more technology in LBW decisions
Steve G Smith
Opinion mounts for more technology in LBW decisions
Cape Town - Cricket experts and umpires agree that television technology
should be utilised more to eliminate flagrant umpiring errors.
According to Brian Basson, the United Cricket Board's umpire director, the
pressure on test umpires has become unbearable since wrong decisions can
clearly be seen by television viewers and commentators, due to new
technology.
Former South African fast bowler, Fanie de Villiers, thinks he has the
solution to eliminate bad decisions drastically: "The third umpire should
have the power to reverse an umpire's decision if a blatant mistake was
made.
"This should ensure that no more than two or three doubtful decisions are
made in a test match, instead of the twenty in the recent match at
Headingley," De Villiers said.
He wants to launch a Cricket Supporters' Union. "If I represent a million
people, the United Cricket Board will have to pay attention if I come up
with recommendations that are advantageous to players and umpires," De
Villiers said.
In the meantime, researcher Prof. Tim Noakes of Cape Town's Sport Institute,
said he and his colleagues can develop computer technology which would
accurately determine whether a player was trapped leg before wicket.
He said they needed six ultra rapid cameras, at a cost of R1,5 million, that
could photograph 2 000 frames per second and which would be able to
determine the trajectory of the ball.
Graham
Steve G Smith wrote in message <35d17...@job.acay.com.au>...
>August 11, 1998
>Opinion mounts for more technology in LBW decisions
>
>Section snipped
Perhaps there is an answer that eliminates the need for technology and will
also elimintae the poor decisions? (Purists should not read this)
The rules should be looked at closely :-
- Perhaps LBW should be eliminated altogether or perhaps the batsman should
be allowed more than 1 LBW before being given out - sort of like a warning
card in soccer.
- To compensate, perhaps the stumps should be wider so that the batsman has
less chance of blocking the ball.
-The 1-day pitch width should be adopted for all cricket to make the bowlers
bowl on-line.
A question - Have there always been 3 stumps at the current width? I seem
to recall that the current configuration developed over the years.
Graham
Mike Hesse
>He said they needed six ultra rapid cameras, at a cost of R1,5 million, that
>could photograph 2 000 frames per second and which would be able to
>determine the trajectory of the ball.
Hardly a practical alternative.
Mike Hesse
Email supplied on request
Phil. G. Felton
> On Wed, 12 Aug 1998 21:36:21 +1000, "Steve G Smith" <sm...@cci.net> wrote:
>
> >He said they needed six ultra rapid cameras, at a cost of R1,5 million, that
> >could photograph 2 000 frames per second and which would be able to
> >determine the trajectory of the ball.
>
> Hardly a practical alternative.
>
This is an area in which I do research and my proposed system was similar to
Noakes'. I think it could be done somewhat cheaper than that although I forget
what the $/R rate is. I think that only 2 cameras would be needed at each end
and they'd need to be mounted high up so as to prevent obstruction by fielders.
I agree that high speed cameras would be needed, broadcast standards are not
adequate. Bad light might cause some problems also. It's certainly doable,
I think a suitable system could be developed in a year and then given an
extensive evaluation. It still wouldn't solve all the problems though, a
ball hitting the pads on the full might pose problems.
Phil.
v...@my-dejanews.com
fel...@princeton.edu (Phil. G. Felton) wrote:
> > >could photograph 2 000 frames per second and which would be able to
> > >determine the trajectory of the ball.
>
> Noakes'. I think it could be done somewhat cheaper than that although I forget
> what the $/R rate is. I think that only 2 cameras would be needed at each end
> and they'd need to be mounted high up so as to prevent obstruction by
fielders.
I would be interested to know how well understood the motion of a cricket ball
is understood. Just how reliably do you think you could predict the trajectory
of the ball in all circumstances.
Cheers
Vince Freeman
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
Mike Holmans
decided to opine:
>On Wed, 12 Aug 1998 21:36:21 +1000, "Steve G Smith" <sm...@cci.net> wrote:
>
>>He said they needed six ultra rapid cameras, at a cost of R1,5 million, that
>>could photograph 2 000 frames per second and which would be able to
>>determine the trajectory of the ball.
>
>Hardly a practical alternative.
Seen from here, where that money translates to UKP150K, or Australia,
where it would be AUD400K, it doesn't sound an awful lot, as long as
you only need one set of six cameras which you can tote from ground to
ground.
After all, it's in the same ball-park as the team win bonuses.
However, if we're going to go down that road for Test cricket, how can
we ensure that it's universally available? There was some worrying
evidence in the spring that the lack of money in WI cricket is leading
to a severe deterioration in facilities, and the logistical
difficulties of toting expensive equipment around and calibrating it
when there's only a day or so between Tests seem horrendous.
Cheers,
Mike
Stephen A Devaux
writes:
>On Wed, 12 Aug 1998 21:36:21 +1000, "Steve G Smith" <sm...@cci.net> wrote:
>
>>He said they needed six ultra rapid cameras, at a cost of R1,5 million, that
>>could photograph 2 000 frames per second and which would be able to
>>determine the trajectory of the ball.
>
>Hardly a practical alternative.
Right. But if it WERE to become practical (i.e., economical and quick) would
it be GOOD? I think, yes. Fewer "unfair" dismissals (and unfair
non-dismissals) would make cricket a better game, and if technology can do it,
I'm for it.
