[Lancia037Rally]

Never thought I'd see Antoine L'Estage's speedfactor just .04 from a Corvette's driver. Antoine is always challenging Pat and ACP.

Edit: Antoine's speedfactor has been adjusted.

 

[jake himes]

Are you familiar with Doug Jenkins?

As we all know the speedfactor system is just a model based on data. The more data we have the better the model is. Sometimes competitors feel mistreated by the system but they always have the chance to prove it on the stage.

 

[Doug Woods]

>Sometimes competitors feel mistreated by the system but they
>always have the chance to prove it on the stage.

Antoine will prove it.

Doug Woods

 

[FMF]

I don't think the speed factor should override the seed NO MATTER WHAT. Like this situation you don't run here much maybe have some mechanicals on a stage get reseeded behind seed 8's way back in club and spend your time passing people on stage and tangled up in all their messes.

 

[patr]

speedfactor is totally lame compared to seeding IMHO

 

[John Cassidy]

I think speedfactor would be more meaningful if it were a North American convention, not just used in the US.

Cheers! John

 

[patr]

and took into account stage length, and car type, international results, and ability to finish

it needs to all be taken into account, if it doesn't change, then it should stay what it was, a 'hint' for the steward to organize the cars. Now it seems to be the gospel.

 

[Mad Mike]

>and took into account stage length, and car type,
>international results, and ability to finish
>
>it needs to all be taken into account, if it doesn't change,
>then it should stay what it was, a 'hint' for the steward to
>organize the cars. Now it seems to be the gospel.


Hallelujah Brother Patrick!

Here's what I sent the PRB ...

13. Review Speed Factor changes. Only count speed factor for completed rallies.

I?m in favor of ANYTHING that places value back on finishing rallies rather than further diluting the sport into a single-dimensional quest for outright speed. Speed Factors need to be informational-only for the use by the Stewards instead of mandated in the rulebook. We need to find a way to stop comparing apples to oranges and we REALLY need to identify (by way of asterisks or whatever) those speed factors established by the twice-diluting procedure we use at stand-alone ClubRallys. While we?re at it, it would be good to figure out some formula to compute a more realistic speed factor when changing classes (up or down) even if it?s only a fraction of the hard numbers generated by said computation. Lastly, where did we discover the rationale that says an event or two?s inactivity means you are slower?

I suggest you and anyone else keen to avoid leaving speed factors the only ranking system should fire off comments to the board NOW!

Halley ...
ProRally #86
http://www.realautosport.com

 

[randyzimmer]

SCCA Speedfactor

Well, took a look at the numbers.
Between the first rating listed at R.A. and Sno Drift, Pat lost from .9 to .89 = .01 and I lost .04 from .86 to .82 from not running SCCA events. After that, I stayed the same at .82. Is the deterioration of SF a thing of the past or will I be moved down more since I pointed it out?
Anyway, the whole system is a mess as guys who could never stay with Diggins are equal or close and if deterioration sets in, will be SF'd ahead of him in a couple months!
Without a dominant driver and team, there will be a whole gaggle of guys at .95 and above, the two minute window will end up covering half the field before long. Looks like this SF experiment has hit the skids.
rz
edit: by the way, a change from .86 to .82 puts ~70 people eligible to start ahead!

 

[patr]

RE: SCCA Speedfactor

the only real issue I have with it is this, say you win a 25 mile stage, but you lose the half mile superspecial twice. Those superspecial losses have just as much weight as the 20 mile stage, even though competition wise, a .01 adjustment for the 25 miler equals a lot more 'actual time' and in fact is what matters event wise. This is all easy to fix, to use a weight factor for stage length. It doesn't matter much except that you do get seeded in odd ways because of this.

 

[John Cassidy]

RE: SCCA Speedfactor

I know I'll open another can of worms here, but why is Seeding determined differently in the US and Canada as well?

I've had to petition the SCCA to have my Seed increased on this side of the border as it's increased in Canada.

So, it seems to me that neither system is perfect. Obviously, the import of a knowledgeable steward in terms of determining start orders is key.

Cheers! John

 

[starion887]

RE: SCCA Speedfactor

You and RZ have pointed out some of the oddities of the SF system. I've been seeing this in the ESRC this year; get stuck in one mud hole: you drop like mad. Have Seamus drop off the road; and the survivirs' SF's all shoot up fantastically!

