◆ SNAIL [Spec] Racing - Currently Recruiting for GT7 - JOIN TODAY!!Open

[hooker4]

Had a good weekend. Got to meet Eric aka Dragonwhisky in person. He and I finished up some mods to his rig. Had some good times chatting. Enjoying so Chivas Regal 18. Hopefully his wife doesn't want to kill me after he got to try triples and dual buttkickers. I started to film him just a lil late. His earlier reactions prompted me to chase down my phone to document this. Anyway here is a video of Eric's maiden voyage on my rig.

Surely he will give his review of how it was.

How do you run 3 monitors? Thats awesome!!👍

 

[kcheeb]

That still doesn't explain the Owens/Worst/Martin issue. I would assume it has something to do with pole bonus points not being taken into account when the individual races are adjusted and then added up.

Wasn't trying to explain the differences, just remove some of the clouds I'll leave the explaining to zer0

Each of the three calculation methods have the pole bonuses calculated as advertised.

 

[zer05ive]

I built this spreadsheet from the first two weeks of August for zer0 yesterday

The data it has is:

1. the current adjustment formula
2. the current adjustment formula, with a way to change the values used, factor and driver base count.
3. the new formula

It compares each of the categories nine ways from Sunday, or thereabouts

Thanks cheeb. As always your data-fu is simply amazing!

In fact, when comparing your data with the adjusted results I listed at the bottom of this post, I realized that I made a serious error when adding up EAO-HarleyMan's score. I'll get that corrected asap.

Okay, I've corrected EAO-HarleyMan's score at the bottom of this post. It changes the "what if" prizes a bit, but nothing too drastic. Again, this was just due to an error in my math when doing it manually.

If I'm looking at the spreadsheet right, I see some serious flaws with what is labeled as the new adjustment. How is it possible, or right in any way, for someone who scores 81 pts in a division to be adjusted to a better position than someone who scored 88 in the same division? That should be a giant red flag that something is wrong with that adjustment scenario.

It's because myato1 actually scored 90 points. The only reason he's listed at 81 is because he collected 9 penalty points. With that in mind, is it okay to lower that giant red flag?

I'll put the big red flag down but I will wave a smaller one. On Aug. 3 Owens scored 84 pts., Worst and Martin both had 85 in the same division and Owens would have been ranked higher than both of them. Nobody in that group received a penalty so how is that one explained?

I guess that begs the question, how would you compute the adjusted total (for prizes) if a division winner received a penalty but still won their division? How would the penalty be adjusted out of the equation?


I'm only ten lines in and looking at this in between doing the honey-do list. There could be more that would need to be looked at. At first glance, it appears to favor winning individual races over being consistent for the entire night.

That's because the grid size in D1 that night fell off the friggin' table at the end of the night! They went from 10 drivers on the grid down to only 5 at the end. And since each position lost becomes more detrimental the smaller the grid gets (as it should IMO), the result is an abnormality like the one you found. Simply put, that was the proposed formula's way of "fixing" the issue I pointed out in my original proposal:

For example, if a driver finished last in a field of 15, he only gets 2 points. However, if a driver finishes last in a field of 7, he gets quadruple the amount with 8 points?? That's doesn't seem right..

So although Worst and Martin scored a single point more than Owens using unadjusted points, the new proposed system is saying that if you consider the size of the grids in each race, Owens actually had a slightly better night. But yeah I'd agree that a division going through a drastic change in grid size in the middle of the night definitely complicates the issue. Not exactly sure how we could work around an abnormality like that.

As for your question about how penalties would be adjusted, I don't see a need to adjust them any differently to how they are adjusted now. Currently, I believe the adjustment formula deducts the penalty from the unadjusted score first, and then applies the multiplier afterwards. The new proposed system would do it the same way.

 

[JLBowler]

The work-around is to only apply the adjustment to the division winners to determine the prize winners. Drivers in the same division are on equal footing no matter how many drivers are on the grid. They are all racing against the same number of drivers.

The adjustment is only to compare division winners to each other when they may or may not be on equal footing based on size of grid. There are other factors that put division winning drivers on less equal footing than the winners of other divisions but that's a completely different discussion.

