Appendix

Below are the principles behind the choices, with supporting stats and how they map into the rules.

These rules were developed out of the official Disney’s Galaxy’s Edge rules, the ones packaged with the deck of cards sold at the Disney Parks, and they aim to stay as close to that version as possible. The goal here was not to reinvent the game, but to preserve its character while only changing what was truly necessary to make it fully consistent, strategically well aligned, and suitable for clear casino-style play.

Download the full calculations document

Ranking by rarity. In poker, the rarer hand ranks higher; Sabacc should follow the same logic. My simulations and combinatorics show that when we look across families of ranked hands (e.g., all Rhylet variants together, all Straights together), rarer families are placed higher than more common ones. This keeps incentives and expectations aligned: hard things are rewarded accordingly.

Higher key beats lower key inside the same rank. Some argue “lower integer” should win because “the goal is to be closer to zero anyway.” But that overlooks risk. Example: drawing from [+2,+2,-2] vs [+9,+9,-9] to try to get a Squadron: It’s not that the probability to get a Squadron with 2s is higher than with 9s. These are the same. It’s that misses with the high-key chase push you much farther from zero on average, and misses are much more common. If you do land the higher key, you have succeeded despite greater expected downside along the way. Ranking the higher key higher correctly rewards that risk.

It keeps one compact tiebreaker chain for all outcomes: Nulrhek and Sabacc, ranked and unranked. If multiple at zero, ranked-hand beats unranked, higher rank wins, then higher key, then general tiebreakers; for equal-distance Nulrhek, positive beats negative, then general tiebreakers.

1. Most cards wins
2. Highest Σ|cards| (sum of absolute values; look at all numbers, ignore colors)
3. Highest |card| (compare highest to lowest; look at all numbers, ignore colors)
4. Highest positive card
5. Suited (last numeric resort; Sylop doesn’t break suited)
6. Single-card draw (only if still tied)

This mirrors the production notes of the film rules in Solo and avoids per-rank “if/then/else/unless/otherwise” branches. It’s easy to learn, fast to adjudicate, and watertight at a casino table.

Why highest Σ|cards|, and not only highest sum of all positives? Using Σ|cards| matches the film’s “numbers first” spirit and avoids ignoring half the hand too early.

Why highest |card|? It resolves hands that are clearly not the same hand by values, preventing ties and a single card draw if possible.

Why suits and why at the end? Suits are not the focus of any realistic strategy. It is only introduced as a last tiebreaker to shave off some single card draws if possible.

Ranked hands: families, special cases, and clear placement. I keep all the familiar named hands (Rhylet, Yee-Haa, etc.) but recognize families where the original Galaxy’s Edge list feels incomplete or too constrained. Example: Wild Rhylet generalizes Full House patterns beyond matching signs; Full Straight fills an obvious sequence gap; Five Card Squad distinguishes rarer four-of-a-kind patterns at 5 cards.

Pure Sabacc (0,0) is not top rank. It’s elegant, but comparatively common in long runs, so it ranks below much rarer patterns, consistent with the rarity-first philosophy.

Spike Dice: less whiplash, more play. Non-Spike doubles are treated as a forced single discard+draw instead of a wipeout. That keeps the Spike Sabacc flavor, but creates more opportunities to improve hands and reduces resets. Only double Spikes trigger a wipeout.

Betting belongs after Spike Dice. For real-money play, your last information update should come before betting. That way, your wager reflects the hand you’ll actually reveal, with no 1-in-6 “my hand just changed” lottery between bet and showdown.

Here are a few datapoints from long runs (multi-million hand simulations, 4 players), plus a closed-form count across five sequential draws. These support the ranking-by-rarity stance and show, for example, why Pure Sabacc shouldn’t sit at the top, and Yee-Haa is also totally overrated.

If you want to explore the simulator itself, including the code and the generated hand-combination data, have a look at the Corellian Spike Simulator repository.

Galaxy’s Edge rules, 8 million games per targeted ranked hand, 4 players

This (Casino) version, 8 million games per target hand, 4 players

• Very rare families: Full Sabacc and Fleet, Rhylet and Wild Rhylet, Gee Whiz and Full Straight, Sylop Straight Khyron.
• Rarer mid-tier: 4 of a kind , Sylop Rule of Two , Banthas Wild , Pure Sabacc.
• More common families: Straight Khyron, 2 pairs, Yee-Haa, Pair.

Algebraic draw counts, all 5 card draw Sabacc matches, 6,471,002 total possible combinations

This method is how probabilities are usually given for Poker style games. (Contrary to my earlier version of this analysis, after closer examination and 136 million more simulations with targeted runs, it turns out The Yee-Haas are not rarer than Rule of Two, and still by far more common than for example Banthas Wild.) So, the combinatorics and real play match in every case. This is the only correction I had to make to the ranked hands after deeper analysis (And could have been prevented, if I would have believed the math in the first place).

Takeaway: families higher in the list are notably rarer; Pure Sabacc, while iconic, is far more frequent than many upper-tier families, so it shouldn’t outrank them. As Kay Vess would say “Remember, this isn’t Kessel Sabacc now!”

(For comparison, the Galaxy’s Edge rules simulation over 88 million games shows Pure Sabacc is more frequent than Fleet, Full Sabacc, Rhylet, Gee Whiz, Squadron, and Banthas Wild combined! again reinforcing that “rarity → rank” keeps expectations honest.)

(Note: The big frequency sims shown here evaluate card formation, not behavioural betting edges, on purpose, so you can compare base combinatorics)

Open all downloads