Infinite, Lightning, Classic, VIP: Which Live-Dealer Blackjack Tables You Can Actually Count

The live-dealer lobby is not one game. It is a wall of tables that all look like blackjack, and most of them were designed so that a count never becomes useful.

That is the part almost every “can you count cards online?” article skips. They treat live dealer as a single category, then conclude it does not work. We have already covered why RNG blackjack is structurally uncountable, and why a public calculator for a real shoe is legal. This post is the missing filter: which live tables still have a depleting shoe, which are a waste of a count, and how to tell the difference from the stream in about five minutes.

Do not trust the lobby name. Watch the shoe.

The lobby is not one game

Open any large live-casino lobby and you will see some mix of Classic, Speed, VIP, Salon, Infinite, Lightning, First Person, and a rotating cast of branded variants. They share a dealer, a camera, and a felt. They do not share a dealing system.

Three questions decide whether a table is even a candidate:

1. Are the cards physical, and do they stay out of play after they are dealt?
2. How deep into the shoe does the dealer go before shuffling?
3. Is the main bet still standard blackjack, or has an RNG overlay or a rule variant taken over the expected value?

If the answer to (1) is no, stop. No calculator, no counting system, and no amount of patience creates information that is not there. If (1) is yes, (2) and (3) tell you whether sitting is worth the time.

Crypto casinos do not get a special exemption. A Bitcoin cashier does not create a depleting shoe. The same tells apply on a crypto lobby as on a licensed one.

The three dealing systems

Forget product names for a moment. Every live blackjack table you will actually sit at falls into one of three mechanical setups. You can see all three on camera.

1. Reshuffle every hand, or a continuous shuffler

Discards go back into a machine, or the dealer completes a full shuffle before the next round. Every hand starts from a fresh composition. The running count resets to meaninglessness.

What it looks like: used cards are fed into a shuffler after the round, or you watch a complete shuffle between every hand. There is no cut card sitting in a shoe that slowly empties. The discard tray, if it exists, never fills.

Verdict: skip. This is RNG blackjack with a human attached. The math is the same as shuffling after every hand.

2. Mid-shoe cut card (the typical studio shoe)

A real 6- or 8-deck shoe. A plastic cut card is placed somewhere in the middle. The dealer deals until that card comes out, then shuffles. Discards stay in the tray until then.

This is the game the rest of this post is about. Deck depletion is real. The constraint is how early they cut.

What it looks like: a clear or smoked shoe on the table, a cut card, a discard tray that fills as the shoe empties, and a shuffle only when the cut card appears — not after every hand.

Verdict: maybe. Sit long enough to measure the cut, then decide. Most studio tables cut around half the shoe. That is countable in principle and modest in practice, which I will quantify below.

3. A deeper shoe

Same hardware as (2), but the cut card is closer to the back of the shoe. You see well past the halfway point before the shuffle.

What it looks like: identical to a mid-shoe table until you watch one full shoe. The only tell is where the cut card sits, or how many decks have come out when they shuffle.

Verdict: sit, if the rules are also standard (3:2, not a novelty variant). These are uncommon in high-volume studios and more common on some VIP or slower tables. Confirm it. Do not assume VIP means deep penetration.

Named products: sit, skip, or maybe

Studio catalogs change, and the same title can be configured differently across operators. Use this as a starting map, then verify with the tells above.

A 7-seat Classic table on a major provider is the place most people should start looking, not because it is generous, but because it is the one format that still uses a real shoe often enough to be worth checking.

Infinite is the trap. It looks like the most “live” product in the lobby — a dealer, a crowd, a physical deal — and it is the one recreational players open first because they never have to wait for a seat. If that table reshuffles every round, counting it is how you convince yourself the whole idea is fake.

How to measure penetration from the stream

Penetration is the percentage of the shoe dealt before the shuffle. It is the single number that decides whether a countable table is worth sitting.

You do not need a formula to start. You need one full shoe of observation, without betting if you do not want to.

Watch for the cut card. On an 8-deck shoe it is a plastic card inserted into the stack. When it comes out, that round is usually the last of the shoe. Estimate how much of the shoe is still in the box when it appears. Halfway is 50%. A thin remaining slab is deeper. A fat remaining slab is worse.

