# Multi-Hand Video Poker Strategy: What Changes With Several Hands

> Multi-hand video poker strategy explained: choose a hold, compare round returns and spread, and check the rules and total stake on your game.

- Source: https://toolsgambling.com/blog/multi-hand-video-poker-strategy
- Publisher: ToolsGambling
- Published: 2026-10-26
- Updated: 2026-10-26
- Cite as: "Multi-Hand Video Poker Strategy: One Hold, Many Hands", ToolsGambling, https://toolsgambling.com/blog/multi-hand-video-poker-strategy

Multi-hand video poker is a format in which several hands of the same game are played from one deal, for example 5 hands in IGT’s Five Play. How holds carry between hands depends on the game. The strategy principle stays the same with 5 hands or any other count: under the same rules and strategy, expected return per hand does not change; the total stake per round and the spread of its result change.

The practical answer to multi-hand video poker strategy is to choose the best hold for the original hand, then assess the cost and spread of the whole round. Adding hands does not improve a poor hold or increase expected return per coin staked.

## What Multi-Hand Video Poker Is

Multi-hand video poker plays several hands of one game from a single deal. In Five Play, those hands start from the same original cards and the same hold decision; other games may work differently. You are applying one strategic choice to several draws, rather than solving a separate starting-hand problem for every hand.

A hand receives its own final combination and payout under the paytable. A round contains all the hands played together, so its total stake and total payout are larger than the corresponding amounts for a single hand at the same stake.

That distinction matters when comparing games. A larger payout across a round does not by itself indicate a better return: more coins were also staked. The paytable and game rules determine the value of a hold; the number of hands determines how widely that decision is repeated.

## How a Deal Works: Holds and Draws

In Five Play, the hold selected on the main hand carries over to the other hands, and replacement cards are dealt separately to each hand. That is the mechanic described in the Pokernews review of Five Play, pokernews.com, rather than a confirmed rule for every multi-hand game.

The sequence is straightforward:

- An original hand is dealt.
- You select the cards to hold.
- Those held cards appear in the other hands under the game’s rules.
- Each hand receives replacement cards for the discarded cards.
- Each completed hand is evaluated against the paytable.

Other descriptions may provide less detail. The support article for 7-Play, champion.helpscoutdocs.com, describes playing from one to seven hands and setting the stake per hand. It does not explain whether holds carry between hands. That omission leaves part of the mechanic unconfirmed; it is not evidence that the game follows the Five Play model.

## One Strategy for All Hands

With identical rules and the same paytable, the best hold is determined by the original hand, not by how many hands you play. Repeating the hold across more hands changes the scale of the result, not which hold has the highest expected return.

**Calculation assumption: each hand draws its replacement cards from its own copy of the deck without the cards of the original deal.** This is an assumption, not a verified statement about the internal operation of all multi-hand machines.

Consider **K♠ K♥ 7♦ 4♣ 2♠**. In the educational calculation, the best hold is **the pair of kings**. For Jacks or Better 9/6 at the maximum bet of five coins, its expected return is **1.54 per coin staked**, rounded to two decimal places from **1.5365**.

For the same original hand, the paytable changes the estimate as follows:

- **8/5:** 1.53 per coin.
- **7/5:** 1.52 per coin.
- **6/5:** 1.51 per coin.

The pair of kings remains the best hold in each of these cases. These are estimates from the paytable, not guarantees of the next result. Expected return describes the average payout across the possible draws, not a payout you must receive on this round.

The calculated variance per hand for this particular deal and hold is **2.58**. It applies to this example, rather than to all possible original hands.

Different game rules can change the hold itself. The [Deuces Wild strategy](https://toolsgambling.com/blog/deuces-wild-video-poker-strategy) guide explains decisions under a different ruleset, illustrating why the original hand must be evaluated within the correct game.

## What Changes With Several Hands

More hands increase the total stake, expected round payout, and absolute spread of round results. They do not raise expected return per coin when the rules, paytable, stake per hand, and strategy remain unchanged.

For one hand we use the exact Jacks or Better 9/6 values: expected return under optimal play is **99.54%** (the bankroll guide gives 99.5439%) and the standard deviation is **4.42** per coin at the maximum bet. Everything in this article is per coin at the maximum bet: five coins per hand, with the royal flush paying 800 to 1.

