Performance Index Calculation?

[kl3640]

Of course you subject yourself to all the "sanbagger awards" in such a case. 

There are some really fast Amateurs who are in their first full AM season, and they just haven't asked for advancement, though they certainly will make the bump list at the end of the season.  Unless you're winning all of the AM races and then decline advancement when you make the off season bump list, it probably wouldn't be sandbagging since you wouldn't be winning too much anyway.

The real sandbaggers are those who cleaned up as AM's, then took a season off or something and come back as AM's instead of EX's and clean up.

[ktd]

It seems to be:

[(T-A)/T]*1000 = performance index

where T = the total number of people in your races - including you.
        A = the total number of people that finished ahead of you.

That formula isn't adding up for me.

[kl3640]

That formula isn't adding up for me.

Why not?

10 riders.  You finish first.

((10-0)/10) = 1 * 1000 = PerfInd of 1000

10 riders.  You finish fifth.

((10-5)/10) = .5 * 1000 = 500 = PerfInd of 500

10 riders.  You finish dead last.

((10-10)/10) = 0 * 1000 = 0 = PerfInd of 0

[ktd]

Why not?

10 riders.  You finish first.

((10-0)/10) = 1 * 1000 = PerfInd of 1000

10 riders.  You finish fifth.

((10-5)/10) = .5 * 1000 = 500 = PerfInd of 500

10 riders.  You finish dead last.

((10-10)/10) = 0 * 1000 = 0 = PerfInd of 0

I got you there.  It just didn't match what they gave me.
That is for one race.  Do the average your last race score with the latest race score?
I got to do my last score I think and average them.

[ktd]

It seems to be:

[(T-A)/T]*1000 = performance index

where T = the total number of people in your races - including you.
        A = the total number of people that finished ahead of you.

Ok A = Total number of people that finished ahead of you.  If I finished 5th 4 people finished ahead of me.  Should A be My position as in your example?

[Chris410]

Yes, you could enter a number of races however, you need to score points in order to apply that towards becoming expert. 

Here are how the points are broken down per the 08 rulebook:

8.1 CCS POINTS - Points will be awarded based upon the fi nal results of each class
run at each race using the following scale:
First 35 Ninth 16 Seventeenth 8
Second 30 Tenth 15 Eighteenth 7
Third 26 Eleventh 14 Nineteenth 6
Fourth 23 Twelfth 13 Twentieth 5
Fifth 21 Thirteenth 12 Twenty-fi rst 4
Sixth 19 Fourteenth 11 Twenty-second 3
Seventh 18 Fifteenth 10 Twenty-third 2
Eighth 17 Sixteenth 9 Twenty-fourth 1

8.1.1 Points as described above are awarded based upon order of fi nish for all
riders, irrespective of number of entries in the class.
8.1.2 Points as described above are awarded based upon order of fi nish as
well as rider status (Expert or Amateur) except in those classes that
have no divisions

[ktd]

I'm just lost. sorry.  Not even thinking of expert yet.  Just learning the whole deal.

Supertwins they have me 8th 37points 474.3 PI
Formula 40 12th 35Points  615.4 PI
I was trying to figure the PI out using your formula out of curiosity of what it means. 

[kl3640]

I'm just lost. sorry.  Not even thinking of expert yet.  Just learning the whole deal.

Supertwins they have me 8th 37points 474.3 PI
Formula 40 12th 35Points  615.4 PI
I was trying to figure the PI out using your formula out of curiosity of what it means. 

I owe you an apology, I made a mistake that may have contributed to your confusion:

In my first example, I used A as the # of riders who finished in front of you.  In the second and third, I used it as the absolute finish position.  If in fact A is supposed to be the # of riders who finished ahead of you, what you could do is simply change the formula to substitute (P-1) for A where P is your finish position, as such:

( (T-A)/T)*1000=PI, and IF A=(P-1) then
( ( (T-(P-1))/T)*1000=PI

where

T=Total riders on the grid (I'm not sure if they use riders on the grid or entries, I think the former though)
P=Your finish (e.g., 5th place)

So if you finished 5th out of a grid of 10, then PI = ( ( (10-(5-1))/10)*1000=600.

I'm not sure if A is supposed to be the # of riders who finished ahead of you or the position in which you finished, although someone indicated earlier in this thread that it should be the # of riders who finished in front of you.  If the former, then, even if you finished last, you'd get something for your PI (presumably a credit for just having completed the race).  If the latter, then you'd get nothing for finishing the race dead last.

