Hi-Opt II is a multi-level (level-2), balanced, ace-neutral count designed for the highest possible accuracy on insurance and play decisions. Powerful, demanding, and best for serious counters who can run a side count cleanly at table speed.
Memorize the table below. Note that two card groups are worth +2 or −2 instead of +1 or −1.
| Card | Tag Value |
|---|---|
| 2, 3 | +1 |
| 4, 5 | +2 |
| 6, 7 | +1 |
| 8, 9 | 0 |
| 10, J, Q, K | −2 |
| A | 0 (track separately if side counting) |
The system is balanced: a full deck sums to (4×1 + 4×2 + 4×2 + 4×1 + 4×1 + 4×0 + 4×0 + 16×(−2) + 4×0) = 0.
The point of weighting 4s, 5s, and 10s at ±2 is that those cards have the largest impact on dealer outcomes. A 10 hurts the dealer most when she’s showing a stiff card; a 5 helps the dealer most. Tagging them at twice the weight of 2s, 3s, 6s, and 7s captures more information per card and produces a count that is much better correlated with insurance decisions and most index plays.
The cost is mental load: every time you see a 4, 5, or a 10, you have to add or subtract 2 instead of 1. This sounds trivial in isolation but adds up over a six-deck shoe at table speed. Many counters struggle to run Hi-Opt II reliably for several hours.
| Metric | Hi-Lo | Hi-Opt I | Hi-Opt II (with ace SC) |
|---|---|---|---|
| Playing Efficiency | ~0.51 | ~0.61 | ~0.67 |
| Betting Correlation | ~0.97 | ~0.97 (with ace SC) | ~0.99 |
| Insurance Correlation | ~0.76 | ~0.85 | ~0.91 |
| Memorization | One rule | Two rules (2 and A separate) | Five rules (level-2 weighting) |
| Mental load | Low | Medium | High |
Hi-Opt II beats both simpler systems on every accuracy metric, but only if you can run it without errors. A 1% miscount rate on Hi-Opt II is worse than a 0% miscount rate on Hi-Lo. Self-honesty matters here.
With aces tagged 0, the main running count tells you nothing about ace density. For betting, ace density matters a lot — aces drive blackjack frequency, which in turn drives a meaningful chunk of your edge at high counts.
Without a side count, Hi-Opt II’s betting correlation drops to roughly 0.91 (similar to a basic level-1 count). With the side count, it climbs to ~0.99. The difference is real money on a serious bet spread.
Setting it up:
Insurance correlation is the metric that measures how well a counting system predicts when taking insurance is profitable. Hi-Opt II’s ~0.91 correlation is among the best of any practical system. Hi-Lo’s ~0.76 means roughly one in four insurance decisions made on the count alone is sub-optimal.
For high-frequency play, especially at games with many dealer aces showing, Hi-Opt II catches more of the correct insurance moments. This is the single strongest argument for choosing Hi-Opt II over Hi-Lo.
If insurance accuracy is not a priority for you (you bet small, the insurance side bet is unavailable, etc.), Hi-Opt II’s extra mental load is hard to justify.
The Blackjack 3000 trainer supports Hi-Opt II in Settings. Speed count, true count, and gauntlet drills will all score against Hi-Opt II tag values when selected.
Be honest about the trade-offs. Hi-Opt II is the strongest practical balanced count, but it asks for serious commitment.
Use Hi-Opt II if:
Stick with Hi-Lo or Hi-Opt I if:
Hi-Opt II has the highest insurance correlation of any common counting system and very high playing efficiency. It is worth the extra effort only if you can run it and the ace side count without errors at table speed. Most players are better served sticking with Hi-Lo or Hi-Opt I.
Level-2 means the tag values include numbers other than +1, 0, and −1. Hi-Opt II uses +2 for 4s and 5s, and −2 for 10s. The wider range of values creates a more sensitive count but increases the mental workload because you must add or subtract 2 instead of just 1 for many cards.
Strongly recommended. Hi-Opt II tags aces as 0, so without an ace side count the betting correlation drops to about 0.91 versus ~0.99 with the side count. Skipping the side count negates much of Hi-Opt II’s edge over simpler systems.
The same way as Hi-Lo — divide the running count by decks remaining. Because Hi-Opt II tags some cards at ±2, the running count swings wider, so you will see larger numbers. Round to the nearest integer for index decisions.
No. Hi-Opt II has its own index numbers, which are roughly 1.5x to 2x the magnitude of Hi-Lo’s for the same plays. Always use a Hi-Opt II-specific index chart from a verified source.
Most serious Hi-Opt II practitioners run the side count constantly. Some players drop it during periods of low concentration or when not at the table for serious betting decisions, but the math says it pays.