For **N hands**, the table gives the expected round return and round spread: the sum of the payouts of all hands, each measured per coin at the maximum bet. With 5 hands, the expected round return is **4.98** and the round spread is **11.6**. The key redraw assumption is an independent deck copy per hand, with the shared deal already included. Additional correlation between hands would raise the spread, but the calculation does not quantify that increase; the shared-deal shares below show the dependence already included.

| Hands | Round return (per coin at the maximum bet) | Round spread (SD, per coin at the maximum bet) |
|---|---|---|
| 1 | 0.995 | 4.4 |
| 5 | 4.98 | 11.6 |
| 10 | 9.95 | 19.1 |
| 25 | 24.89 | 40.3 |

![Table: return and spread per round for 1, 5, 10 and 25 hands on 9/6, per coin at the maximum bet (educational calculation)](https://toolsgambling.com/media/article/2026/10/834cb928c4fe59e8adf07ec49239c81c.webp)
*Educational calculation, assuming each hand draws its replacement cards independently: return grows in direct proportion to the number of hands, spread grows more slowly.*

These are whole-round payouts, not net profits. The higher return column must be read alongside the larger amount staked. To obtain a net result, the round stake must be deducted from the payout.

Spread here means standard deviation, abbreviated SD. It is not interchangeable with variance. The shared deal accounts for approximately **35%** of round variance with 5 hands, **52%** with 10 hands, and **73%** with 25 hands. These shares are an estimate, not exact values: they rest on a sample of random deals in which rare large payouts are poorly represented, so the exact contribution of a dealt royal flush was added. As the hand count rises, the common starting cards account for a larger share of the round's variation.

## Bankroll for Several Hands

Plan your budget around the full round stake, not just the stake displayed for one hand: **round stake = number of hands × stake per hand**.

The [video poker bankroll guide](https://toolsgambling.com/blog/video-poker-bankroll-strategy) continues the relationship between budget and coin-in: keeping the stake per hand unchanged while adding hands increases the amount staked each round. That is the budgeting consequence to resolve before choosing a hand count, rather than treating a larger expected payout as extra spending capacity.

## What the Calculation Cannot Show

The calculation cannot identify an unknown machine's dealing method, predict the next result, or establish an exact return from a short session. Its results describe this calculation and its sample.

A Las Vegas Advisor Q&A discusses a dispute over one random-number generator per line versus separate decks, while noting that the regulator does not disclose its methods. That discussion does not establish a universal dealing rule.

Do not mix exact values with estimates. For one hand, the standard deviation of **4.42** is the exact value from the bankroll guide. For 5, 10 and 25 hands the spread comes from a formula in which the shared-deal share is an estimate, so those values are approximate and may shift slightly.

Finally, the published return figures are rounded. The table is computed from the exact return of 0.995439 per coin and shown rounded: do not multiply the rounded 0.995 to reconstruct more precise totals or “correct” the rows. Use the table values as shown.

## How to Check It on Your Game

Check the game’s rules and paytable first, evaluate the hold on one hand, and then assess the full round stake. This separates the card decision from the decision about how many hands to fund.

1. **Read the game rules.** Look for an explanation of whether holds carry across hands and how replacement cards are dealt. Leave unanswered details unresolved rather than borrowing rules from another game.
2. **Inspect the paytable.** Confirm the payouts relevant to the game you are evaluating. Similar graphics or the same hand count do not establish identical payouts.
3. **Check the original hand.** Use the [video poker calculator](https://toolsgambling.com/poker/video-poker-calculator) to check a hold on one hand. That check evaluates the card decision; it does not verify a machine's internal dealing method.
4. **Check the stake per hand and hand count.** Establish the full cost of the round before comparing its expected payout with another format.
5. **Separate stake, payout, and net result.** The stake is committed to play; the payout comes from the final combinations; the net result accounts for both.

## FAQ

### Do you need a separate strategy for each hand?

With the same rules and paytable, choose the hold from the original hand. The number of hands alone does not change the best hold.

### Do more hands mean a higher return per coin?

No. The expected return for the whole round increases, but so does the total stake. Adding hands does not increase the return per coin staked.

### Can the hands be treated as fully independent?

No. Even with separate copies of the deck for replacement cards, the hands remain connected through the shared deal and the same hold.

### What is the expected return of Jacks or Better 9/6 under optimal play?

99.54% per coin at the maximum bet (0.9954), with a standard deviation of 4.42 for one hand. These are the exact values from the site’s bankroll guide. The number of hands does not change the return per coin; it changes the round’s total stake and spread.

### Is holding the pair of kings correct in every video poker variant?

Not necessarily. The example applies to the paytables used in the calculation. With different rules, the best hold needs to be checked separately.