# When the Dealer Wins Ties: What Each Version Costs

Some blackjack versions let the dealer win ties. Wizard of Odds prices each one: losing 17 ties alone costs 1.87 percent of expected return, and losing every tie costs 8.86.

Source: https://21trainer.app/articles/rules-where-ties-lose/
Updated: October 12, 2026
Publisher: 21 Trainer, makers of Blackjack Strategy: 21 Trainer, an educational blackjack trainer for iPhone and Android (App Store: https://apps.apple.com/app/blackjack-strategy-21-trainer/id6759527695, Google Play: https://play.google.com/store/apps/details?id=app.blackjacktrainer)

> **Educational content, not gambling advice.** 21 Trainer and this article teach blackjack strategy and card counting as skills, using virtual chips only. No strategy or counting system guarantees winnings, and nothing here encourages real-money play. If gambling is a problem for you or someone you know, call 1-800-522-4700 (National Council on Problem Gambling).

Of everything that can happen to a blackjack bet, the push gets the least attention. Your total matches the dealer's, your chips slide back, and the hand is over. It registers as a pause in the game rather than an outcome.

That quiet outcome is doing structural work. Commonly published figures, which our piece on [push frequency](https://21trainer.app/articles/blackjack-push-frequency/) covers in detail, put the push rate at roughly 8 to 9 percent of hands in standard six-deck games where the dealer stands on soft 17, about one hand in twelve. A rule that turns those hands into losses is reaching into a big slice of the game.

Several versions of blackjack reach exactly that way, and a push is the thing they reach for. The bet is returned untouched, so a push is a round where your money passes through the game without being touched by the house edge. Ties are not spread evenly either: the same article puts more of them on 20 than on any other total. Remember that, because it shows up again in the price tags below.

## The ladder: what each version costs

Wizard of Odds' blackjack rule variations page prices each tie rule as an effect on the player's expected return, with correct strategy adjustments, against one baseline: eight decks, dealer stands on soft 17, double on any first two cards, double after split, split to four hands. The ladder reads like this.

| Rule | Effect on expected return |
| --- | --- |
| Player loses 17 ties | -1.87 percent |
| Player loses 17 and 18 ties | -3.58 percent |
| Player loses 17 through 19 ties | -5.30 percent |
| Player loses 17 through 20 ties | -8.38 percent |
| Player loses 17 through 21 ties | -8.86 percent |
| Dealer bust on 22 is a push | -6.91 percent |

Working down the ladder total by total, each step adds a distinct price. Losing ties on 17 alone costs 1.87 percent. Adding 18 costs 1.71 more, since 3.58 - 1.87 = 1.71. Adding 19 costs 1.72 more, since 5.30 - 3.58 = 1.72. Adding 20 costs 3.08 more, since 8.38 - 5.30 = 3.08, by far the largest single step. Adding the final total, 21, costs only 0.48 more, since 8.86 - 8.38 = 0.48.

Two features of that arithmetic deserve a second look. The 17, 18 and 19 steps are nearly the same size, then the 20 step jumps, then the 21 step collapses. Both shapes fit where ties actually happen.

## Scale: the familiar rules beside it

The same Wizard of Odds page prices the rule changes players hear about constantly: blackjack paying 6 to 5 costs 1.39 percent, blackjacks paying 1 to 1 costs 2.27 percent, and the dealer hitting soft 17 costs 0.22 percent. Set those beside the tie rules and the perspective shifts fast.

Losing ties on 17 alone, at 1.87 percent, costs more than the 6 to 5 payout everyone complains about, at 1.39 percent, and that payout is the change our [3:2 versus 6:5 guide](https://21trainer.app/articles/blackjack-3-2-vs-6-5/) calls the worst common rule in blackjack. Losing every tie from 17 through 21, at 8.86 percent, costs about 6.4 times that payout change, since 8.86 / 1.39 is about 6.37, and about 40 times the soft 17 rule, since 8.86 / 0.22 is about 40.3. Even the full even-money cut on blackjacks, 2.27 percent, is smaller than the 20 step on its own, at 3.08 percent. The full tour of payouts sits in our piece on [payout variants](https://21trainer.app/articles/blackjack-payout-variants/).

## Why the steps are uneven

The 20 step is the giant, and the likely reason is that 20 is the total both sides reach most often. Two published figures line up with that. In a six-deck game, a dealer showing a ten finishes on exactly 20 about 36.8 percent of the time once the peek has ruled out a blackjack, per our article on [the dealer's final totals](https://21trainer.app/articles/dealer-final-total-distribution/). And hard 20, two ten-value cards, is the single most common two-card start, at 9.40 percent of starting hands in six decks, per our piece on [starting hand frequencies](https://21trainer.app/articles/starting-hand-frequencies/). Those figures are consistent with the step sizes rather than a proof of them, but the direction matches: the total both seats reach most often carries the biggest price.

The 21 step is the tiny one. The gap between Wizard of Odds' last two figures is 0.48 percent, and ties on 21 are comparatively rare, which fits the smallest step on the ladder.

## A dealer 22 that pushes instead of busting

The last row of the table works a different way. Instead of taking ties away from the player, this rule converts a would-be player win into a push: a dealer who finishes on exactly 22 does not bust, and every remaining player hand gets its money back. Wizard of Odds prices it at -6.91 percent, larger than the 6 to 5 payout, larger than even-money blackjacks, and larger than any single step of the tie ladder.

