Driver Indexed Class? (for Classic "Pro" series)
Posted: Tue Jun 17, 2014 4:13 pm
So, our Classic Event poll revealed mostly that people like what we're doing with the Classic events. They like a little variety, and they like lots of runs. Cool.
But, enough people want to see a "Classic Pro Championship" that we're looking into ways to make it happen. We got enough feedback from people who like the casual "practice" nature of the Classics and DON'T want it to be serious competition that if we do this, it's going to be an "opt-in" thing. The people who want to take it seriously and run for points will have to opt-in early in the season. And it will be a "serious" competition. We'll make the rules such that there won't be a lot of dropped events, and there won't be a bunch of points awarded if you "win" when there's no competition in your class that day. No "attendance awards" here. If you win the Classic Pro championship, you earned it.
Now, some of the SCCA guys have suggested making it a single "pro class" and using the standard "PAX index" to equalize the drivers. (I'll leave it to the reader to Google that if you don't know what it is) I'm going to tell you right now, we WON'T be doing that. I think the PAX is a cool thing, and IF you're dealing with people who have cars that are a) "the car for the class" and b) well-prepared to the SCCA rules, it's reasonably accurate. But, on the local level, hardly anybody drives a car that is that well-prepared. And FAST has a pretty big number of people who have cars that are prepared with no regard for SCCA classing... and they'd all get screwed to the wall. No, PAX is not for us.
However... a long time ago, in a galaxy far, far away... I dreamed up something called the Driver Index Class. We ran it for a couple years when I was in Kansas, and it was pretty cool. It's simple, but the reality of how it ends up working is a little complicated. I'll try to explain.
Driver Indexed Class
The whole idea is to create a class where the driver's ability shines above all else. It works similar to the PAX index, but instead of indexing the car, we index the driver. How do we do that? We go back 6 months and look at that driver's results over the past 6 competition events. For each event, we divide the FTD by that driver's time (and we'll probably use street tire FTD, because it's a little more stable than race tire FTD that varies wildly depending on who shows up). So, if the driver *has* FTD, their index would be 1.000. If FTD was 49.000 seconds and our driver drove 50.000 seconds, their index would be .980. We average the most recent 6 events, and that is the Index for that driver.
Here's the kicker... to account for any changes that a driver might make to their car, or epiphanies that driver might have throughout the season, AND to minimize the effect of sandbagging, the index is updated after each event. So, for each event, you'll get a new index that includes your performance from the last event. I'll be honest, it's a fair bit of work to administer this class. But, it's fun, I don't mind doing it for a small group. If you're consistent, you will do well. If you have a bad day, it REALLY shows up.
Some of the things we learned from it were that good drivers are very consistent. And with this format, you're constantly competing against YOURSELF. If you can manage to improve your position relative to the FTD, you "beat" your index for that event, and you'll finish better among the DIC class (go ahead and start the DIC jokes). But, then you've set a new "standard" for yourself, so if you don't do as well next time, you've made your index harder... and you could lower your standing in the next results. The only way to really beat the system is to improve slightly with each and every event, which is why it's not possible to allow novice drivers in this class. By the nature of being a novice and constantly learning and improving, a novice could actually win this class! The other thing is that for it to be at least reasonably accurate, we can't allow car-hopping. A one-time change in car or a single "major modification" to a car will rectify itself over a couple events by the nature of the self-adjusting index. But, shuffling between several cars would never give accurate results.
Quick example:
The coolest thing about it is that it allows a driver in a "slow" car to compete against a driver in a "fast" car without regard to how well either car is prepared relative to the rules. You'll both race "head-to-head" correcting your actual time by your Index factor. The thing people didn't like about it was that to really know how you were doing against your DIC competition (keep laughing, I'll wait), you had to know your current index AND their current index, and do a couple of calculations. Simple solution to that? Just drive your ass off and drive to the best of your ability each and every time and don't worry about it! Of course, with the technology we have today, it would be easy to keep the current indexes published on the website, and for anyone to check them on their phone at any time. So, I think it could work a little better now than it did 15 years ago.
Anybody like the idea of doing a Driver Indexed Class, or do you all think I'm nuts? (I guess the two options are not mutually exclusive)
But, enough people want to see a "Classic Pro Championship" that we're looking into ways to make it happen. We got enough feedback from people who like the casual "practice" nature of the Classics and DON'T want it to be serious competition that if we do this, it's going to be an "opt-in" thing. The people who want to take it seriously and run for points will have to opt-in early in the season. And it will be a "serious" competition. We'll make the rules such that there won't be a lot of dropped events, and there won't be a bunch of points awarded if you "win" when there's no competition in your class that day. No "attendance awards" here. If you win the Classic Pro championship, you earned it.