Here is the reverse issue with using this system within the division. What if a race starts with 13 on the grid and there are massive connection issues and the race finishes with 7? Because points are awarded based on number of starters, you would have points awarded to the driver that finished last of those that finished as a 7th place finish in a room of 13. That's an extreme example but you get the point. You would have the same problem with even one driver getting dropped.

 

[kcheeb]

Off topic a little, What kind of discount do snail members get on the granstand? Finally able to afford one.

GrandStand is offering a 10% discount to S.N.A.I.L.s check this post out.

Contact Bill at bill@granstand.com and he'll get you racing in a rig 👍

 

[zer05ive]

The work-around is to only apply the adjustment to the division winners to determine the prize winners. Drivers in the same division are on equal footing no matter how many drivers are on the grid. They are all racing against the same number of drivers.

The adjustment is only to compare division winners to each other when they may or may not be on equal footing based on size of grid. There are other factors that put division winning drivers on less equal footing than the winners of other divisions but that's a completely different discussion.

That's a perfect work-around. 👍 Since this adjustment is for the prize winners only, why not just compare the division winners like we do now? Problem solved.

Here is the reverse issue with using this system within the division. What if a race starts with 13 on the grid and there are massive connection issues and the race finishes with 7? Because points are awarded based on number of starters, you would have points awarded to the driver that finished last of those that finished as a 7th place finish in a room of 13. That's an extreme example but you get the point. You would have the same problem with even one driver getting dropped.

Why would you consider that 7th finisher as last place? All those guys who hypothetically disconnect are treated the same as drivers who have flat tires, wreck, have mechanical failures etc in a RL race. That 7th place finisher isn't last, he's still 7th out of 13. It's not his fault that the rest of the field couldn't get their cars to the finish line!

 

[Handlebar]

There could be more that would need to be looked at. At first glance, it appears to favor winning individual races over being consistent for the entire night.

That would be a step back, imho, if it turns out as Bowler suspects.

Mathematically, it is possible this system even favors winning a race and taking a minor penalty to do it over doing it clean, there is an incrementally larger reward in some cases.

 

[JLBowler]

That's a perfect work-around. 👍 Since this adjustment is for the prize winners only, why not just compare the division winners like we do now? Problem solved.






Why would you consider that 7th finisher as last place? All those guys who hypothetically disconnect are treated the same as drivers who have flat tires, wreck, have mechanical failures etc in a RL race. That 7th place finisher isn't last, he's still 7th out of 13. It's not his fault that the rest of the field couldn't get their cars to the finish line!

Because all the data got put out there for me to see and I had to look at it and try to flesh out any issues that might be there.

Because the 7th place finisher didn't "beat" anyone on the track. Just measuring on track performance, I would call that last place when you don't beat anyone that also finishes the race. That's part of the luck equation in racing. Sometimes you get a good finish without a good effort and sometimes you get a bad finish with good effort. It is racing after all.

 

[zer05ive]

There could be more that would need to be looked at. At first glance, it appears to favor winning individual races over being consistent for the entire night.

That would be a step back, imho, if it turns out as Bowler suspects.

Mathematically, it is possible this system even favors winning a race and taking a minor penalty to do it over doing it clean, there is an incrementally larger reward in some cases.

I feel like I've debunked that suspicion by explaining how that happened in the second section of this post. However, let me know if you disagree. If you want to look at it mathematically, let's break down the formula. I feel it speaks for itself:

16 - z(16/y) = points awarded​

As you can see, 16 is divided by y. Nothing more, nothing less. This means that every driver receives their fair a proportionate amount or points based on where they finish the race. There are no other variables that put a premium on 1st place or even getting on the podium. If there were, there would be more to the formula. There simply isn't. The difference in points between 1st and 2nd place (the value of x) is identical to the difference between 14th place and 15th place. Simple as that. It's like Fox News news.. "Fair and Balanced"

EDIT: If anyone is wondering whether or not this formula favors winning over being consistent, simply look at this post and go to this section:

