2023 United States Grand Prix - COTA, Oct 20 - 22

[ValeVida46]

The regulations allow them to check specific cars or specific things when they suspect something is illegal. In this case they clearly did suspect something and were right.. Is that not proof that the system works? You may not like the outcome but the proof is in the result here

They checked and failed to find anything 4 times previously.
You cannot call something a random check when the selection process is anything but random.
Because if they're admitting they use bobbling heads as a metric, it isn't a random check. I can elaborate further if this is not understood.

And what kind of proof is not checking 2 teams at all, as evidence it works?
You check something more times you have more chance to find something. It really is as simple as that.

[organic]

The regulations allow them to check specific cars or specific things when they suspect something is illegal. In this case they clearly did suspect something and were right.. Is that not proof that the system works? You may not like the outcome but the proof is in the result here

They checked and failed to find anything 4 times previously.

You cannot know that. We only heard about it when they overstepped the mark. It's quite possible that they were close to the limit on wear previously and that's why they've kept checking them. Furthermore, if you consider Max has been scrutinized twice in the region. Again Ferrari were "marginal" in the plank wear as in almost disqualified @ Austria but that didn't even make headlines apart from in smaller italian media publications. Who's to say Merc weren't similarly close to the limit at a previous race and that's why they've been scrutinized more times.

Besides, humans are not very good at looking at data and deciding how random something is. The distribution in OP isn't that unlikely - it's not like all teams apart from 2 were scrutinized at all. In fact when data is deliberately made less random we often think it appears more random.. We are not good at it so I wouldn't get hung up on this so much when so much context is lacking

Again the regulations specifically don't claim it's random. Are you not reading what i'm writing? This feels plainly like trolling. The regulations stipulate that if they see something suspicious in the data they can investigate as they see fit. If that is a car bottoming a lot getting more plank investigations, I don't see what's wrong with that?

[dans79]

I don't think FIA claim it is a totally random selection.. I think considering everything they've said below I believe they mean that the selection is arbitrary rather than truly random.

I was talking more about how Williams & AlphaTauri haven't been inspected at all this season.

At one point it was random, but then Joe could decide to add additional cars at his discretion. However 20% of the grid not getting checked at all, is pretty much proof its not even close to random now. I'm going to bet this got changed when they decided teams should scrutineer their own cars.

Imo, this is just more proof that the FIA is inept and useless!

[dans79]

Again the regulations specifically don't claim it's random. Are you not reading what i'm writing? This feels plainly like trolling. The regulations stipulate that if they see something suspicious in the data they can investigate as they see fit. If that is a car bottoming a lot getting more plank investigations, I don't see what's wrong with that?

I'll have to check and see if I still have some of my actuarial science books packed away some place tonight. however after 18 races having 20% of the grid not checked at all, has a very low probability of being random.

[ValeVida46]

You cannot know that. We only heard about it when they overstepped the mark.

Is marginal illegal? So we do know that. Especially them passing the previous tests and failing the last one.

Furthermore, if you consider Max has been scrutinized twice in the region. Again Ferrari were "marginal" in the plank wear as in almost disqualified @ Austria but that didn't even make headlines apart from in smaller italian media publications.

Besides, humans are not very good at looking at data and deciding whether or not something is random. In fact when data is deliberately made less random we often think it appears more random.. We are not good at it so I wouldn't get hung up on this so much when so much context is lacking

Context isn't lacking at all when the data spins out 20% of the grid getting 50% of the supposedly random checks.
Marginal is legal. If it wasn't they'd have been DQ'd. If suspicions are raised from previous checks it is no longer random.
What this demonstrates is a double tiered process of scrutiny that relies on previous suspicions as a pretext for future checks. 8 teams can then push the boundaries while 2 others keep getting checked.
That isn't right at all.

[TFSA]

They checked and failed to find anything 4 times previously.
You cannot call something a random check when the selection process is anything but random.
Because if they're admitting they use bobbling heads as a metric, it isn't a random check. I can elaborate further if this is not understood.

And what kind of proof is not checking 2 teams at all, as evidence it works?
You check something more times you have more chance to find something. It really is as simple as that.

We use the term word "random" for computer generated random numbers as well, despite them in reality only being pseudo-random (ei, generated by an computer algorithm, typically a Hash-function which is fed some data).

"Random" in this case means that the FIA can, at their discretion, subject different cars to checks for whatever reason they feel like.

That they 4 selected cars at the US GP was the 3 podium cars + the polesitter, and that they selected this specific race to do extra plank checks (often they don't check them at all, or maybe 1-2 cars at the most) certainly wasn't random. We can't exactly say what is behind their decision (data, gut feeling), but it was clearly driven by them suspecting something, at least with some of the cars.

What this demonstrates is a double tiered process of scrutiny that relies on previous suspicions as a pretext for future checks. 8 teams can then push the boundaries while 2 others keep getting checked.

You can't safely conclude that. If the FIA is basing their decisions on cars to check partly on data, then 8 other teams pushing the rules is gonna get figured out - maybe not immediately, but at some point.

I know the FIA sometimes presents themselves as somewhat incompetent, but they're not that stupid.