Now, some of the SCCA guys have suggested making it a single "pro class" and using the standard "PAX index" to equalize the drivers. (I'll leave it to the reader to Google that if you don't know what it is) I'm going to tell you right now, we WON'T be doing that. I think the PAX is a cool thing, and IF you're dealing with people who have cars that are a) "the car for the class" and b) well-prepared to the SCCA rules, it's reasonably accurate. But, on the local level, hardly anybody drives a car that is that well-prepared. And FAST has a pretty big number of people who have cars that are prepared with no regard for SCCA classing... and they'd all get screwed to the wall. No, PAX is not for us.
However... a long time ago, in a galaxy far, far away... I dreamed up something called the Driver Index Class. We ran it for a couple years when I was in Kansas, and it was pretty cool. It's simple, but the reality of how it ends up working is a little complicated. I'll try to explain.
Driver Indexed Class
The whole idea is to create a class where the driver's ability shines above all else. It works similar to the PAX index, but instead of indexing the car, we index the driver. How do we do that? We go back 6 months and look at that driver's results over the past 6 competition events. For each event, we divide the FTD by that driver's time (and we'll probably use street tire FTD, because it's a little more stable than race tire FTD that varies wildly depending on who shows up). So, if the driver *has* FTD, their index would be 1.000. If FTD was 49.000 seconds and our driver drove 50.000 seconds, their index would be .980. We average the most recent 6 events, and that is the Index for that driver.
Here's the kicker... to account for any changes that a driver might make to their car, or epiphanies that driver might have throughout the season, AND to minimize the effect of sandbagging, the index is updated after each event. So, for each event, you'll get a new index that includes your performance from the last event. I'll be honest, it's a fair bit of work to administer this class. But, it's fun, I don't mind doing it for a small group. If you're consistent, you will do well. If you have a bad day, it REALLY shows up.
Some of the things we learned from it were that good drivers are very consistent. And with this format, you're constantly competing against YOURSELF. If you can manage to improve your position relative to the FTD, you "beat" your index for that event, and you'll finish better among the DIC class (go ahead and start the DIC jokes). But, then you've set a new "standard" for yourself, so if you don't do as well next time, you've made your index harder... and you could lower your standing in the next results. The only way to really beat the system is to improve slightly with each and every event, which is why it's not possible to allow novice drivers in this class. By the nature of being a novice and constantly learning and improving, a novice could actually win this class! The other thing is that for it to be at least reasonably accurate, we can't allow car-hopping. A one-time change in car or a single "major modification" to a car will rectify itself over a couple events by the nature of the self-adjusting index. But, shuffling between several cars would never give accurate results.
Quick example:
- My per-events indexes for the past 6 events are all .9 (neat, huh?), so my current DI is .900
- For this event, I run a time of 50 seconds. We multiply that by the DI to get my Indexed time of 45.000 seconds. And that's what we'll compare to Indexed time of all of the other driver's in class to determine my finish position for this event.
- For the next event, my per-event index from this event (FTD/my time, let's say FTD was 44.555... so, 45.555/50=0.911) is averaged with the 5 previous .900's to give my new index of 0.902. As you can see, I "beat my index" for this event by driving a little faster relative to FTD, and because of that, my index for the next event went up a little. If I continue that trend or at least hold to that standard, my index will continue to slowly increase as the older/lower indexes drop out of my average. If I slip up at the next event, then my index will drop slightly.
The coolest thing about it is that it allows a driver in a "slow" car to compete against a driver in a "fast" car without regard to how well either car is prepared relative to the rules. You'll both race "head-to-head" correcting your actual time by your Index factor. The thing people didn't like about it was that to really know how you were doing against your DIC competition (keep laughing, I'll wait), you had to know your current index AND their current index, and do a couple of calculations. Simple solution to that? Just drive your ass off and drive to the best of your ability each and every time and don't worry about it! Of course, with the technology we have today, it would be easy to keep the current indexes published on the website, and for anyone to check them on their phone at any time. So, I think it could work a little better now than it did 15 years ago.
Anybody like the idea of doing a Driver Indexed Class, or do you all think I'm nuts? (I guess the two options are not mutually exclusive)