Points awarded when 16 drivers in the race
y = 16
x = 16/y
x = 16/16
x = 1
z = number of positions behind the 1st place
16 - zx = points awarded
1st = 16 - 0x = 16
2nd = 16 - 1x = 15
3rd = 16 - 2x = 14
4th = 16 - 3x = 13
5th = 16 - 4x = 12
6th = 16 - 5x = 11
7th = 16 - 6x = 10
8th = 16 - 7x = 9
9th = 16 - 8x = 8
10th = 16 - 9x = 7
11th = 16 - 10x = 6
12th = 16 - 11x = 5
13th = 16 - 12x = 4
14th = 16 - 13x = 3
15th = 16 - 14x = 2
16th = 16 - 15x = 1 (last place always wins the value of x)

Is there anything disproportionate about the above points distribution? I would argue not. In fact, it's identical to our unadjusted points distribution, which is an indication to me that the formula does exactly what we'd want it to do. 👍

Because all the data got put out there for me to see and I had to look at it and try to flesh out any issues that might be there.

Because the 7th place finisher didn't "beat" anyone on the track. Just measuring on track performance, I would call that last place when you don't beat anyone that also finishes the race. That's part of the luck equation in racing. Sometimes you get a good finish without a good effort and sometimes you get a bad finish with good effort. It is racing after all.

You're bringing forth arguments that have nothing to do with this formula versus the current one. The validity of how well a driver actually did if half the field disconnects could be argued under the current system as well. But more to the point, I don't think it matters if a driver achieves a given result based on on-track passes or based on other drivers not getting to the finish line. As you said, "it is racing after all." When F1, NASCAR, and Indycar etc start caring about whether drivers passed cars on the track or passed cars because other cars DNF'd, then I'll know the racing world has been flipped on it's head and and having this discussion would make more sense to me.

 

[Handlebar]

Coming in 3rd place in a field of 16, 14 points. In a field if 8, it is only 12 points.

Coming in 6th place in a field of 16, 11 points. In a field of 8, it is 5 points less, only 6 points. With each position being worth twice the points in a field of 8, incurring a minor penalty of 1-2 points to aggressively get a position or two is 'worth it' at least in terms of points.

Use the same field-size multiplier for the penalties in that race? Maybe that would tweak it, but... I can't imagine how many nudge penalties could get filed in a small field.

I think the proposed system rewards 'swing for the fences' racing by inflating the score differentials over a plan of rewarding consistency. I am not for it, personally.

 

[Rallywagon]

Coming in 3rd place in a field of 16, 14 points. In a field if 8, it is only 12 points.

Coming in 6th place in a field of 16, 11 points. In a field of 8, it is 5 points less, only 6 points. With each position being worth twice the points in a field of 8, incurring a minor penalty of 1-2 points to aggressively get a position or two is 'worth it' at least in terms of points.

Use the same field-size multiplier for the penalties in that race? Maybe that would tweak it, but... I can't imagine how many nudge penalties could get filed in a small field.

I think the proposed system rewards 'swing for the fences' racing by inflating the score differentials over a plan of rewarding consistency. I am not for it, personally.

I think adding a equalizing multiplier to the penalties is a good deterrent, lets not forget though, if you purposely hit someone, no matter how hard, it can be leaved as a 5. Intentionally hitting someone is far different than accident bumping, can is generally pretty obvious.

 

[JLBowler]

Like I said in the very beginning. What we had before what we have now wasn't perfect. What we have now isn't perfect and what is proposed isn't perfect either. It just comes down to how do we want to be imperfect.

I'm kind of on board with Handlebar on this one. The larger value gap between each position could create a risk/reward decision where a 1 or 2 point penalty would be worth it to get that one position if the driver was in a position toward the end of the night to win the division and be in contention for a prize.

 

[Handlebar]

The other consideration... (My perspective, sadly)...


In a room of 8, if I come in 6,7,8,6,7,8... Looking at that point total of 36 points is unlikely to encourage new drivers when they see the suggested point differential. Would be 60 pts in current system.

 

[JLBowler]

The other consideration... (My perspective, sadly)...


