to the average pot. Capped rake is nonlinear, so:

Fee(average pot) ≠ average of individual pot fees

A game containing many small pots and a few very large pots can have the same average pot as a tightly clustered game but produce a different average drop.

Label the source of the distribution:

• Observed: Calculated from recorded hands
• Modeled: Generated using a stated model
• Assumed: Chosen for an illustration or sensitivity test
• Proportionally scaled: Adapted from another stake or format

For example, the PrimeDope rake calculator says its no-limit hold’em estimates use observed distributions at selected stakes, proportionally scale stakes outside that sample, and apply modeled adjustments to other formats. Because the underlying hand histories and calculations are not published there for independent checking, those outputs should be treated as the publisher’s modeled estimates rather than audited samples for every listed game.

Worked example: the same schedules under two pot distributions

This is a hypothetical exercise, not a comparison of current poker rooms.

Initially ignore rounding and separate drops:

• Schedule A: 10% capped at $4
• Schedule B: 5% capped at $6

Schedule A reaches its nominal cap at a $40 pot. Schedule B reaches its nominal cap at $120.

Small-pot-heavy distribution

Assume qualifying pots are:

• 50% at $10
• 30% at $20
• 20% at $100

Schedule A charges $1, $2, and $4:

(0.50 × $1) + (0.30 × $2) + (0.20 × $4) = $1.90

Schedule B charges $0.50, $1, and $5:

(0.50 × $0.50) + (0.30 × $1) + (0.20 × $5) = $1.55

Schedule B is cheaper by $0.35 per qualifying pot under this distribution.

Larger-pot-heavy distribution

Now assume:

• 20% at $10
• 30% at $20
• 50% at $100

Schedule A averages:

(0.20 × $1) + (0.30 × $2) + (0.50 × $4) = $2.80

Schedule B averages:

(0.20 × $0.50) + (0.30 × $1) + (0.50 × $5) = $2.90

The ranking reverses. Schedule A is cheaper by $0.10 per qualifying pot because its lower cap applies to a larger share of this distribution.

Now add a hypothetical $1 jackpot drop to Schedule B and assume every example pot triggers it:

Distribution	A base rake and cap-hit rate	B base rake and cap-hit rate	B total with $1 jackpot
Small-pot-heavy	$1.90; 20%	$1.55; 0%	$2.55
Larger-pot-heavy	$2.80; 50%	$2.90; 0%	$3.90

The cap-hit denominator is all qualifying pots in each hypothetical distribution. Schedule B never reaches its $6 cap because the largest modeled pot is $100.

If only a fraction q of the observations triggers the $1 jackpot drop, add:

$1 × q

A 40% trigger frequency therefore adds $0.40, not $1, to the average.

As a separate historical illustration, a 2012 Southern California example described a $5 house drop plus a $1 jackpot drop once a flop was dealt. Removing $6 from a $20 pot consumed 30% of that pot, according to Crush Live Poker’s account. That example demonstrates how heavily a fixed charge can weigh on a small pot; it is not evidence of a current or typical fee schedule.

Convert the result into useful comparison metrics

Report dollars per qualifying pot and dollars per dealt hand separately.

If 70 out of 100 dealt hands qualify and those qualifying pots average $3 in total drop:

Drop per qualifying pot = $3

Drop per dealt hand = 70 × $3 ÷ 100 = $2.10

The second figure includes the 30 zero-drop hands. It is the correct starting point for table-level hourly and per-100-hand calculations.

For observed table pace H:

Table drop per hour = Average drop per dealt hand × H

If the observed sample contains 28 hands per hour:

$2.10 × 28 = $58.80

Do not import a supposed universal live-poker pace. Record the actual pace of the game being evaluated.

To express table-level drop in big blinds per 100 hands:

Table drop in bb/100 = 100 × average dollars dropped per dealt hand ÷ big blind

At $1/$2 with an average drop of $2.10 per dealt hand:

100 × $2.10 ÷ $2 = 105 bb/100

That is the amount removed from the whole table, not one player’s personal rake burden. Personal estimates require a declared attribution method:

• Winner-based: Assign the drop to the pot winner
• Contributed: Divide it among players who contributed to the pot
• Weighted-contributed: Allocate it in proportion to each player’s contribution

Report the mean and median drop alongside the cap-hit rate. Two games can have similar averages while imposing very different patterns: one may take modest amounts from most pots, while another takes nothing from many hands and a large amount from a smaller group.

For timed collection, put both systems on the same basis. Compare the observed hourly timed charge with the pot-raked game’s average drop per dealt hand multiplied by its observed hands per hour. Keep table-level charges separate from per-seat charges.

Why the named real-game comparison remains unresolved

When accessed for this article on September 12, 2026, the supplied PokerStars rake URL displayed a FanDuel promotional homepage rather than a usable poker fee schedule. On that access date, it supplied no applicable rake percentage, cap, jackpot rule, rounding method, occupancy tier, or pot distribution.

The Reddit page is likewise only an example of an unresolved comparison. It provides neither verified rules for the anonymous games nor auditable hand records from which to calculate pot-size and trigger frequencies. Current PokerStars fees—or the fees of either anonymous room—should not be inferred from those pages.

A defensible real-game verdict requires:

1. Dated official fee sheets for the exact game and stake
2. Applicable caps for every occupancy tier
3. Exact rounding rules and collection order
4. Base-rake and promotional-drop triggers
5. A consistent definition of pot size
6. An auditable sample containing qualifying and zero-drop hands
7. Pot-size, street-reach, and cap-hit frequencies

Lower gross fees do not guarantee a better or profitable game. Opponent skill, table pace, stack depth, tips, individual decisions, and ordinary poker variance also affect results. Treat this calculation as cost information, not a promise of winnings, and play only within personal limits.

Choose a fee structure only after applying its complete rules to an observed—or clearly labeled assumed—distribution. Report gross drop, eligibility, and cap-hit frequency separately. If any schedule field or distribution input remains unknown, publish a sensitivity range rather than naming a cheapest game.