Also, there was what I thought was a good suggestion (I forget by whom)
regarding caught out decisions that the batsman does not like (edges, above the
gloves, bump balls, etc.). My adapted version would be this:
If the batsman thinks he's not out after the umpire has given him out, he may
appeal to the third umpire, who will rule based on the video. If the
video is uncertain, the batsman is allowed to continue batting. If the
video shows that the batsman was clearly out, the batsman is penalized
by:
1. A SIGNIFICANT fine.
2. Suspension from the rest of the match (replacement by an emergency
fielder permitted.)
3. Suspension from the team's next equivalent international match
(Test or ODI).
If these sanctions prove insuffiently Draconian to stop frivolous
appeals, they could be increased to include larger fines and NO
replacement by an emergency fielder.
Fraternally in cricket,
Steve the Bajan
Neil Selby
> Right. But if it WERE to become practical (i.e., economical and quick) would
> it be GOOD? I think, yes. Fewer "unfair" dismissals (and unfair
> non-dismissals) would make cricket a better game, and if technology can do it,
> I'm for it.
> Also, there was what I thought was a good suggestion (I forget by whom)
> regarding caught out decisions that the batsman does not like (edges, above the
> gloves, bump balls, etc.). My adapted version would be this:
> If the batsman thinks he's not out after the umpire has given him out, he may
> appeal to the third umpire, who will rule based on the video. If the
> video is uncertain, the batsman is allowed to continue batting.
Problem is, there's probably at least one 'caught behind' in every
game which is clearly out but the TV replay can't see.
If the
> video shows that the batsman was clearly out, the batsman is penalized
> by:
> 1. A SIGNIFICANT fine.
> 2. Suspension from the rest of the match (replacement by an emergency
> fielder permitted.)
> 3. Suspension from the team's next equivalent international match
> (Test or ODI).
Which supposes the batsman *always* knows when he's nicked it. Some
people say they always feel it, some don't...
Neil
Jackie Hewitt
>
> Meanwhile prof John Noakes of Wood Lane London said he could do the same using
> 6 washing-up bottles, a tin of dog food and a roll of sticky-back plastic.
>
Ah- some one of my generation who remembers the heady days when john
Noakes was on Blue Peter. Used to do all sorts of dare devil things,
then look absolutely terrified when baking a cake.
--
Jackie Hewitt
Courtney Walsh's Biggest Fan
See my tribute to Courtney at
http://users.ox.ac.uk/~jackieh/courtney_walsh.html
Mike Hesse
>This is an area in which I do research and my proposed system was similar to
>Noakes'. I think it could be done somewhat cheaper than that although I forget
>what the $/R rate is.
R10.50 buys £1.00
R6.40 buys us$1.00
> I think that only 2 cameras would be needed at each end
>and they'd need to be mounted high up so as to prevent obstruction by fielders.
>I agree that high speed cameras would be needed, broadcast standards are not
>adequate. Bad light might cause some problems also. It's certainly doable,
>I think a suitable system could be developed in a year and then given an
>extensive evaluation. It still wouldn't solve all the problems though, a
>ball hitting the pads on the full might pose problems.
How would the cameras determine the length that still has to be travelled until
the wickets are hit?
However, the technology must be affordable at all test venues. Some still do
not have pan eye cameras for run outs. (SA tour to Aus in Dec).
Linda
Jackie Hewitt <jackie...@oucs.ox.ac.uk> wrote in article
<35D300...@oucs.ox.ac.uk>...
> David Shepherd wrote:
> >
> > Meanwhile prof John Noakes of Wood Lane London said he could do the
same using
> > 6 washing-up bottles, a tin of dog food and a roll of sticky-back
plastic.
> >
> Ah- some one of my generation who remembers the heady days when john
> Noakes was on Blue Peter. Used to do all sorts of dare devil things,
> then look absolutely terrified when baking a cake.
Well hell. I never thought I had anything in common with John Noakes, but I
live and learn.
Linda
Phil. G. Felton
What the cameras do is give you a 2-d side view of the balls path, with
another view from another angle you have enough information to determine
the ball's
position in space. The cameras would be accurately positioned before hand
and calibrated with a target positioned in a known position wrt the stumps.
One of the reasons you need a fast camera is the distance moved by the
ball per frame, a fast bowler bowls at about 40 m/s so at 60 fps the ball
travels
2/3 of a meter, whereas at 1000fps it moves 40mm per frame which is more
reasonable to work with. The cameras would need to be accurately synched with
each other which is fairly standard technology and the images sent to a computer
for analysis. 100 frames would cover the last 4m of the trajectory which should
be enough.
Phil.
Phil. G. Felton
> On Wed, 12 Aug 1998 19:14:06 -0400, fel...@princeton.edu (Phil. G.
Felton) wrote:
>
> >This is an area in which I do research and my proposed system was similar to
> >Noakes'. I think it could be done somewhat cheaper than that although I
forget
> >what the $/R rate is.
>
> R10.50 buys £1.00
> R6.40 buys us$1.00
>
> > I think that only 2 cameras would be needed at each end
> >and they'd need to be mounted high up so as to prevent obstruction by
fielders.
> >I agree that high speed cameras would be needed, broadcast standards are not
> >adequate. Bad light might cause some problems also. It's certainly doable,
> >I think a suitable system could be developed in a year and then given an
> >extensive evaluation. It still wouldn't solve all the problems though, a
> >ball hitting the pads on the full might pose problems.
>
> How would the cameras determine the length that still has to be
travelled until
> the wickets are hit?
>
> However, the technology must be affordable at all test venues. Some still do
> not have pan eye cameras for run outs. (SA tour to Aus in Dec).