In a room of 8, if I come in 6,7,8,6,7,8... Looking at that point total of 36 points is unlikely to encourage new drivers when they see the suggested point differential. Would be 60 pts in current system.

Adjusted total for anyone but the division winners shouldn't be shown. This isn't a proposal for a new points system, it's just a proposal to change the formula for comparing division winners in different divisions for the purpose of ordering prize winners.

 

[Rallywagon]

Imho, if I finish like that, I'm not really following points. But that might just be me.

 

[Rallywagon]

Division 3
The room is now open for business.
I will be cycling tracks every 15 minutes.

 

[Nicktune]

Just got my email for GTA!!!!!!

Congrats dgaf!

You did pretty much every GTA practice session....It was a good idea.

 

[dabneyd]

Unfortunately for another week, I can't make racing. I started some work on my car later than I'd hoped and of course it's taking longer than I'd hoped... anyway, someone in D1, please save replays and then PM me for access to Google drive.

 

[nmcp1]

Unfortunately for another week, I can't make racing. I started some work on my car later than I'd hoped and of course it's taking longer than I'd hoped... anyway, someone in D1, please save replays and then PM me for access to Google drive.

Already have acces to the drive so I'll handle D1 replays tonight.

 

[hillracing]

Just got home to find the break in on my Viper didn't work. Does anyone have one they could send?

 

[zegrated]

I'm gonna back you up on this one zer0.

It seems like some people just don't want change, and I'm not sure why, because the new proposal looks very good to me. I think the old system is a little too simplistic in that it has subtracts a set number of points for any positions after first. You are completely right in saying that that is a flawed system because it varies tremendously based on how many are on the grid.

Having just a simple multiplier based on grid size was not really sufficient to negate those flaws and find out the "rightful" winners of the prizes.

I like the new system because instead of being a simple multiplier on the old and flawed system, it brings a whole new, and IMO, more accurate formula to provide points for each position that depend on the grid size. It does not have a set number of points for every position, which was the main flaw in the old system for determining prizes -for example, a 2nd place in a grid of 16 was counted as the same as a 2nd in a grid of 8 in the old system, which should not be the case!

Again, my hopes would be that the new system could be implemented to be the official "published" points, but that may be too drastic a change.

Anywho, good work and well thought out. You got my support.

 

[calden592]

Hey guys! New to the thread here but i heard that six other SNAILs qualified for the national finals, congrats! (h5)

 

[brianm3]

Hi Brian, thanks for your interest. Skills will be along shortly with your official welcome post, but in the interest of time, you might want to take a peek at his previous welcome post.

Based on the rather impressive credentials in your sig, your time trial placement is likely a given, but we require all new members to run it all the same. Once you do that, start breaking in the three cars, or just post a request for some help with duping cars if there's not enough time (league night starts in five hours).

EDIT: Ninja'd by Skills!

Thank you, I appreciate it. So just read the welcome letter, ran through the rules rather quickly and just purchased some fresh cars. I'll try to put some mileage on them but time is a little short so I'll roll with it.

Thanks again for adding me, I have heard nothing but great things about this league! Really looking forward to this...

Cheers,

Brian

 

[kcheeb]

Just got home to find the break in on my Viper didn't work. Does anyone have one they could send?

I have a send available hill. Send me an fr.

 

[pyxen]

sorry D4, I'm out tonight. My wife just got back from being away for 4 days, and I'm going to couch surf with her.

If I get online because she crashes, then I may be there for the later races..

 

[Wolfsatz]

I have a send available hill. Send me an fr.

Viper cannot be sent....NVM... The SP1 cannot be sent

 

[hillracing]

Viper cannot be sent

Just got one from socalracer310. Thanks.

 

[Rallywagon]

If there is one thing I can say about SNAIL, it is not afraid of change. We have changed Just about every aspect including the meaning of the name. The only thing I can think of that has remained untouched is the spec part. We are just sure to fully deliberate any change before it takes place, and to be sure all sides have been thoroughly represented and presented.

 

[fzappa]

D2 now open.

 

[bandit0627]

This is a late announcement but I will not be making it for tonights races. My trip to the city took a little longer than expected so I won't be home for another hour and a half! Good luck everyone!