Cracking the Code: The Likely Drop Rates Behind Rocket League Tournament Cups

35,000 drops told us what the cups give out. Here is my best guess at the numbers behind them, and what changes if I'm right.

In my last big analysis, I logged 35,296 tournament cup drops from more than 850 YouTube videos and measured exactly what each cup gives out. That answered what happens when you open a cup.

It didn't answer the question that kept bugging me afterward: What numbers did the developers actually type in?

Nobody picks a drop rate of 38.40%. Somewhere in Rocket League there is a reward table, and someone filled it in. My measured rates are what that table produced across thousands of openings, plus some noise. So in this final post of the series, I'm working backward from the data to the most likely settings behind every cup. Then I'll see whether those settings change which cup is best to buy.

Before we start: Everything in this post is an inference, not official odds. Psyonix hasn't published them. The measured rates in my main analysis are still the real data. This post is my best explanation for where they came from.


Working Backward Like a Designer

A good guess has to pass three tests. It has to look like something a person would choose, it has to make sense as game design, and it has to fit the data.

Test 1: Would a person actually pick these numbers?

When people set up probabilities, they reach for familiar numbers:

  • Whole percentages like 25% or 30%
  • Simple fractions like 1/2, 1/4, and 1/8, and combinations of them
  • Small whole-number weights that the game turns into percentages
  • A few reusable tables instead of many nearly identical ones

Weights are more useful than they might sound. A designer doesn't have to write "37.5%." They can give Very Rare a weight of 12, Import 9, Exotic 8, and Black Market 3. The game adds those up (32) and divides. Those four simple numbers become 37.5%, 28.125%, 25%, and 9.375%. A strange-looking percentage can come from a very ordinary choice.

Test 2: Does it make sense as game design?

The All-Star and Chromatic Cups do the same job: they're the high-end rewards. Look at how close their results are:

RarityAll-StarChromatic
Exotic25.26%25.22%
Black Market9.09%8.59%

The simplest way to build two cups like this is to make one rarity table, use it for both, and then add a rule to the Chromatic Cup that makes the item painted. That's easier to build and easier to balance than keeping two tables that are almost the same.

Test 3: Does it fit the data?

Every guess has to land inside the 95% margins of error from the real drops. I also ran a chi-square test, which checks whether the gap between a guess and the data is bigger than random chance would explain.

There is a catch: those tests assume every drop is independent, and these drops aren't perfectly independent. The real uncertainty is probably wider than the margins of error suggest because of:

  • Many drops coming from the same video
  • Videos clustered by player, season, or reward pool
  • Players choosing which openings to record and upload
  • Differences in video quality
  • Detection or classification errors in my script
  • Drops that couldn't be identified and were filtered out
  • Possible changes to the game across seasons

So a guess has to fit the data, but that isn't the only thing that decides the ranking.


The High-Tier Cups: All-Star and Chromatic

Here are the four explanations I considered, ranked from most to least likely.

#1: One shared table with weights of 12:9:8:3 (my top pick)

RarityWeightTheorized RateAll-Star (Measured)Chromatic (Measured)Combined (Measured)
Very Rare1237.5%38.40%35.83%37.24%
Import928.125%27.24%30.36%28.65%
Exotic825%25.26%25.22%25.24%
Black Market39.375%9.09%8.59%8.87%

In plain terms: out of every 32 high-tier cups, you'd expect 12 Very Rares, 9 Imports, 8 Exotics, and 3 Black Markets. Both cups would use this table, and the Chromatic Cup would then paint whatever you got.

Why it's my top pick:

  • The weights are small whole numbers, and the rates are clean fractions.
  • When I combine all 7,907 All-Star and Chromatic drops, all four rates land inside their margins of error.
  • It explains why the Exotic and Black Market rates are almost identical between the two cups.
  • It follows a common pattern in game design: reuse a table, then add a rule on top.

The evidence against it: If you treat every drop as independent, the All-Star and Chromatic results are different enough that random chance would rarely produce the gap (p ≈ 0.014). On its own, the Chromatic sample's 35.83% Very Rare and 30.36% Import both fall outside their margins of error for this model. Most of that difference is Very Rare and Import trading a few percentage points with each other. That's a real strike against a shared table. It isn't a knockout, though, because drops from the same videos, players, and seasons aren't truly independent, and accounting for that would widen the margins.

#2: One shared table fitted to the data (30:23:20:7)

RarityTheorized Rate
Very Rare37.5%
Import28.75%
Exotic25%
Black Market8.75%

These are weights of 30:23:20:7 out of 80. This version fits the combined data almost perfectly, and every value is a multiple of 1.25%.

I rank it second because it fits a little too well. 30:23:20:7 looks like numbers fitted to my data after the fact, not numbers someone sat down and chose. 12:9:8:3 is the more believable human choice, even though it doesn't fit quite as perfectly.

#3: Two separate, rounded tables

CupVery RareImportExoticBlack Market
All-Star37.5%27.5%25%10%
Chromatic35%30%25%10%

Both rows use friendly numbers and add up to 100%. The Chromatic table could be set up on purpose to move some odds from Very Rare to Import.

It ranks lower because it needs two tables where one would do. Also, a 10% Black Market rate for the Chromatic Cup is quite a bit higher than the 8.59% I measured.