PS to my earlier response you could have a travelling system, every ground
wouldn't need to have one. Rather like automatic timing at some track meets
which is set up by a contractor for individual events rather than being a
permanent fixture. As stated above it won't be cheap!
Phil.
Simon Thompson
wrote:
> What the cameras do is give you a 2-d side view of the balls path, with
> another view from another angle you have enough information to determine
> the ball's
> position in space. The cameras would be accurately positioned before hand
> and calibrated with a target positioned in a known position wrt the stumps.
> One of the reasons you need a fast camera is the distance moved by the
> ball per frame, a fast bowler bowls at about 40 m/s so at 60 fps the ball
> travels
> 2/3 of a meter, whereas at 1000fps it moves 40mm per frame which is more
> reasonable to work with. The cameras would need to be accurately synched with
> each other which is fairly standard technology and the images sent to a computer
> for analysis. 100 frames would cover the last 4m of the trajectory which should
> be enough.
What about some of the contentious decisions where the ball hit bat
before pad, thith the bat adjacent to the pad. The ball may not travel
far enough for the above system to realise it's trajectory had
changed, and the pad would obscure at least one camera. This system
would have said out, LBW, as the ball was on path to the stumps.
You would, therefore, need a manual review as well.
All very time-consuming.
-----------------------------------------
Simon Thompson
Christchurch
New Zealand
Phil. G. Felton
(Simon Thompson) wrote:
Such a system will not work in all cases as you point out, there will always
be a need to manually review, for example in the case of a ball pitching
outside the off stump the umpire will still need to decide if a shot was
played. I consider this an advantage, I view the automated system as an
accurate aid to the umpire not a replacement. The siting of the cameras
should minimize the possiblity of the cameras' view being obscured. The
system would have a provision to review the frames by the third Ump. Of
course if the ball hits bat & pad nearly simultaneously that's probably
a good case for giving the batsman the benefit of the doubt.
To reiterate the umpire would make the decision with input from the computer
I envisage the following info: 'ball pitched 1" inside the line of offstump,
probability of striking the wicket 95%, struck pad in front of middle stump.'
Phil.
Drewy
<felton-1208...@pgfelton97.princeton.edu>...
> In article <35d7ff8b...@jhb-news.iafrica.com>, Mike Hesse wrote:
>
> > On Wed, 12 Aug 1998 21:36:21 +1000, "Steve G Smith" <sm...@cci.net>
wrote:
> >
> > >He said they needed six ultra rapid cameras, at a cost of R1,5
million, that
> > >could photograph 2 000 frames per second and which would be able to
> > >determine the trajectory of the ball.
> >
> > Hardly a practical alternative.
> >
>
to
> Noakes'. I think it could be done somewhat cheaper than that although I
forget
each end
> and they'd need to be mounted high up so as to prevent obstruction by
fielders.
> I agree that high speed cameras would be needed, broadcast standards are
not
> adequate. Bad light might cause some problems also. It's certainly
doable,
> I think a suitable system could be developed in a year and then given an
> extensive evaluation. It still wouldn't solve all the problems though, a
> ball hitting the pads on the full might pose problems.
>
technology that could interpret the current LBW law using one, two or a
thousand cameras! The number of variables that need to be assessed (point
of pitching, point of contact, object first contacted and DEFINITE future
trajectory including height) make this a mammoth task. I would be
interested in any inside knowledge you might wish to share. Like how 10
cameras can DEFINITIVELY say that this ball WOULD NOT have bounced over the
stumps but THAT ONE would have.
NOT possible!
Drewy
Drewy
<6qupqq$1...@dfw-ixnews7.ix.netcom.com>...
wrote:
> >
> >>He said they needed six ultra rapid cameras, at a cost of R1,5 million,
that
> >>could photograph 2 000 frames per second and which would be able to
> >>determine the trajectory of the ball.
> >
> >Hardly a practical alternative.
>
would
> it be GOOD? I think, yes. Fewer "unfair" dismissals (and unfair
> non-dismissals) would make cricket a better game, and if technology can
do it,
> I'm for it.
>
> Also, there was what I thought was a good suggestion (I forget by whom)
> regarding caught out decisions that the batsman does not like (edges,
above the
> gloves, bump balls, etc.). My adapted version would be this:
>
> If the batsman thinks he's not out after the umpire has given him out, he
may
> appeal to the third umpire, who will rule based on the video. If the
> video shows that the batsman was clearly out, the batsman is penalized
> by:
>
> 1. A SIGNIFICANT fine.
> 2. Suspension from the rest of the match (replacement by an emergency
> fielder permitted.)
> 3. Suspension from the team's next equivalent international match
> (Test or ODI).
>
> appeals, they could be increased to include larger fines and NO
> replacement by an emergency fielder.
>
you would rather watch a game under this proposal than under the current
system?? Really??
Drewy
Stephen Devaux
>
> Stephen A Devaux <ap...@ix.netcom.com> wrote in article
> <6qupqq$1...@dfw-ixnews7.ix.netcom.com>...
> > My adapted version would be this:
> >
> > If the batsman thinks he's not out after the umpire has given him out, he
> may
> > appeal to the third umpire, who will rule based on the video. If the
> > video is uncertain, the batsman is allowed to continue batting. If the
> > video shows that the batsman was clearly out, the batsman is penalized
> > by:
> >
> > 1. A SIGNIFICANT fine.
> > 2. Suspension from the rest of the match (replacement by an emergency
> > fielder permitted.)
> > 3. Suspension from the team's next equivalent international match
> > (Test or ODI).
