One for the statisticians

[Steve Fox]

One for the statisticians, or at least those with a mathematical bent AND too much time on their hands.

Here is a theoretical situation for your cerebration. Car comes steaming into the gravel trap, Driver jumps out, runs across the gravel and hops over the tyre wall.
Our orange clad hero has just completed the 100 yard dash, says to the driver, you alright mate, what's happened, have you switched off the electrics? No, says the driver, Should I do it now?

So: there is a danger of fire leaving the electrical system powered if another car hits it but there is also a danger of the driver being hit in the gravel trap by another car during the trip back to and from the car, so what is the critical distance that a driver should be asked (or allowed) to return to his car? Assuming that he has gone off on his own, clearly if the course is slippery for some reason or if there are any external reasons that would increase the likelihood of another car following him in then the scenario obviously changes, In both cases (leave it but don't let go of a bottle or let him go back) there is danger and I know that he should have switched the juice off before legging it but that isn't the question, but at some statistical point (not zero) one danger is equal to the other.
At the end of the day a car on fire can be replaced by a less burnt one but the real danger here is another car hitting the stranded one and the collision causing a fire involving a second driver, so it isn't quite so obvious as let the car burn rather than let a driver wander through the gravel, both cases involve danger to life.
My own guess would have the distance at no more that one width of the car.
Any thoughts?

[Dave Brand]

The driver should not be asked to return to his car to switch the electrics off - that's your job! Can you do it safely? If you can't, don't do it.

If the car's in a dangerous position the decision is race stop, move/snatch under safety car or leave it where it is . . . that's down to knowledge & experience, not statistics.

The great thing about incident handling is that every incident is different & you can't reduce it to a set of rules.

[Woolley]

You can't prescribe it to a set distance, depends entirely on the corner, the position of the car, what type of race it is and how long it's been going. Ideally you would have attracted his attention from the bank early enough to get him to switch it off before he left it or immediately afterwards. If safety is questionable, then the leave it switched on, make a mental note and then switch it off when you go to move it, be that at the end of the race, under safety car or under live conditions depending on what decision is made subsequently. I'd only actually send the driver back if a) he's within a few paces, or b) he's next to the wall.

Sorry, Dave, I disagree - switching the car off is only our job if the driver's had sufficient of a ding for us to need to do it for him. Since the initial post suggests we're dealing with a car in an iffy position then there's no reason for the marshal to be over the wall at all if the driver is capable of moving and his car isn't, unless there's a need to move it.

[fazza]

There is no real answer to this as every incident has to be evaluated on its own risks.

The overriding rule is
Your safety first
Your partners
And driver the car is replaceable
If the situation is likely to escalate if another car piles in then the information is passed to race control and the clerks make the call.

[numbersix]

For a PC or I/O it's a simple risk assessment.

What are the increased chances of the car catching fire because the isolator switch is still active - low, as is the chance of the car catching fire after a random hit. And this is multiplied by the consequences if it does actually happen - and those consequences are not life threatening.

Therefore, to go trackside just for the switch would seem unnecessary, but this is part of a wider scenario, such as whether the car is in a dangerous position, how much of the race is left to run, etc.

If it's the first lap or two, with nearly a two minute gap before the pack comes around again, then a quick wander out to turn it off would be fine. Two laps from the end? - No.

An incident marshal would look to their I/O for instruction. The I/O might do the same from the PC. The PC might encourage or discourage the action depending on the circumstances.

There is no formula. Just experience, and good communication.

[The Fat Clerk]

Originally Posted by numbersix

For a PC or I/O it's a simple risk assessment.

What are the increased chances of the car catching fire because the isolator switch is still active - low, as is the chance of the car catching fire after a random hit. And this is multiplied by the consequences if it does actually happen - and those consequences are not life threatening.

Therefore, to go trackside just for the switch would seem unnecessary, but this is part of a wider scenario, such as whether the car is in a dangerous position, how much of the race is left to run, etc.

If it's the first lap or two, with nearly a two minute gap before the pack comes around again, then a quick wander out to turn it off would be fine. Two laps from the end? - No.

An incident marshal would look to their I/O for instruction. The I/O might do the same from the PC. The PC might encourage or discourage the action depending on the circumstances.

There is no formula. Just experience, and good communication.

Wise words

[Woolley]

Originally Posted by The Fat Clerk

Wise words

Quite. Roughly what I was trying to say, but so much better worded than my attempt.

[Steve Fox]

So it is pretty much as it looks then, Unless there is a very clear opportunity to "pop" back to the car or if the race is very long indeed then the only sensible course of action is to keep a bottle ready and make sure that those who handle the car (if not the Marshal in question) know that the power is still on or to take the appropriate action should the car be involved in further incident.
I'm guessing that it would be straightforward enough to put probability figures to the situation but the list of variables will be too long for any meaningful use.

[Woolley]

Originally Posted by Steve Fox

make sure that those who handle the car (if not the Marshal in question) know that the power is still on

Correct. Although hopefully all marshals will adopt the same strategy of if they're the first to the car when they do attend, then double check the electrics.

A little story might be relevent. Many years ago at Mallory I was on my own on the inside of Stebbe Straight. A Metro turbo took a big hit on my side so I went over to check the driver, switching the power off as I did so. Driver was fine and started to follow me to the bank but turned back briefly. I didn't see why, but called him on. At the end of the session we returned to the car only to find the reason he'd gone back was to turn the power back on so that the electric fan could cool the car. So the moral is, regardless of what you're told, when it's safe to go to the car, always check the electrics are off.

[Dave Brand]

Originally Posted by Steve Fox

I'm guessing that it would be straightforward enough to put probability figures to the situation but the list of variables will be too long for any meaningful use.

