Live Dealer Cards Dealt Face-Down Stall Bet Sizing 21%
Players who sit at live dealer blackjack tables where the dealer's second card is dealt face-down size their bets roughly 21% larger than players at face-up tables, according to a 14-month analysis of 2.3 million wagering decisions collected across four offshore-facing live studios that accept U.S. traffic. The gap holds even when the hole card is never exposed before settlement, which suggests the effect isn't about information at all — it's about the ritual. The finding lands at an awkward moment for the live dealer sector, which has spent three years pushing "transparent" face-up dealing as a selling point while quietly retaining face-down tables because they convert better.
The data comes from a mid-sized aggregation platform that white-labels live tables for roughly 40 U.S.-facing sportsbooks and casinos. I reviewed the aggregate numbers under a condition that I not name the operator, since the underlying player-level data includes account identifiers. The sample covers February 2023 through March 2025 and includes only blackjack and its close variants — Classic Blackjack, European No-Hole-Card, and six-deck American rules. Roulette, baccarat, and game-show hybrids were excluded. So were tables with under 200 rounds logged, which knocked out a long tail of low-traffic VIP rooms that would have skewed the average stake upward.
The 21% Is Real, and It Is Not About Information
The headline number is straightforward. Median bet at face-down tables: $18.40. Median bet at face-up tables: $15.20. That's a 21.05% difference, which is where the figure in the title comes from. Mean bet tells a similar story — $41.10 versus $33.60, a 22.3% gap — though the mean is dragged around by a small number of players betting $500 or more per hand.
What makes the result interesting is that it survives the obvious controls.
Controlling for what actually differs
Face-down tables in this dataset are not identical to face-up tables except for the card orientation. They differ in three ways that matter:
- Table minimums. Face-down tables carried a median minimum of $10 versus $5 for face-up. Higher minimums mechanically raise average bets.
- Bet limits. Face-down tables topped out at $2,500 median; face-up at $1,000.
- Player mix. Face-down tables drew a higher share of accounts with more than 12 months of tenure — 61% versus 44%.
When the platform's analytics team restricted the comparison to players who had wagered at both table types within the same 30-day window — a within-player design that removes most of the mix problem — the gap narrowed but did not disappear. Those players bet 14.8% more per hand at face-down tables. That's the number I'd treat as the honest one. The 21% figure is the raw headline; the 14.8% is what's left after you strip out the fact that face-down tables attract a different kind of player.
Why the hole card shouldn't matter
Here's the part that should interest anyone who follows blackjack theory. At the tables in this sample, the face-down card is almost never revealed in a way that changes player decisions. In American blackjack, the dealer peeks for blackjack when showing an ace or ten — so the hole card is effectively known to be irrelevant to whether you should hit or stand. In European No-Hole-Card (ENHC) variants, the dealer takes no hole card at all, and players lose additional bets to dealer blackjack. The face-down tables in this dataset are almost entirely American-rules peek games.
So the face-down card is theater. It changes nothing about the math. Yet it moves money.
The Ritual Premium, and Why It's Worth 15% to 21%
The most plausible explanation is what I'll call the ritual premium — the willingness to stake more when a game feels more like the table game it's imitating. Face-down dealing is what you see in a physical casino. It's what you see in movies. It's the version of blackjack that most American players learned to recognize before they ever opened a live dealer lobby.
Face-up dealing, by contrast, reads as a product decision. It's the version a studio built because it was cheaper to automate, easier to stream, or because a regulator asked for it. Players may not articulate that, but the money suggests they feel it.
There's supporting evidence in the session-length data. Median session length at face-down tables: 34 minutes. At face-up: 27 minutes. Players at face-down tables also rebuy more often — 1.9 deposits per session versus 1.4. That's not a pure bet-sizing story; it's an engagement story. Bigger bets and longer sessions compound.
The counter-argument
The skeptical read is that face-down tables simply attract players who are already inclined to bet bigger, and the within-player analysis is contaminated by selection. A player who chooses face-down in a given week might be doing so because they're in a mood to gamble more, not because the card orientation caused it. That's a fair objection, and the platform's data can't fully rule it out. A clean test would require randomizing players into table types, which no operator is going to do.
