Fia and plank tests in 2023

[ThijsMuis]

So in actuality what we have here is the following: A 100% hit rate on cars flagged by censor data. And a 100% miss rate on cars selected as controls to verify the accuracy of their data. In other words, the system worked perfectly. This entire story has been a lesson in the media being lured into a "controversy" by delusional fans

Trying to find any links to data.
Has it been released?

[Cs98]

So in actuality what we have here is the following: A 100% hit rate on cars flagged by censor data. And a 100% miss rate on cars selected as controls to verify the accuracy of their data. In other words, the system worked perfectly. This entire story has been a lesson in the media being lured into a "controversy" by delusional fans

Trying to find any links to data.
Has it been released?

Were you expecting diagrams? The reporting was on Sky. Essentially 16 and 44 were flagged by the oscillation meter installed last season for porpoising (ironic), 4 and 1 were selected as controls.

[dialtone]

https://twitter.com/dr_obbs/status/1717967614079619268

So much for the "random" checks if what Sky says is true.

In that case, it's what I've always said: The FIA actually knew what they are doing. And there is no need to check other cars, including Russell and Sainz.

I do wish the FIA was more open on this here though. That could have avoided more controversy, if they had just admitted that they might be using data to support their choice of cars to scrutineer.

That’s not how it works unless you are omniscient and know the only correlated data possible is the onboard accelerometer.

This is the first time since 1994 that cars are dsq for plank wear. Just because of this you should have to check them all.

If not, that’s some awful confidence, not backed up by much, and frankly they missed a golden opportunity to validate their accelerometer hypothesis properly as picking those 2 cars that lifted themselves plenty high added confounding variables more than resolve doubts.

I have no clue why this is even a discussion point… Even Max made a reasonable argument but some RBR supporters here just need to be right.

[ThijsMuis]

So in actuality what we have here is the following: A 100% hit rate on cars flagged by censor data. And a 100% miss rate on cars selected as controls to verify the accuracy of their data. In other words, the system worked perfectly. This entire story has been a lesson in the media being lured into a "controversy" by delusional fans

Trying to find any links to data.
Has it been released?

Were you expecting diagrams? The reporting was on Sky. Essentially 16 and 44 were flagged by the oscillation meter installed last season for porpoising (ironic), 4 and 1 were selected as controls.

Sorry was asking there was data included because I can't find it.

[Cs98]

https://twitter.com/dr_obbs/status/1717967614079619268

So much for the "random" checks if what Sky says is true.

In that case, it's what I've always said: The FIA actually knew what they are doing. And there is no need to check other cars, including Russell and Sainz.

I do wish the FIA was more open on this here though. That could have avoided more controversy, if they had just admitted that they might be using data to support their choice of cars to scrutineer.

That’s not how it works unless you are omniscient and know the only correlated data possible is the onboard accelerometer.

This is the first time since 1994 that cars are dsq for plank wear. Just because of this you should have to check them all.

If not, that’s some awful confidence, not backed up by much, and frankly they missed a golden opportunity to validate their accelerometer hypothesis properly as picking those 2 cars that lifted themselves plenty high added confounding variables more than resolve doubts.

I have no clue why this is even a discussion point… Even Max made a reasonable argument but some RBR supporters here just need to be right.

Is this the new spin now after a week of running with "50% of the cars checked got DQd"? 100% of the cars flagged got DQd, 100% of the controls did not get DQd. We had a working system all along, shame salt got in the way of recognizing that.

[TFSA]

That’s not how it works unless you are omniscient and know the only correlated data possible is the onboard accelerometer.

This is the first time since 1994 that cars are dsq for plank wear. Just because of this you should have to check them all.

If not, that’s some awful confidence, not backed up by much, and frankly they missed a golden opportunity to validate their accelerometer hypothesis properly as picking those 2 cars that lifted themselves plenty high added confounding variables more than resolve doubts.

That argument assumes that the FIA are completely new to this.

The sensors have been installed since halfway through 2022 (if i recall), and the FIA have been checking planks all that time. They have had plenty of races to correlate the sensor data with their plank checks, and therefore very likely know how to interpret those data at this point.

They flagged on two cars, picked two cars for control, and very likely got confirmation that their method is effective.

I have no clue why this is even a discussion point… Even Max made a reasonable argument but some RBR supporters here just need to be right.

That makes zero sense. Red Bull fans would be arguing for more checks, since the most likely culprits for more disqualifications were the other Ferrari and the other Merc - at least according to the logic of most people here (and Max).

Sounds more like the pot calling the kettle black. What you interpret as Red Bull fans sound like more "Not Mercedes fans" to me, because Mercedes/HAM fans are the ones who seem to be most vocal on this (not just on this forum, but I've been reading around, including the cesspool that is X/Twitter), because they feel like some great injustice has been done here against Hamilton again. But not being a Mercedes fan doesn't turn one into a Red Bull fan.