> >
> > If these sanctions prove insuffiently Draconian to stop frivolous
> > appeals, they could be increased to include larger fines and NO
> > replacement by an emergency fielder.
> >
>
> I almost never disagree with you, Steve. But can you HONESTLY tell me that
> you would rather watch a game under this proposal than under the current
> system?? Really??
>
> Drewy
Drewy, I would RATHER not have to. But I think that the gamesmanship
may be reaching an unfortunate level. The technology now exists to
correct a lot of bad calls (which, I agree, have ALWAYS gone on), and
the gamesmanship is on the rise. I would rather use the technology in
the way suggested above, to catch and discredit, and ultimately deter,
the cheats and/or illegitimate whiners, than to have the present number,
or greater, of demonstrably incorrect decisions.
A situation in which technology can so discredit umpire decisions as to
seem to make them arbitrary would be DISASTROUS for the game.
Ultimately, in such a context, the ability to appeal confidently would
threaten to surpass batting or bowling well as a valuable cricketing
skill. Could acting lessons for teams be far behind?
Ramaswamy
> Sorry Phil but I don't believe we are within a million miles of the
>technology that could interpret the current LBW law using one, two or a
>thousand cameras! The number of variables that need to be assessed (point
>of pitching, point of contact, object first contacted and DEFINITE future
>trajectory including height) make this a mammoth task. I would be
>interested in any inside knowledge you might wish to share. Like how 10
>cameras can DEFINITIVELY say that this ball WOULD NOT have bounced over the
>stumps but THAT ONE would have.
>
>NOT possible!
And sitting in my arm-chair (not really), I tend to agree. I said it
before, and I'll repeat just once -- technology can be a trap. The
premise behind automation is that if a machine can do what a man can do,
the machine can do it better. Unfortunately, a machine cannot always do
better than a man in all cases. Note the failures of the CIA in recent
times because of an excessive dependence on remotely operated gadgetry
over a man on the scene.
In Baseball, they already have the technology to trace the arc of a
pitch as it crosses the plate and can project the box that represents
the batter's strike zone (it varies according to the height and
proportions of the batter), and can tell if a pitch was a ball or a
strike But this is done from behind the catcher at ground-level, from no
more than 30 feet away, and dealing with a ball that is in the air
throughout (can't recall if the catcher's re-positioning during the
pitch was a factor in any of this). Things should be a lot more
difficult for a set of cameras behind the bowler, beyond the boundary
(and in some cases seeing through the bowler during the initial arc of
the ball).
I don't think a 1000 cameras will have to be looked at -- after all
umpires are not flies -- but the 3rd umpire has to sample information
from all to rule out the possibility of contradictory information. That
takes time, and on every appeal? Whether out or not? Somehow I don't
think the principles used on the Space Shuttle will work in an
emotion-charged situation like an LBW -- on the Shuttle if 4 of 5
parallel computations agree, the 5th is discarded even if it
contradicts. Does that mean we go on an 80% chance of an LBW? Maybe that
will speed up the game by shortening the innings. And years from now,
The Bajan will do a statistical analysis on averages before and after
the LBW calls changed :-)
The cost of the setup will also be a factor. We may be talking about,
say 4 cameras initially, but with more controversy will come more
equipment in the guise of sophistication. Say 8-12 cameras at EACH end
(LA/RA, over/round the wicket, with redundancy), and heaven help the
authorities if an 'unofficial' camera later shows a contradiction (as in
the WC red-card incident). And I am not even touching the issue of
richer vs poorer nations. Only certain grounds will have these gadgets
and so will be favoured for international competition, initially anyway.
That would amount to a reversal of current trends outside England with
its 6 fixed venues. Okay, make the contraption portable. That requires
setup time and crew. And the more movable parts there are in any system,
the greater the rate of failure. The whinging about incompetent umpires
will become whinging about broken equipment.
But I suppose there's no harm in trying :-)
Apropos nothing, the International Skiing Federation stopped measuring
50Km Cross-Country races down to the 100th of a second, because it
simply didn't make sense. Ditto in swimming where times are only to the
100th of a second now, after an Olympics gold medal was decided by
2/1000th of a second; that too was a farce.
Cheers,
Ramaswamy
Stephen Devaux
> I don't think a 1000 cameras will have to be looked at -- after all
> umpires are not flies -- but the 3rd umpire has to sample information
> from all to rule out the possibility of contradictory information. That
> takes time, and on every appeal? Whether out or not? Somehow I don't
> think the principles used on the Space Shuttle will work in an
> emotion-charged situation like an LBW -- on the Shuttle if 4 of 5
> parallel computations agree, the 5th is discarded even if it
> contradicts. Does that mean we go on an 80% chance of an LBW? Maybe that
> will speed up the game by shortening the innings. And years from now,
> The Bajan will do a statistical analysis on averages before and after
> the LBW calls changed :-)
Damn right! or the S. Deveaux from the Bahamas will!
>
> Cheers,
> Ramaswamy
Phil. G. Felton
<md...@tig.com.remove.au> wrote:
> Phil. G. Felton <fel...@princeton.edu> wrote in article
> <felton-1208...@pgfelton97.princeton.edu>...
> > In article <35d7ff8b...@jhb-news.iafrica.com>, Mike Hesse wrote:
> >
> > > On Wed, 12 Aug 1998 21:36:21 +1000, "Steve G Smith" <sm...@cci.net>
> wrote:
> > >
> > > >He said they needed six ultra rapid cameras, at a cost of R1,5
> million, that
> > > >could photograph 2 000 frames per second and which would be able to
> > > >determine the trajectory of the ball.
