The newest post in the Challenges forum prompted us to create this topic - specifically, the fact that participants are always asked to post the upgrade level of their car (15/27, for example).
Early on in the design process for MRR, we knew we wanted to make our own leaderboards, one for each event. That would essentially automate (and deprecate) the Challenges forum. But from our experience with other challenge forums, and with FM's failed implementation of TimeTrials, we realized there's one missing component we feel is needed to make the results lists more fair.
The problem is that different players might be racing with different cars, or even the same car at different upgrade levels. But using the 15/27 method doesn't really give us anything useful. Our initial idea was to incorporate the PR values already present in RR3 (and backport those values to 1.1.2). Then, in the results pages, we would sort by time, *then* by PR - the idea being that, if two drivers turn in the exact same time, the one driving the inferior car (lower PR) must be the superior driver, and should, therefore, have a higher position in the results. But then we thought that, like most other things the Monkeys did, PR probably isn't all that reliable. That is to say, two cars of equal PR are probably not really equal in performance.
The solution?... Well, that's what this post is all about. We're not sure we can come up with a realistic version of PR based on the specifications of the various cars because the specs from one car to another are probably just as flawed as the PR numbers. But for the same car, we have some ideas...
Consider the Base Specs of a car to be "level zero"... and the Max Specs (fully upgraded) to be 100% - in each of the spec categories: top speed, acceleration, stopping, and grip. So for any given car, you'll get 4 different numbers between 0 and 100. Averaging those 4 numbers will give us a single number from 0 to 100%. Take the Ford Focus as an example:
| Top Speed | Acceleration | Stopping | Grip | |
|---|---|---|---|---|
| Base: | 163 mph | 5.9 s | 127 ft | 0.94 g |
| Max: | 183 mph | 5.3 s | 117 ft | 1.01 g |
| Diff (= 100%): | 20 | 0.6 | 10 | 0.07 |
| Example values: | 175 mph | 5.7 s | 119 ft | 0.96 g |
| Example %: | 55% | 33% | 80% | 29% |
So this would give our example Ford Focus RS an "Upgrade Rating" of 49%. [We're also wanting suggestions about what to call this value. We want to avoid "PR" so as not to confuse it with RR3.]
Another option would be to take into account the fact that upgrading some "categories" give better overall performance benefits than others. Most players agree that acceleration and grip are more important than top speed or stopping. So the categories could be "weighted" to give more importance to some over the others. But then you could argue that top speed *is* important on some circuits - like Spa and Mount Panorama, to name only two.
And then there's still the issue of whether this is a realistic comparison between cars of different type. We think not. Just as an example... the Silvia at 50% probably doesn't perform as well as the Focus at 50%.
So we would welcome feedback from the community on this. Our ultimate goal is to present a "Leaderboard" that is as representative of actual driver abilities as possible.