[ValeVida46]

They checked and failed to find anything 4 times previously.
You cannot call something a random check when the selection process is anything but random.
Because if they're admitting they use bobbling heads as a metric, it isn't a random check. I can elaborate further if this is not understood.

And what kind of proof is not checking 2 teams at all, as evidence it works?
You check something more times you have more chance to find something. It really is as simple as that.

We use the term word "random" for computer generated random numbers as well, despite them in reality only being pseudo-random (ei, generated by an computer algorithm, typically a Hash-function which is fed some data).

"Random" in this case means that the FIA can, at their discretion, subject different cars to checks for whatever reason they feel like.

That they 4 selected cars at the US GP was the 3 podium cars + the polesitter, and that they selected this specific race to do extra plank checks (often they don't check them at all, or maybe 1-2 cars at the most) certainly wasn't random. We can't exactly say what is behind their decision (data, gut feeling), but it was clearly driven by them suspecting something, at least with some of the cars.

Over the course of 1 event yes. I agree.

The other 4 event checks though are totally at odds with what the FIA say their selection process should be.

Instead it's "feel like" gut feeling" and "head bobbling" as metrics for inspection. Any wonder some teams have no checks while others have 5.
It's pretty damning on a governing body to be so inept.

[dans79]

from the sporting regulations.

https://www.fia.com/sites/default/files ... -10-25.pdf

31.8

The stewards will publish the findings of the scrutineers each time cars are checked during the
Competition. These results will not include any specific figure except when a car is found to be
in breach of the Technical Regulations.

[TFSA]

Over the course of 1 event yes. I agree.

The other 4 event checks though are totally at odds with what the FIA say their selection process should be.

Instead it's "feel like" gut feeling" and "head bobbling" as metrics for inspection. Any wonder some teams have no checks while others have 5.
It's pretty damning on a governing body to be so inept.

What you call inept, i call smart.

It makes perfectly reasonable sense for the FIA to use data to make smarter decisions here. If a car has indications that it is breaking the rules, it makes sense to focus scrutiny on it.

Similarly, it made sense to check the entire podium at US, even if Max and Lando didn't show signs of wearing their plank (from what we know, they ran a rather conservative ride height), just to avoid controversy.

It's generally a good idea to include some randomness, just to make sure that teams don't figure out a way to reliably avoid scrutiny and game the system. But using a data- or evidence-driven approach still makes sense.

[ValeVida46]

You can't safely conclude that. If the FIA is basing their decisions on cars to check partly on data, then 8 other teams pushing the rules is gonna get figured out - maybe not immediately, but at some point.

I know the FIA sometimes presents themselves as somewhat incompetent, but they're not that stupid.

If there is bias raised from previous races being projected into a future race, that would alleviate others from the same scrutiny. That much is clear from the numbers of inspections we've seen some teams get over others.

I mean we literally just had a driver go off track 28 times without sanction, this was picked up by a random on twitter.
I'd say the FIA at this both point are both incompetent and stupid.

[organic]

You can't safely conclude that. If the FIA is basing their decisions on cars to check partly on data, then 8 other teams pushing the rules is gonna get figured out - maybe not immediately, but at some point.

I know the FIA sometimes presents themselves as somewhat incompetent, but they're not that stupid.

If there is bias raised from previous races being projected into a future race, that would alleviate others from the same scrutiny. That much is clear from the numbers of inspections we've seen some teams get over others.

I mean we literally just had a driver go off track 28 times without sanction, this was picked up by a random on twitter.
I'd say the FIA at this both point are both incompetent and stupid.

No doubt, but the track limit thing is a separate problem and sporting issue. I believe others in this forum like Juzh have shown countless examples of track limits not being picked up by the FIA that would've changed results. What about the Hamilton track extending in 2021 Bahrain? They've never been good at it - not new

The planks and scrutineering is a technical reg issue which is much clearer and has been regulated consistently - Ferrari and Merc seem perfectly fine taking the DSQ on the chin

[ValeVida46]

Similarly, it made sense to check the entire podium at US, even if Max and Lando didn't show signs of wearing their plank (from what we know, they ran a rather conservative ride height), just to avoid controversy.

They could very well have been very marginal. Maybe this means they get another random check this weekend....

[AR3-GP]

Italian media now concludes that McLaren and RB sacrificed over 1 tenth in terms of performance to ensure plank legality and that "Hamilton's plank was in worse shape than Leclerc"

Well it's no wonder that there were no leaks on the amount of plank wear on the Merc, whereas Ferrari admitted to "tenths of a millimeter" if that can be believed. I figured as much that was the reason...and of course AMUS didn't bother getting the inside scoop either...

[AR3-GP]

Merc and Ferrari are notorious for plank scraping noises in their F1TV onboards this year. RB is much quieter. This could be part of why Merc and Ferrari are so frequently under suspicion.

[ValeVida46]

Maybe the FIA need to change it from "random checks" to "suspected infringement checks".

If English is to be respected of course.

Instead, it says its long-standing protocol of randomly checking various parts from cars has long worked because teams never know what components are being looked at each race - so they can't risk trying to get around the rules