> > >
> > > Hardly a practical alternative.
> > >
> >
> > This is an area in which I do research and my proposed system was similar
> to
> > Noakes'. I think it could be done somewhat cheaper than that although I
> forget
> > what the $/R rate is. I think that only 2 cameras would be needed at
> each end
> > and they'd need to be mounted high up so as to prevent obstruction by
> fielders.
> > I agree that high speed cameras would be needed, broadcast standards are
> not
> > adequate. Bad light might cause some problems also. It's certainly
> doable,
> > I think a suitable system could be developed in a year and then given an
> > extensive evaluation. It still wouldn't solve all the problems though, a
> > ball hitting the pads on the full might pose problems.
> >
> Sorry Phil but I don't believe we are within a million miles of the
> technology that could interpret the current LBW law using one, two or a
> thousand cameras! The number of variables that need to be assessed (point
> of pitching, point of contact, object first contacted and DEFINITE future
> trajectory including height) make this a mammoth task. I would be
> interested in any inside knowledge you might wish to share. Like how 10
> cameras can DEFINITIVELY say that this ball WOULD NOT have bounced over the
> stumps but THAT ONE would have.
>
> NOT possible!
I disagree, I believe I could come up with a system which would be as good
as a top class human umpire with better consistency (no distractions). The
point of impact could be assessed to within an inch as could point of first
contact, assessment of object first contacted would be down to the 3rd ump.
but he would have 2 views with far superior time resolution than existing
cameras and complete recall ability. If he still can't decide it's not out
just like now but he has had better data at his disposal than any one else
and can produce evidence (hard copy) to shut up the naysayers! As I said
elsewhere the ability to project the future trajectory will depend on how
far the ball has travelled since bouncing and how much further it has to
travel to reach the stumps. The quality of the projection can be assessed
and a uncertainty assigned, based on experience accumulated in the testing
phase you would say that if the probability of contact with the stumps was
below a certain value (say 80% for the sake of argument) then its not out.
This after all is basically what the umpire decides in his head except he
has to rely on an imperfect recall. This system doesn't take away the umpires
role, it gives him the ability to accurately replay the situation and make
a decision based on the best available data. Let's face a human umpire
can't meet the challenge you pose above!
I'm fairly confident that the above can be done based on my experience with
electonic imaging and computers. I'd excpect about a 2 yr lead time before
it could be ready 1 yr development and 1 yr evaluation.
Phil.
Jim Garner
"Steve G Smith" (sm...@cci.net) writes:
> August 11, 1998
>
> In the meantime, researcher Prof. Tim Noakes of Cape Town's Sport Institute,
> said he and his colleagues can develop computer technology which would
> accurately determine whether a player was trapped leg before wicket.
>
> could photograph 2 000 frames per second and which would be able to
> determine the trajectory of the ball.
Actually, there was some lad on this group two or three years ago who was
studying ballistics science. IIRC, he said ballistics is already so
advanced it is possible from a single ordinary camera to devise a computer
program that would accurately tell whether a delivery a) pitched between
the stumps and b) would have hit the wicket.
Of course, special cameras would be better.
This would revolutionize batting, if introduced. At present, the argument
is that when a forward-playing batsman, is struck on the front pad, it is
not possible for the ump to determine whether the ball would have gone on
to the wicket or not, so the batsman gets the benefit of the doubt. This
leads a lot of players to stick their front foot into the path of the
ball, feeling reasonably safe. Likewise, players playing back often
shoulder arms and let the ball go by, sticking their pads out in case it
breaks back.
With a thoroughly accurate system for determining the path of the ball,
these batting tactics would get to be a lot more iffy.
Jeez! We might even see some players attempting to HIT the ball.
--
Jim Garner, sage and dogsbody.
an...@freenet.carleton.ca http://www.ncf.carleton.ca/~an410
(613) 526-4786; 759B Springland, Ottawa, ON K1V 6L9 Canada
"The best-laid femmes go oft astray"
Mike Hesse
>Ahem.. but seriously.. academics have been claiming since cricket began that
> they can measure and model ball movement.
> It's patently obvious they can not.
> The debacle of the 'sniper' was a recent example,
> and no-one with a pencil has yet managed to explain why reverse
> swing works.
> Even if they did finally figure out it was to do with non-laminar
> airflow they then have to model it, which is 'impossible'.
>
> In other words, forget it.
I would be against using an untried system in international matches. I am
however also against blanket statements based on statements that it
is patently clear etc. With all due respect it is not patently clear. Science
has done far more marvelous things than work out the trajectory of a
cricket ball.
They do not have to explain why reverse swing happens, simply measure
and track it. Sure, we would love to simply put our faith in umpires but the
game is at an prof level where bad umpiring decisions (not biased) are not
something the spectators want to see. Ultimately it is the spectators that pays
the bills.
Cricket is well beyond the view of "it is good sportsmanship to take the bad
with the good" When I play in a social on Sunday, then sportsmanship
prevails. In the international arena where more money is being earnt than
most of us can dream of, the same principle does not hold. That is what
gambling joints are for.
The 3rd umpire can assist more than is currently being used & in my
view the logic of "they will never get it right" does not wash.
Phil. G. Felton
an...@FreeNet.Carleton.CA (Jim Garner) wrote:
> "Steve G Smith" (sm...@cci.net) writes:
> > August 11, 1998
> >
> > In the meantime, researcher Prof. Tim Noakes of Cape Town's Sport Institute,
> > said he and his colleagues can develop computer technology which would
> > accurately determine whether a player was trapped leg before wicket.