As far as your question about the PI over multiple events, I believe that it is just a rolling arithmetic mean of all previous PI's, irrespective of any factors outside of A or T.  If you want to check CCS's math you could easily look at your previous finishes in a class this season, calculate the PI for each, and then average those PI's (add up all the PI's and divide by the # of events).

Where you may be getting confused regarding your standing in those classes is that you're mixing points with PI, and there is no relationship.  As the 2008 Rulebook's section 8.1.1 (quoted previously in another post in this thread) indicates, the points are awarded without respect to the # of entries, i.e., a certain place finish gives the rider the same # of points in each event, regardless of the # of entries in those events.

So a situation could exist where you have a lot of points relative to the # of events run in a class, but a low PI, because of a low # of entries.  For example:

In a given class X, there are always 5 entries, and you consistently finish 4th in every event.  After 10 events in that class, you'd have:

Points = 230 = (23 * 10)
PI = 400 = ( ( (5-(4-1)/5)

If, on the other hand, there were always 10 entries in that class, and you consistently finished 5th, after 10 events you'd have:

Points = 210 = (21 * 10)
PI = 600 = ( ( (10-(5-1))/10)*1000

The rulebook (Section 8.2) explains the use of points for determining class and overall championships, and the purpose of the Perf Indx (which appears to be limited to Regional Top Ten Numbers assignment and Regional Overall Championships, not individual class championships, if I'm reading the rulebook correctly).

[SVbadguy]

Nobody has been able to tell me how the final P.I., that is multiplied against the points, is calculated.

An average of my per race P.I. is quite a bit higher that what was used to calculate my adjusted points last year.
103 and 131 point differences result with two different ways I calculated final P.I. compared to how CCS does it.

[kl3640]

Nobody has been able to tell me how the final P.I., that is multiplied against the points, is calculated.

An average of my per race P.I. is quite a bit higher that what was used to calculate my adjusted points last year.
103 and 131 point differences result with two different ways I calculated final P.I. compared to how CCS does it.

How did you arrive at that - did they actually tell you what PI they used to calc your AP, or did you back in to the PI that they must have used by reversing the formula from the section 8.2.2 and 8.2.3 of the rules, using as one of the factors the non-adjusted points listed for you in the Championship Standings?

Perhaps I missed it, but I could not find in the 2008 Rulebook any reference to Performance Index other than in the below referenced two subsections, and they do not include the formula for calculation of the per-event PI itself, let alone the season average PI; rather, they only addresses how the PI is used for the determination within each region of EX Top Ten Numbers (with the caveat that regions run by affiliate organizations may have different processes for assigning Top 10 Plates) and Overall Regional EX & AM Championships.

...[Top Ten Numbers for Experts/Regional Overall Championships for EX/AM]... are assigned in each series by overall points earned in all [Expert only for Top Ten Numbers, EX & AM for Overall Championships] SS, SB, GP, and GT classes, multiplied by the riders [sic] performance index and then divided by 1000.  Expert Top Ten Numbers will be assigned to those riders who rank in the ten highest positions by adjusted points for each region/The rider with the highest adjusted point total will be declared the "Overall Regional Champion" in the Expert or Amateur category.

Note: Affiliated organizations may use their own system for issuing numbers.

[tzracer]

It seems to be:

[(T-A)/T]*1000 = performance index

where T = the total number of people in your races - including you.
        A = the total number of people that finished ahead of you.

It should be quite clear what A and T represent.

The total PI is not an average of you PI in each race. Total PI is calculated the same as for the individual races. T would be the total number of people in all your races, A would be the total number of all people that finished ahead of you in all races. The PI is a type of average, you do not average averages.

Also not all classes may be used to calculate your total PI. Classes such as supertwins is not used to calculate points towards the number one plate, these classes may not be used when calculating the total PI.

[kl3640]

It seems to be:

[(T-A)/T]*1000 = performance index

where T = the total number of people in your races - including you.
        A = the total number of people that finished ahead of you.

It should be quite clear what A and T represent.

The total PI is not an average of you PI in each race. Total PI is calculated the same as for the individual races. T would be the total number of people in all your races, A would be the total number of all people that finished ahead of you in all races. The PI is a type of average, you do not average averages.

Also not all classes may be used to calculate your total PI. Classes such as supertwins is not used to calculate points towards the number one plate, these classes may not be used when calculating the total PI.

Quite correct, A=(p-1), where p= your finish position - my apologies for any confusion.

Also, the calculation of seasonal PI by adding up finishes rather than averaging PI's would make sense, as it allows larger/smaller events to be weighted proportionately.