Every Number on the CSR2 Car Screen, Explained
PP, EVO, stars, dyno, tier and rank — what each number in CSR Racing 2 measures, what it predicts, and how much it is actually worth in seconds.
CSR Racing 2 puts six numbers in front of you and explains none of them. Worse, they disagree with each other, so the obvious strategy of picking whichever looks highest gets you a car that loses to something with lower figures.
This is a reference for all of them. Every value quoted comes from measuring the full record sheet of 1,309 cars, and each section links to the full working if you want to check it.
PP — Performance Points
What it measures: how much performance has been fitted to the car. It rises when you install parts.
What it predicts: very little. Across Tier 5 it tracks recorded time at a correlation of only -0.20, and in Tier 1 the figure is +0.06, meaning no relationship at all. Pick any two Tier 5 cars at random and there is a 31.8% chance the quicker one has the lower PP.
The structural reason is concrete. Across 70 cars recorded at multiple Elite Customs levels, PP does not change in a single one, while the car gets a median 0.907 seconds faster. PP is blind to the single biggest upgrade in the game.
It also stops varying at the top: all 100 cars in the global top hundred carry 712, 713 or 714 PP.
EVO
What it measures: overall build strength, including whatever Elite Customs does that PP cannot see.
What it predicts: a great deal. Across Tier 5, EVO tracks recorded time at a correlation of -0.96. It explains roughly 92% of the variation in times, against about 4% for PP.
A rough conversion from the data: every 100 EVO points is worth about 0.185 seconds in Tier 5. You can estimate a car's time as 11.4264 minus 0.001849 times its EVO, and 41% of the field lands within five hundredths of that.
Where it fails is the sharp end. The second-fastest car in the game carries 2,525 EVO while cars behind it carry more. Once everything is built out, EVO stops ordering cars.
Stars — count and colour
These are two separate pieces of information in one symbol, and conflating them causes most of the confusion.
Star count describes the car's standing within its own tier. It tracks time well inside a tier: five-star Tier 5 cars post a median 7.232 against 8.473 for four-star. It does not travel across tiers — no Tier 1 or Tier 2 car on the sheet has five stars, while 91% of Tier 5 does.
Star count is also not a build meter. 614 five-star Tier 5 cars carry less EVO than the best four-star car.
Star colour is fusion state, and it is measurable because 226 cars appear on the sheet in both gold and purple. The purple copy is quicker in 218 of those pairs, by a median of 0.050 seconds.
Dyno time
What it measures: the game's prediction of your time on a flawless run.
What it predicts: a floor, not a target. 1,230 of the 1,246 cars with both figures beat their dyno, by a median of 0.115 seconds. Only 13 cars finished slower than it, and the worst shortfall in the whole database is 0.011 seconds.
The gap between dyno and real time is the part a driver controls, and it is wildly uneven. It is worth a median 0.316 seconds in Tier 1 and 0.101 in Tier 5. On a car under 1,500 EVO it is worth 0.184; above 2,500 EVO it is worth 0.048.
Dyno is also what Live Races match you on, which is why two cars paired as equals can be half a second apart on real pace.
Tier
What it measures: which class the car races in. Race distances differ between tiers.
The consequence people miss: times are not comparable across tiers in either direction. The median Tier 4 car runs 10.285 and the median Tier 5 car runs 7.353, and that three-second difference is not a measure of anything you can bank.
The tiers are also very different shapes. Tier 1 spans 0.890 seconds from fastest to slowest; Tier 4 spans 6.660. Car choice matters enormously in one and barely at all in the other.
Rank
What it measures: position in the order, globally or within the tier.
What to watch out for: it is far less precise than it looks. 1,251 cars produce only 1,004 distinct times, 247 adjacent pairs are exact ties, and another 318 are separated by a single thousandth. A thirty-place difference in rank can represent no measurable difference in speed.
The global top hundred is also entirely Tier 5, so a global rank is not a target a car in any other tier can pursue.
The tune: nitrous, final drive, tyres
Three settings, and they behave very differently from one another.
Tyre pressure is close to solved: 70% of the sheet runs 0/100, with smaller clusters at 52/48 and 53/47. Final drive is largely decided by tier, with a median of 3.89 in Tier 1 falling to exactly 2.00 in Tier 5, where 516 cars share that one value.
Nitrous is the genuinely open one, with 404 distinct values and two incompatible philosophies — a hard hit over about four seconds, or a long gentle burn over twelve. Both appear on records, which is a good sign that it is where experimenting still pays.
Which number to use for what
| Question | Use | Do not use |
|---|---|---|
| Which car is faster? | Recorded time | PP, stars |
| How built is this car? | EVO | Stars, PP |
| Is my build finished? | EVO, star colour | PP |
| How much can I gain by driving? | Dyno minus recorded time | PP, EVO |
| Who will I face in Live Races? | Dyno | Recorded time |
| Is this car good for its class? | Star count, rank in tier | Global rank |
The single rule underneath all of it: only recorded time is a result. Everything else on the screen is a description of the car or the build, and descriptions lose information by design.
Each number above has a full write-up with the workings behind it, and every car on the sheet shows all of them side by side in the record database.