> >
> > He said they needed six ultra rapid cameras, at a cost of R1,5 million, that
> > could photograph 2 000 frames per second and which would be able to
> > determine the trajectory of the ball.
>
>
> Actually, there was some lad on this group two or three years ago who was
> studying ballistics science. IIRC, he said ballistics is already so
> advanced it is possible from a single ordinary camera to devise a computer
> program that would accurately tell whether a delivery a) pitched between
> the stumps and b) would have hit the wicket.
(a) yes (b) no, with normal broadcast standard cameras at 60 fps for a
ball travelling at 40 m/s will travel .67m (2ft) in each frame. What the
ball does
before pitching is not very relevant so you could have as little as 1 frame of
useful information!
>
> Of course, special cameras would be better.
>
> This would revolutionize batting, if introduced. At present, the argument
> is that when a forward-playing batsman, is struck on the front pad, it is
> not possible for the ump to determine whether the ball would have gone on
> to the wicket or not, so the batsman gets the benefit of the doubt. This
> leads a lot of players to stick their front foot into the path of the
> ball, feeling reasonably safe. Likewise, players playing back often
> shoulder arms and let the ball go by, sticking their pads out in case it
> breaks back.
>
> With a thoroughly accurate system for determining the path of the ball,
> these batting tactics would get to be a lot more iffy.
>
> Jeez! We might even see some players attempting to HIT the ball.
>
Not really the tactic will still be a good one, assuming you had a good camera
system which was able to track the ball accurately you would need to compute
its future trajectory based on the data from the instant of the ball hitting the
ground to the instant of hitting the pad. The less data there is the more
uncertainty in this trajectory and the further the ball has to travel to hit
the stumps the more this uncertainty will manifest itself. Thus playing forward
will still be a good strategy although maybe not quite as good as formerly.
Phil.
Bob Dubery
It seems to me, after reading the above from Phil, that there is a
potential problem with the prospective new technology.
When it comes to adjudicating runouts or other decisions involving
lines the technology is almost self-proving: the action is merely
slowed down and it can (in most cases) be ascertained whether or not
the line in question was touched or crossed.
With the deployment of technology for LBW decisions we will be in a
position where we have to take a computer's word that the ball would
have gone on to hit the stumps or not.
This worries me. There must be a way that the computer's
extrapolations can be proven to be reliable. How do we do this, short
of setting everything up in an indoor nets and letting it track
several thousand balls bowled by a bowling machine?
---
The e-mail in the headers is bogus ;-)
un-knot mega...@KNOTglobal.co.za for
my e-mail address
Jack Bramah
<mike...@nospam.com> writes
Who's earning the fabulous sums of money? Not the players, certainly!
>
>The 3rd umpire can assist more than is currently being used & in my
>view the logic of "they will never get it right" does not wash.
>
>
>Mike Hesse
>Email supplied on request
--
Jack Bramah <ja...@catland.demon.co.uk>
Ser galego é máis ca nacer en Galicia
MG
Phil. G. Felton wrote:
> In article <6r7hcj$j...@freenet-news.carleton.ca>,
> an...@FreeNet.Carleton.CA (Jim Garner) wrote:
> > This would revolutionize batting, if introduced. At present, the argument
> > is that when a forward-playing batsman, is struck on the front pad, it is
> > not possible for the ump to determine whether the ball would have gone on
> > to the wicket or not, so the batsman gets the benefit of the doubt. This
> > leads a lot of players to stick their front foot into the path of the
> > ball, feeling reasonably safe. Likewise, players playing back often
> > shoulder arms and let the ball go by, sticking their pads out in case it
> > breaks back.
> >
Simon Hughes wrote an interesting article in the Telegraph in which he said that
the introduction of cameras to help in caught behind decisions might bring back
walking, as batsmen who knew they were out would be unwilling to be exposed as
cheats by the camera.
--
Mike Gooding
"You're pretty brave in cyberspace, flameboy"
"Step inside"
Mike Holmans
to opine:
>
> Simon Hughes wrote an interesting article in the Telegraph in which he said that
>the introduction of cameras to help in caught behind decisions might bring back
>walking, as batsmen who knew they were out would be unwilling to be exposed as
>cheats by the camera.
Well, Simon is no nitwit, and is obviously very much in touch with
current player thinking, but I can't help feeling that this is a sop
to readers of the Torygraph, persuading that organ's Cretaceous
readership that the introduction of technology will actually foster
the traditional values they consider more important than the results
of games.
Frankly, this seems to me to be highly unlikely. As we know, Aussies
have never walked, saying that's what the umpires are for. The
introduction of cameras, especially if they aren't the sort of 2000
fps cameras necessary to be completely accurate, is much more likely,
in my view, to encourage batsmen to hang around and wait for the
decision.
But we do need something which will deliver the verdict very fast if
we're going to go down this road. If players are now expecting any
runout appeal under two yards to go to the third umpire, they're not
going to trust caught behind decisions except off wild top edges where
the ball goes fifty feet in the air.
Cheers,
Mike
--
"I'm so honoured that you've replied to one of my posts that I'm speechless.
Do I get a certificate, or something?" - Linda Taylor
Ian Dinsdale
>With the deployment of technology for LBW decisions we will be in a
>position where we have to take a computer's word that the ball would
>have gone on to hit the stumps or not.
>This worries me.
