CSR2 Star Ratings: Count, Colour and Their Value
Star count and star colour measure two different things in CSR Racing 2. We checked all 1,232 rated cars to find what each one is actually worth.
Stars in CSR Racing 2 confuse people because they carry two separate pieces of information in one symbol. How many stars a car has is one thing. What colour they are is a completely different thing. Players routinely conflate the two and then wonder why a five-star car loses to another five-star car.
The record sheet has a star count for 1,232 of its 1,309 cars and a colour for all of them, so both questions can be answered properly rather than argued about.
Star count tracks the tier, not the car
Here is how the counts are distributed across the sheet:
| 5 stars | 998 | |
|---|---|---|
| 4 stars | 153 | |
| 3 stars | 50 | |
| 2 stars | 16 | |
| 1 star | 15 |
81% of rated cars are five-star, so the rating separates almost nothing at the top.
Nearly everything on the sheet is a five-star car. That alone should tell you the count is not a quality ranking — if four out of five cars score top marks, the mark is not doing much work.
What it does track is the tier you are in:
| Tier | Median star count | Range |
|---|---|---|
| Tier 1 | 2 | 1 to 3 |
| Tier 2 | 3 | 1 to 3 |
| Tier 3 | 4 | 2 to 5 |
| Tier 4 | 5 | 3 to 5 |
| Tier 5 | 5 | 3 to 5 |
A five-star Tier 3 car and a five-star Tier 5 car are not comparable in any sense. The first is at the very top of what its tier offers; the second is the ordinary case. Star counts are a within-tier signal and nothing more.
Within one tier, the count does mean something
Hold the tier constant and the relationship appears cleanly. Among Tier 5 cars, where race distance is identical:
| 3 stars (8 cars) | 9.293s | |
|---|---|---|
| 4 stars (70 cars) | 8.473s | |
| 5 stars (779 cars) | 7.232s |
Each step up is worth around a second at the median, within the same tier.
Three stars to four is worth 0.820 seconds at the median. Four to five is worth 1.241. These are large numbers compared with anything else we have measured, far bigger than Elite Customs per level or purple stars.
Why almost everything is five-star
The 81% figure looks like rating inflation until you break it down by tier, at which point it turns out to be a composition effect:
| Tier | Rated cars | Five-star | Share |
|---|---|---|---|
| Tier 1 | 15 | 0 | 0% |
| Tier 2 | 29 | 0 | 0% |
| Tier 3 | 79 | 6 | 8% |
| Tier 4 | 231 | 192 | 83% |
| Tier 5 | 878 | 800 | 91% |
No Tier 1 or Tier 2 car on the sheet has five stars. By Tier 5, nine in ten do. The reason 81% of the whole database is five-star is simply that 71% of the database is Tier 5, where five stars is the norm.
So the rating is not inflated. It is tier-relative, and the sheet is dominated by one tier. Read within a tier it still separates cars usefully — within Tier 3, only 6 of 79 cars manage five stars, and those six are worth knowing about.
Colour is the part you can actually measure
Star colour is cleaner, because the sheet contains 235 cars recorded both ways: same model, same tier, same tune, one copy gold and one purple.
Across those pairs the purple copy is quicker 218 times, slower 7 times and level once. The median gain is 0.050 seconds, the mean 0.054, and the range runs from 0.268 seconds down to 0.129 seconds the wrong way.
Of the 1,232 rated cars on the sheet, 786 are purple-star and 446 are gold. So purple is the majority at this level of the game, and increasingly the default assumption at the top — every car in the global top ten carries purple stars.
Purple is a late-game phenomenon
| Tier 5 | 68% | |
|---|---|---|
| Tier 4 | 50% | |
| Tier 3 | 29% | |
| Tier 1 | 25% | |
| Tier 2 | 11% |
Two thirds of Tier 5 is fused. In Tier 2 it is one car in nine.
Colour behaves very differently from count. It climbs steeply with tier because fusion is something players do once they have committed to a car, and almost nobody commits to a Tier 2 car.
The practical read is that purple stars stop being optional somewhere around Tier 4. At 50% of that tier and 68% of Tier 5, turning up without them increasingly means giving away a twentieth of a second to most of the field. Below Tier 3 it is close to irrelevant.
Putting the two together
| What you are looking at | What it tells you | Comparable across tiers? |
|---|---|---|
| Star count | Where the car sits within its own tier | No |
| Star colour | Whether it has been fused | Yes |
| Recorded time | How fast it actually is | No |
| EVO | Build strength, tracks time well | No |
The short version. If two cars in the same tier have different star counts, the higher one is very likely the faster car and the gap is probably large. If two cars have the same count but different colours, the purple one is probably faster by about a twentieth of a second. And if you are comparing cars in different tiers, stars tell you nothing at all.
What to do about it
- Use star count to shortlist within your tier. It is a strong signal there and a useless one across tiers.
- Do not chase purple on a car that is not already competitive. A twentieth of a second does not rescue a mid-pack car.
- Do chase purple at the sharp end. Every car in the game's top ten has it, and at that level 0.050 seconds decides races.
- Treat a five-star rating as ordinary, not special. Four out of five rated cars on the sheet have one.
Every car in the record database shows its star count and colour alongside its time and tune, so you can filter to purple-star cars only and see the difference for yourself.