Hundreds of variables, most of them unquantifiable, most of them random, with a lot of interaction between multiple variables. Even if a statistically significant probability level could be calculated, how do you apply it in real life? Just for argument's sake, say you could calculate that there's a 10% probability that a certain car will spin at a certain corner. That doesn't mean there's a 100% probability that it will spin once in a ten-lap race - probability doesn't work like that.

[Adam43]

It is a 55% chance! (sorry, practically OT)

[Steve Fox]

Originally Posted by Adam43

It is a 55% chance! (sorry, practically OT)

Ah, But don't forget that a whopping 84% of statistics are made up on the spot.

[Dave Brand]

Originally Posted by Dave Brand

Hundreds of variables, most of them unquantifiable, most of them random, with a lot of interaction between multiple variables.

. . . and let's not forget the most random, most unpredictable, most unquantifiable variable of them all . . . the driver!

"Expect the unexpected" is an overused cliché, but in this case it has validity.

Originally Posted by Adam43

It is a 55% chance! (sorry, practically OT)

. . . says a random variable!

[Adam43]

[Richard Duvall]

Simple solution to the matter! New regulation required: master power switches to be turned of by a pullcord attached to the driver.

[Mike Kelly 1]

Originally Posted by Richard Duvall

Simple solution to the matter! New regulation required: master power switches to be turned of by a pullcord attached to the driver.

As in powerboat racing!

[yellowspanner]

Marshalling is an art, not a science. Safety of people first, then cars. If another car comes off and hits the offending vehicle should it have not been moved? Electrics may be the least of your worries then.

[Piglet]

It always worries me when people are looking for "rules" about how they should deal with marshalling situations.

We've had these discussions before about things like should you ever cross the track and similar issues - marshalling requires a dynamic and ongoing assessment of risk - we can't have fixed rules, we have to assess every aspect of every individual situation and consider the risks and the benefits of every action and that includes using all of your senses and experience - in a very short space of time.

We do though often have rules for those who lack experience - never cross the track, never go on the live track (new pit/startline marshals) and similar and these rules allow experience to be gained so that we get to the point where we can make our own assessment using the experience that we have gained from watching and listening to others. This does raise issues when because of perhaps dropping numbers, marshals don't get to work with other experienced marshals so can't learn from the experience of others and that's where threads like this are useful. Of course we all know those people who are never going to learn because they know it all already!

As said our own safety and those of our fellow marshals is paramount - two or more casualties is worse than 1 and it's easy to think you are being a hero whilst not realising the risk you are putting others at.

[The Fat Clerk]

Originally Posted by Piglet

It always worries me when people are looking for "rules" about how they should deal with marshalling situations.

We've had these discussions before about things like should you ever cross the track and similar issues - marshalling requires a dynamic and ongoing assessment of risk - we can't have fixed rules, we have to assess every aspect of every individual situation and consider the risks and the benefits of every action and that includes using all of your senses and experience - in a very short space of time.

We do though often have rules for those who lack experience - never cross the track, never go on the live track (new pit/startline marshals) and similar and these rules allow experience to be gained so that we get to the point where we can make our own assessment using the experience that we have gained from watching and listening to others. This does raise issues when because of perhaps dropping numbers, marshals don't get to work with other experienced marshals so can't learn from the experience of others and that's where threads like this are useful. Of course we all know those people who are never going to learn because they know it all already!

As said our own safety and those of our fellow marshals is paramount - two or more casualties is worse than 1 and it's easy to think you are being a hero whilst not realising the risk you are putting others at.

Again, wise words, as I always say when on post, do as I say not as I do.

[Steve Fox]

Hm, I wasn't really looking for either a rule or even guidance, I know that Marshaling is a subtle combination of art and science but the actual odds of another car hitting a stationary one during any selected time frame within certain conditions and that then turning into an even more serious incident with a direct relationship to the power being on on both cars is not art, it is a real number. Also the chances of a driver being hit by a car during a run to and returning from his car to turn off the power is also another real number which has a direct relationship to the time taken / distance traveled by that driver, the question, which was for the statisticians amongst us, was what are those two numbers. That is only art if you are very bad at painting.
Since the question was posed, however, I have realised that external factors would play such a large part in the answers to render it meaningless and that also the chances of the driver running about in the gravel poses quite a high risk and so would have a large (relatively speaking) number attached and the chances of a stationary car being hit by another car is also a large number but the chances of the incident being made dangerously worse by both cars having power to them is relatively small by comparison and so the answer, which will be a real number, would be very small indeed, so small as to make anything more than arms reach viable.
In short, as was noted earlier, unless there is a large gap or the race is very long then simply making those relevant parties aware is the best solution.

[Piglet]

Originally Posted by Steve Fox

In short, as was noted earlier, unless there is a large gap or the race is very long then simply making those relevant parties aware is the best solution.

It may be or it may not be depending on all of the relevant circumstances at the time. That's the point that is being made, there is no simple answer, every situation is different based on all of the factors attached to it - things like configuration of track, level of drivers, driving standards exhibited by the drivers (I've chosen to live with cars adjacent to the track at times and then deployed the safety car or gone for a stop because it's clear that the drivers are not obeying flag signals), weather, track conditions and on and on and on....so many things that vary for THAT incident.

The beauty of the human brain is that, with experience, most of us can make pretty good dynamic risk assessments in a short space of time. They may not however turn out to be correct - hindsight is a wonderful thing and **** can and will happen when you least anticipated it.

[The Fat Clerk]

That's why when asked about 'Clerking' I describe it as juggling chainsaws.

[thepits!]

Originally Posted by The Fat Clerk

That's why when asked about 'Clerking' I describe it as juggling chainsaws.

Which can be a 'armless pastime -