#4: The measured rates are the real, separate settings

This one says the two cups really do have different tables, and my measured numbers are close to them. The data does lean a little in this direction, since the two samples differ more than chance would usually explain.

But numbers like 38.40% and 30.36% don't look like anything a person would type in, and sampling, clustering, mixed seasons, and detection errors could all create the gap. It's possible, but it's the most complicated explanation.


The Lower-Tier Cups: Prospect and Challenger

Prospect Cup

RarityMeasuredTheorized
Uncommon62.19%62.5%
Rare19.22%18.75%
Very Rare12.41%12.5%
Import4.44%4.5%
Exotic1.44%1.5%
Black Market0.30%0.25%

These rates add up to exactly 100%. Each one falls inside its margin of error, and together they work out to weights of 250:75:50:18:6:1 out of 400. That last weight is the fun one: it means 1 Black Market in every 400 Prospect Cups.

My first guess was a rounder 60/20/12.5/5/1.5/0.25, but that only adds up to 99.25%. Even after fixing the total, 60% Uncommon is well below the measured range of 61.43% to 62.95%. 62.5% is just as friendly a number (it's 5/8), and it fits the data far better.

Challenger Cup

The Challenger Cup is the hardest one to crack. Here's the best guess that fits the data:

RarityMeasuredTheorized
Uncommon27.82%27.5%
Rare26.08%26.25%
Very Rare23.34%23.75%
Import15.88%16.25%
Exotic5.64%5%
Black Market1.24%1.25%

That's weights of 22:21:19:13:4:1 out of 80, which adds up to 100%. But the 5% Exotic rate sits just below its measured range, so I hold this one loosely.

A simpler-looking option is 27.5/25/25/15/6.25/1.25, but it fits the data noticeably worse. I'd need more Challenger drops before picking one.


What Changes If I'm Right

Here's the full picture under my leading explanation:

  1. Prospect and Challenger each have their own table.
  2. All-Star and Chromatic share one high-tier table, probably close to 12:9:8:3.
  3. The Chromatic Cup paints whatever that table gives you.

Painted and certified rates don't change. This is only about rarity. Here's what the theorized rates do to the cups:

Without trade-ups: how many cups per Black Market?

CupMeasuredTheorized
Prospect1 in 333 (~400,000 cr)1 in 400 (480,000 cr)
Challenger1 in 81 (~323,000 cr)1 in 80 (320,000 cr)
All-Star1 in 11 (~132,000 cr)1 in 10.7 (128,000 cr)
Chromatic1 in 11.6 (~233,000 cr)1 in 10.7 (~213,000 cr)

The Prospect Cup gets noticeably worse for Black Market hunting, since 1 in 400 means 80,000 more credits on average. Under a shared table, the All-Star and Chromatic Cups have exactly the same Black Market odds.

With trade-ups: effective cost per item

This uses the same method as my main analysis. You trade up every drop below your target rarity at 5 items for 1.

Target RarityProspectChallengerAll-StarChromatic
Very Rare6,403 → 6,400 cr13,483 → 13,289 cr31,247 → 32,000 cr55,815 → 53,333 cr
Import14,655 → 14,545 cr18,339 → 17,961 cr34,360 → 33,684 cr53,300 → 56,140 cr
Exotic39,048 → 38,095 cr39,978 → 42,310 cr37,210 → 37,354 cr61,120 → 62,257 cr
Black Market130,505 → 136,364 cr123,285 → 127,356 cr77,227 → 75,949 cr132,122 → 126,582 cr

Measured → theorized. The Challenger numbers depend on the lower-confidence Challenger table.

Takeaway: The winners don't change. Prospect is still the cheapest way to get Very Rares and Imports, and All-Star is still the cheapest way to get Exotics and Black Markets. What changes is the middle of the pack. With trade-ups, the Chromatic Cup goes from the most expensive way to get a Black Market to roughly tying the Challenger Cup for second.


The Final Verdict

If the shared-table explanation is right, the cup decision gets simpler than ever:

  1. The Chromatic Cup is an All-Star Cup with paint. You're paying 8,000 extra credits for the guaranteed paint, not for better rarity odds. (One catch: its items are less likely to be certified, at 10.25% versus the All-Star Cup's 18.13%.) If you want painted items, it's still the only cup that guarantees them. If you don't care about paint, buy the All-Star Cup.
  2. The All-Star Cup is still the king for Exotics and Black Markets. About 3 of every 32 cups should be a Black Market.
  3. The Prospect Cup is still the volume play for Very Rares, Imports, and trade-up fodder. Just don't count on it for Black Markets: at 1 in 400, it's a long grind.
  4. The Challenger Cup is still the awkward middle child. Its true settings are the hardest to pin down, and nothing here makes it a better buy.

None of this is confirmed. If Psyonix ever publishes official odds, or someone collects enough season-by-season data to test these guesses properly, I'd love to see how close I got. Until then, 12:9:8:3 is my bet.

This wraps up my Rocket League series. It started as an AP Statistics project in 2020, grew into 35,000+ analyzed drops, and ends with me trying to reverse-engineer the game's settings. Thanks for following along, and good luck with your next cup.

- Dallin King

Keep reading