Max is as much in the dark as us on this btw. He doesn't know the considerations the FIA and Jo Bauer takes on these checks. In reality, we don't really know how the FIA approaches this beyond the sparse public information we have available.

But from what we know, their methodology works. So like you said, i have no clue why this is even a discussion, now that we more or less know that the checks are being done based on actual data.

[dans79]

But from what we know, their methodology works. So like you said, i have no clue why this is even a discussion, now that we more or less know that the checks are being done based on actual data.

The simple answer is because just like so many other things with the FIA, it works right up till someone figures out how to game it. Flexy wings several times over the last 2 decades, the fuel flow meter, the technical directives we had this year and last about about the floor/plank, etc etc.

[TFSA]

But from what we know, their methodology works. So like you said, i have no clue why this is even a discussion, now that we more or less know that the checks are being done based on actual data.

The simple answer is because just like so many other things with the FIA, it works right up till someone figures out how to game it. Flexy wings several times over the last 2 decades, the fuel flow meter, the technical directives we had this year and last about about the floor/plank, etc etc.

The entire idea of behind those loopholes that you mention was that they were NOT picked up during scruteneering, and the cars passed all tests.

Those things were figured out and clamped down on, not by any scruteneering, but by the FIA (or sometimes other teams, who then told the FIA) suspecting something was up, and then dishing out Technical Directives or conducting an investigation.

So you're not making an argument for more extensive scruteneering here.

Edit: Also, just because the FIA can use a data driven approach doesn't mean they have to always. They can conduct random checks that really are random.

[dialtone]

That’s not how it works unless you are omniscient and know the only correlated data possible is the onboard accelerometer.

This is the first time since 1994 that cars are dsq for plank wear. Just because of this you should have to check them all.

If not, that’s some awful confidence, not backed up by much, and frankly they missed a golden opportunity to validate their accelerometer hypothesis properly as picking those 2 cars that lifted themselves plenty high added confounding variables more than resolve doubts.

That argument assumes that the FIA are completely new to this.

The sensors have been installed since halfway through 2022 (if i recall), and the FIA have been checking planks all that time. They have had plenty of races to correlate the sensor data with their plank checks, and therefore very likely know how to interpret those data at this point.

They flagged on two cars, picked two cars for control, and very likely got confirmation that their method is effective.

Unfortunately no, even from an extremely basic statistical standpoint, I would say elementary as in even a kid would do experiments this way, you would want to run a very simple A/B test for your theory, which usually means you keep most of the variables identical except your control between the 2 tests.

What does it mean in practice?
In this case if they wanted to confirm their theory they would, at minimum, test the other car of the same team, which presumably they didn't see in the accelerometer, or they would have tested them as well, and would have provided a vastly better check for their hypothesis.

FIA is absolutely new to this, accelerometers haven't existed in the cars for this use prior to TD39 and there have been exactly 0 floor wear violations since then. It doesn't get any newer than this.

I have no clue why this is even a discussion point… Even Max made a reasonable argument but some RBR supporters here just need to be right.

That makes zero sense. Red Bull fans would be arguing for more checks, since the most likely culprits for more disqualifications were the other Ferrari and the other Merc - at least according to the logic of most people here (and Max).

Sounds more like the pot calling the kettle black. What you interpret as Red Bull fans sound like more "Not Mercedes fans" to me, because Mercedes/HAM fans are the ones who seem to be most vocal on this (not just on this forum, but I've been reading around, including the cesspool that is X/Twitter), because they feel like some great injustice has been done here against Hamilton again. But not being a Mercedes fan doesn't turn one into a Red Bull fan.

Max is as much in the dark as us on this btw. He doesn't know the considerations the FIA and Jo Bauer takes on these checks. In reality, we don't really know how the FIA approaches this beyond the sparse public information we have available.

But from what we know, their methodology works. So like you said, i have no clue why this is even a discussion, now that we more or less know that the checks are being done based on actual data.

I'm not an Hamilton fan nor a Mercedes fan, I'm a Ferrari fan for over 30 years. I don't like bad procedures, inconsistencies and so on. I didn't complain about LEC's DSQ, if his car was off it was off, the procedure followed by FIA for everything else is totally utterly laughable.

[dans79]

The entire idea of behind those loopholes that you mention was that they were NOT picked up during scruteneering, and the cars passed all tests.

Those things were figured out and clamped down on, not by any scruteneering, but by the FIA (or sometimes other teams, who then told the FIA) suspecting something was up, and then dishing out Technical Directives or conducting an investigation.

So you're not making an argument for more extensive scruteneering here.

The fuel flow sensor is a perfect example imo. The FIA was so sure it couldn't be gamed, thats it took a couple of teams well over a year to get the FIA to look into it. If they relied more on manual checking they might have caught it a lot sooner!