If if comes to a choice of the computers word - and remember they've no
reason to lie - and the umpire's opinion, I'd love to know how anyone
favouring the umpire justifies their choice.
Just realised people might read into this more than intended.
I am most emphatically not suggesting umpires lie. They might be
mistaken, but they're fair.
>There must be a way that the computer's
>extrapolations can be proven to be reliable. How do we do this, short
>of setting everything up in an indoor nets and letting it track
>several thousand balls bowled by a bowling machine?
>---
All that would have been done in the development phase before any system
was ever put into practice.
You don't have to it again.
Cheers
Ian D
Mike Hesse
>This worries me. There must be a way that the computer's
>extrapolations can be proven to be reliable. How do we do this, short
>of setting everything up in an indoor nets and letting it track
>several thousand balls bowled by a bowling machine?
At this stage I would be against this technology but I think the moneys
should be allocated to its development. The way to test it is
easy. Film real life situations (a bowler going full tilt at the wickets
in swinging conditions) but only give the testers info uptil the
point of say 1m before the wickets, see what they predict &
compare it to what actually happened.
For this technology to be useful, it must be usable by existing
umpires. Can you imagine the outcry if it were suggested that
umpires be replaced :-)
Phil. G. Felton
Dubery) wrote:
> On Fri, 14 Aug 1998 11:02:07 -0400, fel...@princeton.edu (Phil. G.
> Felton) wrote:
>
>
> >To reiterate the umpire would make the decision with input from the computer
> >I envisage the following info: 'ball pitched 1" inside the line of offstump,
> >probability of striking the wicket 95%, struck pad in front of middle stump.'
>
> It seems to me, after reading the above from Phil, that there is a
> potential problem with the prospective new technology.
>
> When it comes to adjudicating runouts or other decisions involving
> lines the technology is almost self-proving: the action is merely
> slowed down and it can (in most cases) be ascertained whether or not
> the line in question was touched or crossed.
>
> With the deployment of technology for LBW decisions we will be in a
> position where we have to take a computer's word that the ball would
> have gone on to hit the stumps or not.
>
> This worries me. There must be a way that the computer's
> extrapolations can be proven to be reliable. How do we do this, short
> of setting everything up in an indoor nets and letting it track
> several thousand balls bowled by a bowling machine?
I think that this is exactly what you'd have to do, as well as 'live' bowlers.
I think you'd also have some experienced umpires watch the videos also and
get their opinions.
Phil.
mega...@hotmail.com
i...@idinsdal.force9.co.uk (Ian Dinsdale) wrote:
> On Mon, 17 Aug 1998 19:21:21 GMT, el...@theking.org (Bob Dubery) wrote:
>
> >position where we have to take a computer's word that the ball would
> >have gone on to hit the stumps or not.
> >This worries me.
> If if comes to a choice of the computers word - and remember they've no
> reason to lie - and the umpire's opinion, I'd love to know how anyone
> favouring the umpire justifies their choice.
Personally, after having worked with computers for 15 odd years now, the
thought of taking a computer's word on anything - without exhaustive and
extensive testing and validation having been done before hand - seems to me to
be about the silliest thing that a person could do.
I'm not saying it can't be done, just that whatever contraption is contrived
must be rigorously tested and proven first. Adjudicating LBW decisions
*reliably* by camera is orders of magnitude more difficult than deciding on a
line decision.
Adjudicating them without a reasonable degree of reliability and accuracy is
child's play.
Oh.... and it better be Year 2000 compliant too.
-----------------------------------------
ACRONYM: Abbreviated Coded Rendition Of Name Yielding Meaning
RECURSION: See 'recursion'
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
mega...@hotmail.com
pos...@jackalope.demon.co.uk (Mike Holmans) wrote:
> Frankly, this seems to me to be highly unlikely. As we know, Aussies
> have never walked, saying that's what the umpires are for. The
> introduction of cameras, especially if they aren't the sort of 2000
> fps cameras necessary to be completely accurate, is much more likely,
> in my view, to encourage batsmen to hang around and wait for the
> decision.
I can't see anything wrong with a batsman merely not walking. To use a
judicial analogy the batsman is standing in the dock awaiting the judge's
verdict. There's no reason for him to stick his neck in the gallows before
the black cap comes out.
*But* if he commits perjury by rubbing various bits of anatomy or equipment,
then that is a fish of an entirely different feather.
The Aussies are right. It also occurs to me that a lot of the vociferous, not
to mention gratuitous appealing that goes on these days might be intended not
only to persuade the umpire that the batsman is out , but to persuade the
batsman too. Thus it's right (and one's right) to wait for the umpire too
rule.
But the umpire has to be a fairly level-headed sort in order to withstand the
pressure placed on him by loud orchestrated appeals.
Drewy
> In article <35d875cd...@news.global.co.za>, el...@theking.org (Bob
> Dubery) wrote:
>
> > On Fri, 14 Aug 1998 11:02:07 -0400, fel...@princeton.edu (Phil. G.
> > Felton) wrote:
> >
> >
> > >To reiterate the umpire would make the decision with input from the
computer
> > >I envisage the following info: 'ball pitched 1" inside the line of
offstump,
> > >probability of striking the wicket 95%, struck pad in front of middle
stump.'
> >
> > It seems to me, after reading the above from Phil, that there is a
> > potential problem with the prospective new technology.
> >
> > When it comes to adjudicating runouts or other decisions involving
> > lines the technology is almost self-proving: the action is merely
> > slowed down and it can (in most cases) be ascertained whether or not
> > the line in question was touched or crossed.