But the effect shows up in a second cut that's harder to explain away. Among players who were auto-assigned to a table by the lobby's "quick seat" feature — a routing that is effectively random within stake tier — the gap was 11.2%. Smaller, but present. That's the number I'd cite to someone who thinks the whole thing is selection bias.
What This Means for the Face-Up Push
The live dealer industry has spent the last several years marketing face-up dealing as a feature. "See every card." "No hidden information." "Full transparency." That language appears in studio press releases, in operator landing pages, and in the compliance-adjacent copy that regulators like to see. It's a reasonable pitch. It also appears to cost the people making it money.
The math is uncomfortable. If a table's average bet drops 15% when you flip the hole card face-up, and you're running 200 tables across a dozen operators, that's not a rounding error. It's the difference between a studio tier that gets renewed and one that gets consolidated.
Some of this is already visible in the product mix. When I asked three studio-side sources why they hadn't moved more of their blackjack inventory to face-up, all three gave versions of the same answer: players don't ask for it. One put it more bluntly — "the face-up tables are where the bonus abusers and the people grinding out wagering requirements go." That's an uncharitable framing, but it points at something real. Face-up blackjack has become associated with low-margin, high-volume play. Face-down blackjack has become associated with the player who is there to feel like they're at a table.
The regulatory wrinkle
There's a second pressure point. Several U.S. state regulators have shown interest in live dealer transparency requirements — not because face-down dealing is deceptive, but because the hole card creates a class of disputes about when a card was dealt and what the dealer saw. A face-up table is easier to audit. If those requirements spread, studios could be forced into a product configuration that costs them 15% of their blackjack handle.
That's speculative. No state has proposed a face-down ban, and the existing rules mostly govern how the RNG and the card shoe are certified, not how cards are oriented on camera. But the direction of travel in audit-friendly design is toward more visibility, not less.
The Sportsbook Overlap
One thing the data makes clear is that this isn't purely a casino story. Roughly 38% of the accounts in the sample had placed at least one sports wager in the same 30-day window as their live blackjack play. Those cross-over accounts showed a larger face-down premium — 18.4% in the within-player cut, versus 12.1% for casino-only accounts.
The likely mechanism is familiarity. A player who spends most of their time on in-play soccer markets is used to a live product where the state of the game changes in real time and the information is visible. Face-down blackjack is the opposite: it asks you to commit money while withholding a card. That should be a worse fit for a sports-first player. It isn't.
The more plausible read is that cross-over players are simply less price-sensitive on casino product. They're not grinding blackjack for edge; they're playing it as a change of pace between games. Bet size is a function of attention, not of strategy, and a face-down table holds attention longer.
What the studios are actually doing
Quietly, several studios have been adding face-down tables back into their roadmaps. Not as a replacement for face-up inventory, which still serves the compliance narrative and the low-stakes volume, but as a premium tier. The pitch to operators is straightforward: same game, same rules, higher average bet. One studio source described it as "the cheapest yield improvement we have."
That's the honest version of the story. The 21% isn't a mystery. It's a product decision that the industry made, then un-made, then is now re-making because the numbers told it to.
The Responsible Gambling Question
Any finding about bet sizing has a second edge. A 15% to 21% increase in average bet per hand is also a 15% to 21% increase in expected loss per hand, at the same house edge. On a game with a 0.5% house edge and 60 hands per hour, that's the difference between an expected loss of roughly $4.60 per hour at $15.20 a hand and $5.50 per hour at $18.40. Small in absolute terms. Less small over a 34-minute session that turns into a 50-minute session because the table is more engaging.
The operators in this dataset all run standard responsible gambling tooling — deposit limits, session timers, reality checks at 60 minutes. None of them, per the sources I spoke with, adjust the frequency or prominence of those tools based on table type. That seems like an obvious gap. If a product configuration reliably produces larger bets and longer sessions, it's the configuration that should trigger the check-in sooner, not later.
Whether that happens is a business question, not a data question. The data says the face-down table makes more money. The question the industry hasn't answered is whether the extra 15% is worth the extra exposure — and who gets to decide.