[TFSA]

The fuel flow sensor is a perfect example imo. The FIA was so sure it couldn't be gamed, thats it took a couple of teams well over a year to get the FIA to look into it. If they relied more on manual checking they might have caught it a lot sooner!

Manual checking of what? There has to be something to check. Is there any scruteneering check you know of that measures fuel flow beyond the sensor (which Ferrari duly cheated)?

You're doing pure speculation here. Too much "Might have".

[dans79]

The fuel flow sensor is a perfect example imo. The FIA was so sure it couldn't be gamed, thats it took a couple of teams well over a year to get the FIA to look into it. If they relied more on manual checking they might have caught it a lot sooner!

Manual checking of what? There has to be something to check. Is there any scruteneering check you know of that measures fuel flow beyond the sensor (which Ferrari duly cheated)?

You're doing pure speculation here. Too much "Might have".

https://www.formula1.com/en/latest/arti ... qoS0G.html

[TFSA]

Unfortunately no, even from an extremely basic statistical standpoint, I would say elementary as in even a kid would do experiments this way, you would want to run a very simple A/B test for your theory, which usually means you keep most of the variables identical except your control between the 2 tests.

What does it mean in practice?
In this case if they wanted to confirm their theory they would, at minimum, test the other car of the same team, which presumably they didn't see in the accelerometer, or they would have tested them as well, and would have provided a vastly better check for their hypothesis.

FIA is absolutely new to this, accelerometers haven't existed in the cars for this use prior to TD39 and there have been exactly 0 floor wear violations since then. It doesn't get any newer than this.

Unfortunately you cannot conclude that, because you haven't seen the data they have.

If the data they have show a very clear correlation between the sensor and the wear from the plank checks they have done, with very few (or maybe no) outliers, then you don't need many samples. The amount of outliers matter. If pretty much all their samples is close to the mean value, then that's a high confidence value, even with the samples they have at this point.

I'm not an Hamilton fan nor a Mercedes fan, I'm a Ferrari fan for over 30 years. I don't like bad procedures, inconsistencies and so on. I didn't complain about LEC's DSQ, if his car was off it was off, the procedure followed by FIA for everything else is totally utterly laughable.

Only problem with that is that you have no proof that this procedure is bad. That's entirely based on the idea that because 50% of the cars failed a "random test", which is erroneous when the test turns out not to be random, but data driven.

[TFSA]

https://www.formula1.com/en/latest/arti ... qoS0G.html

That's not a post-race scruteneering check they implemented. That's a pre-race procedure that strengtens the FIAs knowledge.

Ironically, it's also data based, just like these oscillation sensors are, so the FIA have data to make sure they know they are doing the correct checks.
So what's your point again?

[dialtone]

Unfortunately no, even from an extremely basic statistical standpoint, I would say elementary as in even a kid would do experiments this way, you would want to run a very simple A/B test for your theory, which usually means you keep most of the variables identical except your control between the 2 tests.

What does it mean in practice?
In this case if they wanted to confirm their theory they would, at minimum, test the other car of the same team, which presumably they didn't see in the accelerometer, or they would have tested them as well, and would have provided a vastly better check for their hypothesis.

FIA is absolutely new to this, accelerometers haven't existed in the cars for this use prior to TD39 and there have been exactly 0 floor wear violations since then. It doesn't get any newer than this.

Unfortunately you cannot conclude that, because you haven't seen the data they have.

I absolutely can. RBR and MCL are 2 different cars than Mercedes and Ferrari, setup differently thus introducing a variable than even at equal accelerometer values could lead to different results. And obviously a good large chunk of the paddock shares the same opinion, including Max. So yeah, I can conclude that and I will.

If the data they have show a very clear correlation between the sensor and the wear from the plank checks they have done, with very few (or maybe no) outliers, then you don't need many samples. The amount of outliers matter. If pretty much all their samples is close to the mean value, then that's a high confidence value, even with the samples they have at this point.

They never tested the same cars, they always tested only 1 car per manufacturer, you cannot eliminate setup or other variables. Once again, almost everyone in the paddock thinks that testing only 1 car and not the team mate car was nonsense. I, in addition, think that not testing all cars was nonsense.

I'm not an Hamilton fan nor a Mercedes fan, I'm a Ferrari fan for over 30 years. I don't like bad procedures, inconsistencies and so on. I didn't complain about LEC's DSQ, if his car was off it was off, the procedure followed by FIA for everything else is totally utterly laughable.

Only problem with that is that you have no proof that this procedure is bad. That's entirely based on the idea that because 50% of the cars failed a "random test" (which was not random at all), which is erroneous.

It may surprise you but I don't need proof, I just need to point out the experiment was bad and I've already done it. A proper experiment has 1 control variable. They NEVER did it.