The rule changes strategy too. In ordinary blackjack a dealer bust always pays every live hand, which is what makes standing on a stiff total against a weak upcard correct. When some of those busts become pushes, the value of standing shifts, and hit and stand decisions shift with it.

## Where these rules actually appear

In the variants this site has covered, tie rules do not travel alone. They show up as payment for something the variant gives away, which is the pattern our [variants comparison](https://21trainer.app/articles/blackjack-variants-compared/) describes: the extras are purchases, not gifts.

[Double Exposure](https://21trainer.app/articles/blackjack-double-exposure/) shows both dealer cards face up and charges for the information: the dealer wins all ties except tied blackjacks, which push or go to the player depending on the casino, and a player blackjack pays even money instead of 3 to 2. The house edge depends heavily on which combination of the remaining rules a property chooses, which is why our article on the game declines to quote a single figure for it.

[Pontoon](https://21trainer.app/articles/pontoon-blackjack-rules/) goes further and takes every tie, while showing neither dealer card. It pays for that with 2 to 1 payouts on pontoons and five-card tricks, and the package nets out at a Wizard of Odds house edge of 0.38 percent under what the Wizard of Odds page on Pontoon calls the usual rules: eight decks, dealer hits soft 17, ties losing.

The dealer 22 push helps fund two of the most generous-sounding offers in the family. [Blackjack Switch](https://21trainer.app/articles/blackjack-switch-variant/) lets you swap the second card between two hands, and [Free Bet Blackjack](https://21trainer.app/articles/free-bet-blackjack/) funds your doubles on 9, 10 and 11 and most of your splits; both push rather than pay when the dealer lands on exactly 22, with a player blackjack the lone exception. Switch charges a second time by paying even money on a player blackjack. Per our Free Bet article, Wizard of Odds puts that game's house edge at 1.04 percent with correct play under its common rules, six decks with the dealer hitting soft 17 and blackjack paying 3 to 2, and that figure prices the whole package, soft 17 rule included, not the 22 rule alone.

That last point is the one to keep. A rule's cost in isolation is not the variant's house edge. Each game bundles the tie rule with compensations, and the published edges were computed under different rule assumptions, so the honest move is comparing the trades rather than lining up the decimals.

## What to ask at an unfamiliar table

Before anything else at a table or variant you do not know, find out what happens on a tie, and what happens when the dealer makes 22. Those two questions separate the standard game from the variants covered here: Double Exposure and Pontoon charge on ties, while Blackjack Switch and Free Bet leave ordinary ties alone and charge on the dealer's 22. If either answer differs from the standard game, a standard chart no longer fits, and every rule in the table above costs more on its own than a 6 to 5 payout.

That judgment is trainable. 21 Trainer plays standard blackjack, where ties push, with full rule customization across deck count, soft 17, double after split, surrender and split limits, and a house edge calculator that shows how ordinary rule changes stack. Learning what standard rules cost makes an unfamiliar rule card easier to read.

## Frequently asked questions

### Does the dealer win ties in blackjack?

Not in standard blackjack. A tie there is a push: the bet comes back and the hand is over. The tie-losing versions live in variants. Double Exposure has the dealer win all ties except tied blackjacks, Pontoon has the dealer take every tie, and Blackjack Switch and Free Bet Blackjack push when the dealer finishes on exactly 22. In each of those games the rule is part of a trade, paid for with other liberal rules.

### How much does it cost when the dealer wins ties?

Per Wizard of Odds, against a baseline of eight decks with the dealer standing on soft 17, losing ties on 17 alone costs 1.87 percent of expected return. Adding 18 and 19 brings the cost to 5.30 percent, adding 20 brings it to 8.38, and losing every tie from 17 through 21 costs 8.86 percent. Those figures assume correct strategy adjustments.

### Why is losing ties on 20 the most expensive step?

Most likely because 20 is the total both sides reach most often. In six decks, a dealer showing a ten finishes on exactly 20 about 36.8 percent of the time once the peek has ruled out a blackjack, and hard 20, two ten-value cards, is the most common two-card start at 9.40 percent of hands. That is consistent with the step sizes rather than a proof of them: the 20 addition costs 3.08 percent against 1.87 for 17 alone.

### What happens when the dealer gets 22 in Free Bet Blackjack or Blackjack Switch?

The dealer does not bust. A total of exactly 22 pushes against every player hand still in play, so hands that would have beaten a broken dealer simply get their money back. The one exception in both games is a player blackjack, which still wins. The rule funds the free doubles and splits in Free Bet and, together with even money blackjacks, the card switch in Blackjack Switch.

### Can you practice tie rules in 21 Trainer?

The app plays standard blackjack, where ties push and the bet is returned. It does not simulate the tie-losing variants. What it does offer is the standard game with full rule customization, deck count, soft 17, double after split, surrender and split limits, plus a house edge calculator, so the cost of ordinary rule changes can be seen directly.

## Keep reading

- [How Often Does a Blackjack Hand Push?](https://21trainer.app/articles/blackjack-push-frequency/)
- [Double Exposure Blackjack: Both Dealer Cards Up](https://21trainer.app/articles/blackjack-double-exposure/)
- [Pontoon: The British Cousin of Blackjack](https://21trainer.app/articles/pontoon-blackjack-rules/)