Count decks the slow way if you are unsure. An 8-deck shoe is 416 cards. A full 7-seat round with a dealer typically puts 15–20 cards on the felt. Time one shoe: if they shuffle after roughly 10–14 rounds, you are in the 50% neighborhood. If they shuffle after 6–8 rounds, walk. If they go 18–20+, that is the uncommon deep shoe.

What 50% actually does to the count. True count is running count divided by decks remaining. Take a running count of +8, which is a noticeable excess of low cards:

A true count of +1 is barely above break-even against a typical 8-deck house edge. +2 is where a modest player edge starts to appear. +4 is the kind of situation physical-casino counting is built around.

At a 50% cut, the same running count never gets to concentrate the way it does at 75%. You will still see plus counts. You will see +3 and +4 much less often, and for fewer hands when you do. That is the honest version of “studios shuffle early.” It does not make the remaining cards imaginary. It makes strongly favorable counts rare.

If you expected a 1% hourly edge from a live-dealer phone session, 50% penetration on eight decks will disappoint you. If you were going to sit the table anyway and would rather play the shoe that is actually in front of you, the count is still information.

Rules that still matter once the shoe is real

Table selection beats a fancier counting system online. Hi-Lo is already a high-betting-correlation count. Switching to Omega II will not rescue a 6:5 table or a Lightning overlay.

Once you have a depleting shoe, check these in the info panel and on the felt:

Blackjack pays 3:2, not 6:5. 6:5 adds roughly 1.4% to the house edge. Counting cannot reasonably overcome that on a 50% shoe. If the payout is 6:5, skip.

Dealer hits or stands on soft 17. S17 is better for the player than H17 by about 0.2%. Many studio Classic tables stand on all 17s. Take it when you can; do not sit an otherwise bad table just for S17.

Double after split. DAS is player-friendly. A lot of high-volume 7-seat tables do not offer it. That raises the baseline house edge. It does not make the shoe uncountable; it means you need a slightly higher true count before the player is actually ahead.

Number of decks. Six is better than eight for the same penetration. Eight is what you will usually get. Set the calculator to match, do not leave it on the wrong deck count.

Surrender. Nice when it exists. Rare on these streams. Do not hunt for it.

Side bets. Perfect Pairs, 21+3, Hot 3, Bust It — ignore them. Their house edges are several percent. Counting the main game does not make a 5% side bet into a good bet.

A countable live-dealer table is usually: 7-seat Classic or VIP, 8 decks, 3:2, a real cut card, discards staying out, and a cut you have measured. Everything else in the lobby is noise.

Sit-or-skip checklist

Run this before you open a calculator, and before you put money on the table.

1. Human dealer, live camera. Not First Person, not a video slot wearing a blackjack skin.
2. Physical shoe. Cards come out of a box on the table, not a hidden generator.
3. Discards stay out. Nothing is fed back into a shuffler between hands. No full shuffle after every round.
4. You can see every seated hand. You need every exposed card, including other players. If the UI hides them, you cannot count the table.
5. You have watched one shuffle. You know roughly where they cut. Under ~40% penetration, leave. Around 50%, sit only if you accept a small, infrequent edge. Deeper is the table you were looking for.
6. Blackjack pays 3:2.
7. No RNG multiplier on the main bet. Lightning and its cousins are a different product.
8. Standard ranks and payoffs. Not stripped tens, not dealer-22-push as the main attraction, not a novelty rule set you have not modeled.

If it fails any of 1–4, 6, or 7, skip. If it fails 5 or 8, skip unless you know exactly why you are making an exception.

If it passes, the shoe is real. That is the game the calculator is built for. The next problem is operational — two windows, dealer speed, which cards to tap — and that is the follow-up walkthrough.

The point of the filter

Most live-dealer blackjack is not a puzzle you failed to solve. It is a product line that removed deck depletion on purpose, then kept a camera so it would still feel like a casino.

The mistake is treating that as a verdict on counting, or on tools that track a shoe. Count Infinite, and you will correctly conclude the count never moved. Count a Classic table with a cut card, and you will see it move — not as often as a 75% pit game, and not enough to get rich on a weeknight, but in the same direction the math predicts.

Five minutes of watching the shoe is cheaper than an hour of tapping cards into a reshuffle. Use the checklist, then decide.

Found a table that passes? Count that shoe in real time, or open the calculator and tap along with the stream. For the underlying RNG vs live-dealer split, start with why counting fails on most online blackjack.