True Count Thresholds for Insurance in Blackjack

Most blackjack players have heard the advice: “Insurance is a sucker bet.” And for the average player sitting at a shoe game with no system, that’s completely true. But for card counters, insurance is one of the most powerful tools in the game — if they know exactly when to use it.

The problem? That “when” is not a single number. It shifts based on the counting system in use, the number of decks on the table, the specific cards in a player’s hand, and even the size of the bet relative to the bankroll. Getting this wrong — in either direction — costs money. This guide breaks down precisely how True Count thresholds work for insurance across every major counting system, and what separates amateur application from professional-level strategy.

Table of Contents

• What Is the Insurance Bet in Blackjack?
• The Math Behind the Insurance Break-Even Point
• True Count Thresholds by Counting System
• How Deck Count Shifts the Hi-Lo Insurance Index
• Composition-Dependent Adjustments: Why Your Hand Changes the Number
• Risk-Averse Insurance: Protecting Bankroll Over Maximising EV
• The Specialised Insurance Count
• Practical Takeaways for Counters
• Frequently Asked Questions
• Conclusion

What Is the Insurance Bet in Blackjack?

When the dealer shows an Ace, players are offered insurance — a side bet that the dealer’s hole card is a ten-value, completing a blackjack. The bet pays 2:1 and costs half the original wager.

For untrained players, the bet carries a house edge of roughly 7.7% in a six-deck game. It’s a bad bet the vast majority of the time. But when the remaining deck is rich enough in ten-valued cards, that equation flips. The challenge is knowing with precision exactly when that flip happens. Players who haven’t yet built a counting system would be better served starting with basic blackjack strategy before worrying about insurance indices.

The Math Behind the Insurance Break-Even Point

Insurance pays 2:1. That means a player needs the dealer to hold a ten more than one time in three to break even — specifically, when tens represent more than 33.3% of the remaining cards.

In a freshly shuffled deck, tens make up just 30.8% of all cards (16 out of 52). The deck needs to become meaningfully ten-rich before insurance crosses into positive expectation. That ten-density threshold is what counting systems measure — they just do it in different ways, using different tags and different index numbers.

True Count Thresholds by Counting System

Different counting systems assign different point values to cards, which means their internal “break-even” signals for insurance land at different numbers. Here is how the major systems compare:

Hi-Lo System

The most widely used counting system sets its standard insurance threshold at a True Count of +3 or higher. At this level, the remaining deck holds approximately 35% ten-valued cards — just enough to make the 2:1 payout a positive-expectation bet. The +3 figure is the one most beginner and intermediate counters learn first, though as this guide explains, it is rarely the complete picture.

KO (Knock-Out) System

KO is an unbalanced system, meaning it does not require conversion to a True Count. Instead, insurance becomes correct when the Running Count reaches +2 per deck remaining or higher. The simplicity makes it accessible, but it sacrifices some of the precision available to Hi-Lo practitioners.

Omega II System

This multi-level system is more granular than Hi-Lo and locates the insurance threshold at a True Count of +2 or higher. Its more precise card-weighting means it reaches the insurance signal slightly earlier in deck-rich situations, giving skilled players a modest edge in identifying profitable spots.

Unbalanced Ten Count (UTC)

The UTC uses a pivot point — a Running Count of 0 — as the exact break-even threshold for insurance. When the Running Count sits at or above 0, the bet has neutral or positive expectation. This system is most used by specialists focusing exclusively on the insurance decision rather than overall game strategy.

Summary table of thresholds across systems:

How Deck Count Shifts the Hi-Lo Insurance Index

The +3 figure taught to most Hi-Lo students is a simplified approximation. In reality, the exact break-even index moves as the number of decks in play changes — and the direction of that shift is important.

The reason comes down to the card-removal effect. In a single-deck game, every card seen has an outsized impact on the remaining composition. Remove an Ace from a 52-card deck and the remaining 51 cards look very different. Remove it from a 416-card eight-deck shoe and the shift is barely measurable. This dilution means that fewer decks require a lower True Count to reach the same ten-density threshold.

The practical takeaway: a counter playing single-deck should take insurance far more often than a shoe-game player, simply because the break-even bar is much lower. Waiting for a +3 count in a single-deck game means passing up profitable bets.

Composition-Dependent Adjustments: Why Your Hand Changes the Number

Here is where intermediate counters often leave money on the table. The True Count reflects the overall density of tens in the remaining shoe — but it does not automatically account for the cards sitting in front of the player. Those matter enormously, especially in single-deck games.

The Hard 20 (10-10) Problem

A player holding two tens has already removed two of the very cards needed for a profitable insurance bet. In a single-deck game, that is two out of 16 ten-valued cards gone before the dealer’s hole card is considered. The impact is significant: the Hi-Lo insurance index for a player holding a pair of tens rises from the generic +1.4 all the way to +3.0 in single-deck play.

In a six-deck shoe, the same two tens shift the index by only +0.3 (from 3.0 to 3.3), because 150-plus other cards dilute the effect of those two removed cards.

The Blackjack (Even Money) Decision

Taking “even money” on a blackjack is mathematically identical to taking insurance when holding a natural. The player already holds one ten and one Ace — two critical cards removed from the pool. This dramatically lowers the required threshold. In a six-deck game, the True Count needed for even money to be profitable drops to approximately +2.0. In single-deck play, the threshold falls to near zero (between 0.0 and +0.1), meaning it is almost always correct to take even money in a single-deck game when counting.