I would like to see no more than a .02 drop due to aging. And the stage miles weighting, or eliminating SF from very short stages, should be added into the formula.

A number of us complanined 2 years about the 'double-secret' formulas used. Dunno if that ever got published, but it should be.

And Patr, it can make a huge difference how this is done; maybe not upfront, but in mid pack. You can start 15 places back in a well attended event like STPR or MFR due to SF; if there is dust in midpack, where there is no 2 minute dust window, then a good fast team can really be screwed. Hence RZ's pointing out how a .04 drop in SF due to aging is dramatic. Not all of the 70 drivers he dropped behind will show up, but if 1/4 do, he has dropped 17 start postions.

Regards,
Mark B.

 

[DaveHintz]

RE: SCCA Speedfactor

The problems with starting based on seed were just as bad or worse. Remember John Lane being forced to start behind Kendall Russell? He wasn't slow or inexperienced, but it was pretty hard to get to a seed 5 doing a couple of divisional rallies per year. Everybody in the know knew it didn't make sense. Finishing in the top 15 of a national rally when there were only 15 finishers wasn't much of an accomplishment and people did get seeding points for that in the mid 90's.
The grass is always greener on the other side of the fence.
Dave
Car # 207

 

[Bruce]

>it needs to all be taken into account, if it doesn't change,
>then it should stay what it was, a 'hint' for the steward to
>organize the cars. Now it seems to be the gospel.

If you think this is true, you clearly haven't watched ANY steward create a start order. Just because you see a "start order" on the Internet based on speed factor (or when the entry came in - or anything else) doesn't mean that's where the car will start.

Speed factor is just a tool...an admittedly imperfect one...but it's the only tool we have besides seed - and seed, while it's a good system, doesn't have sufficient granularity to be effective.

I've established start orders with seed, speed factors, smoke, mirrors, and a little black magic. I'll use any tool I can get. A safe start order is the ONLY goal...road position is neither a reward nor a punishment for anything.

Bruce

 

[gplsek]

I think the speedfactor is a good idea, but it has lots of problems and needs to be tunned. I have seen lots of good ideas here and some decent propoasals, now we just have to hope that the PRB will take some into a consideration.

-george

 

[patr]

bruce, i've sat with stewards many times, especially wrt to the seeding

TWICE this year, there were pre-start decisions justified by 'the speedfactor''. I know that each steward does things differently, but these days the factor is being taken too literally. If antoine doesn't start somewhere near the front, (i.e. top 5-8 cars) it would be a prime example of the problem.

no system is ever perfect, and IT IS A BLACK ART, thus speedfactor SHOULD be a hint, but since the standard practise is to use it verbatim, we should take every effort to improve upon it

i did send in my suggestions to the PRB, by the way (about adding stage length modifiers).

 

[Bruce]

Speed factor is a hint...but if it's the only hint you have, it's your main tool. If a guy comes out of nowhere, and the only information I have is the speed factor (that isn't either self-serving or hearsay) then speed factor is what I have to use.

Bruce

 

[jandreini]

Seeding worked well. Problems were rare, anecdoatal, and mostly confined to beginners. Speed factors have not resolved the problems with beginners(only a requirement for lengthy ClubRally experience can solve that). They have, however created a growing problem at and near the top(oh, boy!). They require massive massaging of start orders at every rally. They should be returned to the antipodal netherworld from which they came.

 

[Lancia037Rally]

>If a guy comes out of nowhere, and the
>only information I have is the speed factor (that isn't
>either self-serving or hearsay) then speed factor is what I
>have to use.

But Antoine didn't come from nowhere. He has been a CARS National podium contender for several years yet someone had decieded to give him a lowball speedfactor out of "nowhere". His SF has since been adjusted on the MaineForest entry list to another 'guess'. Hmmn, whomever just guessed .94 so that he would not be included in the top .95 draw.

If SpeedFactor is to work, all N. American rallies should be factored for drivers like Randy Zimmer, Peter Thomson, ACP (whose non-SCCA existant speedfactor started him way back at SnoDrift although he won many of the stages), John Cassidy, Julien Pilon, and even PatR before he returned to SCCA have fair and consistent speedfactors.

Additionally, recent under-subscribed ClubRallies have boosted speedfactors for the finishing drivers.

 

[jake himes]

Seeds make me want to barf.