***Official*** India's Tour of Sri Lanka 2017

[Spark]

Then how do they determine the peak altitude the ball will reach, especially in the case of spinners?

the same way you calculate the peak altitude of any projectile in free motion once you have an acceptable estimate of its initial (= post pitching) velocity.

 

[srbhkshk]

the same way you calculate the peak altitude of any projectile in free motion once you have an acceptable estimate of its initial (= post pitching) velocity.

I don't think it's that simple, grounds close to Sea for instance will behave differently to Dharamshala because of the air factor.

 

[Spark]

I don't think it's that simple, grounds close to Sea for instance will behave differently to Dharamshala because of the air factor.

that effect will be marginal enough that it should be very easy to compensate for it during calibration. i assume they're not actually trying to implement a first principles simulation of drag on a cricket ball here.

edit: but even if you did, it should be pretty easy to factor in air pressure, temperature, humidity etc into whatever effective air density you're stuffing into your simulation.

 

[srbhkshk]

i assume they're not actually trying to implement a first principles simulation of drag on a cricket ball here.

Well of course not, I was basically referring to this -

that effect will be marginal enough that it should be very easy to compensate for it during calibration.

as some factor of estimated bounce.

 

[Spark]

Well of course not, I was basically referring to this -



as some factor of estimated bounce.

yeah but i assume that they just put it in by including an "effective" drag force into their calculations, i.e. some number x the velocity (etc etc). that number doesn't need to have any relation to actual physics, you can just calibrate it to whatever works on any given location. it's certainly how i would do it.

what i mean by "marginal" is that it just involves changing numbers, it doesn't break the actual model they're using to calculate trajectories.

 

[Spark]

so i actually went to the hawkeye website and checked: http://pulse-static-files.s3.amazon...ponse_to_questions_posed_on_Tendulkar_LBW.pdf

the LBW relevant part of our ball tracking is only from the bounce to where the ball hits the batman, everything pre-bounce is irrelevant. All we do is measure the flight of the ball during this time.

so there you go.

honestly i'm surprised that people only ever focus on the predictive side which, to me, seems borderline trivial, when it's the measurement side which seems so much more shaky and needs much more scrutiny imo.

 

[srbhkshk]

yeah but i assume that they just put it in by including an "effective" drag force into their calculations, i.e. some number x the velocity (etc etc). that number doesn't need to have any relation to actual physics, you can just calibrate it to whatever works on any given location. it's certainly how i would do it.

what i mean by "marginal" is that it just involves changing numbers, it doesn't break the actual model they're using to calculate trajectories.

I was visualizing it by assuming a calculation for altitude if there was no air, call it C then subtract a function of { velocity , distance , a number dependent on ground } from C. The ground dependent number is basically what i was calling bounce factor.

 

[Spark]

I was visualizing it by assuming a calculation for altitude if there was no air, call it C then subtract a function of { velocity , distance , a number dependent on ground } from C. The ground dependent number is basically what i was calling bounce factor.

i don't understand what you mean by this i'm afraid. why would you assume there's no air?

 

[srbhkshk]

the LBW relevant part of our ball tracking is only from the bounce to where the ball hits the batman, everything pre-bounce is irrelevant. All we do is measure the flight of the ball during this time.

This is very weird, so in theory if the ball bounced just before hitting the batsman, the calculations will be the same for a 4 foot tall and an 8 foot tall bowler.

 

[SillyCowCorner1]

Big win for India.

 

[srbhkshk]

i don't understand what you mean by this i'm afraid. why would you assume there's no air?

Basically, calculate the easy part first (no air - no drag), then modify it for drag.

 

[Spark]

This is very weird, so in theory if the ball bounced just before hitting the batsman, the calculations will be the same for a 4 foot tall and an 8 foot tall bowler.

okay you've misunderstood me i think: i meant that once they have a stable measurement for the post-bounce path then they can throw away the pre-pitching part of the trajectory as it's no longer relevant. if they don't have that then of course they have to guess, though in that circumstance you assume they would just estimate based on previous deliveries.

edit: and one of the few genuine hawkeye errors i can remember was because of exactly this, against phil hughes in galle in 2011. the ball basically hit him on the half-volley, but that pitch was a bunsen and it turned sharply. cameras didn't pick that up though, and hawkeye assumed it just went straight on. so you do get issues when the batsman basically gets half-volleyed and the pitch is doing weird things.

 

[Spark]

Basically, calculate the easy part first (no air - no drag), then modify it for drag.

i will tell you right now that it's way, way, way easier to do it all at once. i don't even know how you'd do it this way, unless you mean calculate free trajectory for a small interval, correct for drag, and then move onto the next interval. but this seems to me to be prone to huge errors.

 

[srbhkshk]

okay you've misunderstood me i think: i meant that once they have a stable measurement for the post-bounce path then they can throw away the pre-pitching part of the trajectory as it's no longer relevant. if they don't have that then of course they have to guess, though in that circumstance you assume they would just estimate based on previous deliveries.

Essentially I was trying to say that in the second case they will need to be using bowler specific data which I'm sure they don't always have.

Anyway, clearly there are major gaps in my knowledge about this so I won't speak much on this further.

 

[honestbharani]

Has it been proven about that one ball that hawk eye said will bounce over the stumps when the umpire gave an LBW off Jadeja? I remember Murali Kartik saying on commentary this was not Perth and thee is no way that ball would have bounced over the stumps.

 

[DriveClub]

insulting that drs implements the laws of the game, what horror.

They should have a disclaimer in the hawkeye telecast saying this prediction is based on MCC laws that dictate LBW dismissals but not an accurate prediction of the trajectory in anyway. Someone who doesn't watch cricket or who doesn't know the rules would think this is an accurate prediction.

 

[Shri]

the same way you calculate the peak altitude of any projectile in free motion once you have an acceptable estimate of its initial (= post pitching) velocity.

Posts like this make me remember your first dorky username.

 

[Spark]

Posts like this make me remember your first dorky username.

noooon, rien de rien, noooooon, je ne regrette rien...

 

[mr_mister]

lol @ Simon Doull pretending India didn't delay the declare for Kohli's 100. Saying 'oh the 550 proves that was always the number they wanted they werent concerned about milestones


we aren't idiots

 

[srbhkshk]

lol @ Simon Doull pretending India didn't delay the declare for Kohli's 100. Saying 'oh the 550 proves that was always the number they wanted they werent concerned about milestones


we aren't idiots

Commentators need to understand that people don't think of personal milestones as inexcusable evil every single time, if it has no effect on the match status, go get your hundreds or you fifers, who cares?