Risk-Averse Insurance: Protecting Bankroll Over Maximising EV

Standard card-counting strategy targets maximum Expected Value (EV). Risk-averse (RA) strategy takes a different approach — it targets the Certainty Equivalent (CE), which factors in bankroll volatility alongside raw expectation. For serious players betting meaningful portions of their Kelly bankroll, this distinction changes real decisions.

Why Strong Hands Are More Insurable

A player holding a 19 or 20 has a high conditional expected value — they are likely to win the main hand if the dealer does not have blackjack. Insuring these hands at a slightly negative EV still makes sense because it reduces the risk of losing a large wager to a dealer blackjack. The variance reduction justifies the slight expectation cost.

For a Hard 20 specifically, the theoretically optimal insurance amount to minimise variance is 111% of the bet — more than the full insurance the casino allows. In practice, players cap out at 50% coverage, but the mathematics indicate they should be ensuring these hands eagerly, even at counts somewhat below the standard index.

Why Weak Hands Are Less Insurable

A stiff hand like a 16 has low conditional expectation — it is likely to lose even when the dealer does not have blackjack. Insuring it provides little meaningful protection. Worse, taking full insurance on a stiff hand can actually increase overall bankroll variance. The risk-averse index for weak hands pushes higher than the standard break-even index as a result.

The Bet Size Effect

Risk-averse thinking becomes proportionally more relevant as bet size increases relative to bankroll. A player wagering 1-2% of their Kelly-equivalent bankroll can largely ignore RA adjustments for insurance. A player wagering 4-5% — common in multi-deck shoe games with wide bet spreads — should account for RA shifts, particularly on strong hands.

Practical Takeaways for Counters

• The +3 Hi-Lo rule is a starting point, not the full answer. It applies most cleanly to six-deck shoe games with standard hand compositions.
• Single-deck players should use the +1.4 generic index and apply composition adjustments for any hand containing tens.
• Always adjust for your hand. Holding a 20 dramatically raises the required count; holding a blackjack dramatically lowers it — in some cases to near zero.
• In shoe games with large bets, risk-averse thinking pays. Insuring strong hands slightly early protects bankroll growth even at a minor EV cost.
• Avoid over-insuring weak hands. Stiff totals like 14, 15, and 16 deserve higher index thresholds than basic strategy charts suggest.
• Specialised insurance counts offer precision but add cognitive load — only worthwhile for players whose primary edge comes from the insurance bet at high-count tables.
• Before applying any of the above at a real table, make sure you’re playing at a  best and worst blackjack rules — the game rules, deck counts, and penetration depth vary significantly between operators and directly affect every index on this page.
• For deeper mathematical analysis of blackjack strategy and card counting, see Wizard of Odds

Frequently Asked Questions

What is the standard True Count threshold for insurance in Hi-Lo? The most commonly cited threshold is a True Count of +3 or higher. At that point, roughly 35% of remaining cards are tens, which makes the 2:1 insurance payout mathematically favourable. However, this is a generalised figure for six-deck games — single-deck games break even at a True Count as low as +1.4.

Does the number of decks change when I should take insurance? Yes, significantly. The break-even True Count index rises as more decks are added. In a single-deck game it sits around +1.4; in a double-deck game, +2.4; in a six-deck shoe, +3.0. The more decks in play, the richer in tens the remaining shoe must be before insurance becomes a profitable side bet.

Why does holding a pair of tens change the insurance index? Because those two tens are no longer available as the dealer’s hole card. In a single-deck game, removing two tens from a pool of 16 noticeably reduces the probability that the dealer holds a ten. That raises the required True Count for insurance from around +1.4 to approximately +3.0 — a major shift that many intermediate counters overlook.

What is risk-averse insurance and when does it apply? Risk-averse insurance means taking the bet at a slightly lower count than the mathematical break-even point in order to reduce bankroll variance. It is most relevant for players with strong hands (19, 20, blackjack) and large bets relative to their bankroll. The logic is that protecting a high-value hand from a dealer blackjack is worth a minor negative-expectation cost.

Can the KO counting system be used for insurance decisions? Yes. Because KO is unbalanced and does not use a converted True Count, insurance is triggered by the Running Count instead. Players should take insurance when the Running Count reaches +2 per remaining deck or higher. The system is simpler to apply than Hi-Lo index adjustments but trades some precision for accessibility.

Is a specialised insurance count worth learning? For most counters, no. The improvement in insurance accuracy over a well-applied, composition-adjusted Hi-Lo index is marginal, while the mental overhead of running a second simultaneous count is significant. The dedicated count is most valuable for players at high-limit tables where insurance bets represent a meaningful portion of their expected edge.

When is taking even money correct for a card counter? Taking even money on a blackjack is mathematically identical to taking insurance. In a six-deck game, the break-even True Count for even money is approximately +2.0 — lower than the generic insurance index because the player has already removed one ten and one Ace from the remaining pool. In single-deck play, the threshold approaches zero, making even money almost always correct when counting.

Conclusion

Insurance is not a sucker bet for card counters — but only if the thresholds are applied correctly. The gap between a +1.4 single-deck index and a +3.0 six-deck index is not a rounding error; it represents real money left on the table or lost unnecessarily.

The key variables to internalise are: the counting system in use, the number of decks in play, the specific cards held, and the size of the bet relative to the bankroll. Master those four factors and insurance transforms from a casino marketing trap into one of the most precisely exploitable decisions in the game.

Professional blackjack is built on edges measured in fractions of a percent. Correctly calibrated insurance decisions — taken at the right count, adjusted for the right hand, and weighted for the right bet size — are one of the clearest places where that edge compounds over time.