> >
> > position where we have to take a computer's word that the ball would
> > have gone on to hit the stumps or not.
> >
> > extrapolations can be proven to be reliable. How do we do this, short
> > of setting everything up in an indoor nets and letting it track
> > several thousand balls bowled by a bowling machine?
>
>
> I think that this is exactly what you'd have to do, as well as 'live'
bowlers.
> I think you'd also have some experienced umpires watch the videos also
and
> get their opinions.
>
The thing that continues to amaze me throughout the whole debate is the
inherent belief that if a player is given out by a computer rather than by
an umpire he will accept the decision in good grace and, more importantly,
we won't get 500 posts from disgruntled countrymen on this n/g when his
team loses the match! The computer says there is 95% chance that ball would
have hit the stumps so you're out! Well, I must be out then - and in the
news conference I'll tell all our fans that we were fairly beaten! Yeah,
right!!
Do you remember when tennis introduced the wonderful cyclops? This was a
machine that would take the guess work out of calling faults - it would
beep when a fault was served and the players would have no cause to argue.
Except that it would beep when the ball was in by feet!! And the players
would argue so much that the damn thing was usually turned off. Faults are
now called by human beings!
Have any of you ever got a speeding ticket, been sure you were within the
limit but agreed to cop it sweet when the officer said "But it wasn't me
who thought you were speeding, sir - it was my camera.It recorded your
speed as xx mph"?
Drewy
Phil. G. Felton
<md...@tig.com.remove.au> wrote:
Drewy, I agree that the use of a camera/computer system to evaluate lbw
appeals won't eliminate the complaints, however, the computer can't get
upset/annoyed/disconcerted by repeated appeals. Also it would take out some
of the personal vituperation out of the argument (viz the comments about
Ahktar's eyesight/age/hearing). Kitchen, for example, was clearly upset
by the reaction to his decisions and perhaps by seeing the slow motion replays
himself, judging by his comments afterwards. In my research I'm a firm
believer in getting images as well as electronic data, sometimes the latter
can be interpreted incorrectly, whereas an image can prevent this. This is
why I think an imaging system is superior to the Cyclops, after all the system
proposed could be reviewed by the 3rd umpire and any egregious error overruled,
in the Cyclops type of device there's no basis for review, 'the buzzer just
went off'. Finally, one advantage is that the umpires would have at their
disposal better information than the man in the pub' and could document their
decision. I think that the line cameras used for run outs have taken a lot
of the heat out of that decision and I think a good automated system for
evaluation of lbws could do likewise.
>
> Do you remember when tennis introduced the wonderful cyclops? This was a
> machine that would take the guess work out of calling faults - it would
> beep when a fault was served and the players would have no cause to argue.
> Except that it would beep when the ball was in by feet!! And the players
> would argue so much that the damn thing was usually turned off. Faults are
> now called by human beings!
I wasn't aware that they'd discontinued Cyclops, what principle did they
operate on did it use laser?
>
> Have any of you ever got a speeding ticket, been sure you were within the
> limit but agreed to cop it sweet when the officer said "But it wasn't me
> who thought you were speeding, sir - it was my camera.It recorded your
> speed as xx mph"?
Yes, although radar guns are used in very uncontrolled conditions by rather
inexpert operators, with no evidence presented that the transgressor was
actually you. I would expect the system I'm talking about to be better than
that.
> Drewy
Trudie van Gend
benefit should go to the batsmen?
Personally, I think too much technology ruins the game.
Trudie
Mad Hamish
>
>The thing that continues to amaze me throughout the whole debate is the
>inherent belief that if a player is given out by a computer rather than by
>an umpire he will accept the decision in good grace
I see this as more useful in avoiding the absolutely terrible decisions rather
than making the borderline ones easier for the players to digest. I'd _also_ say
that there aren't too many cases where the _players_ really kick up a fuss.
>and, more importantly,
>we won't get 500 posts from disgruntled countrymen on this n/g when his
>team loses the match! The computer says there is 95% chance that ball would
>have hit the stumps so you're out! Well, I must be out then - and in the
>news conference I'll tell all our fans that we were fairly beaten! Yeah,
>right!!
>
>Do you remember when tennis introduced the wonderful cyclops? This was a
>machine that would take the guess work out of calling faults - it would
>beep when a fault was served and the players would have no cause to argue.
>Except that it would beep when the ball was in by feet!! And the players
>would argue so much that the damn thing was usually turned off.
Obviously then it was rushed into use with insufficient testing. You're looking
at a system which clearly didn't work and saying that it means that people won't
accept a system even if it will work.
>Faults are now called by human beings!
>
Sorry but you're wrong, in the Grand Slams Cyclops is still used to call faults.
>Have any of you ever got a speeding ticket, been sure you were within the
>limit but agreed to cop it sweet when the officer said "But it wasn't me
>who thought you were speeding, sir - it was my camera.It recorded your
>speed as xx mph"?
I've never had a speeding ticket, but I know the people who calibrate the radar
guns in Tasmania and I trust them to do their jobs well. (I have less trust of
the breathalisers since they one had me at 0.00 when I'd had a beer 30 minutes
before hand but the breathalisers aren't designed to be accurate at low alcohol
readings).
I don't trust all technology but I do trust well designed technology. I am
pretty confident that this message will make it around the world without any
real hassles.
****************************************************************************
The Politician's Slogan
'You can fool all of the people some of the time and some of the people all
of the time, but you can't fool all of the people all of the time.
Fortunately only a simple majority is required.'
****************************************************************************
Mad Hamish
Hamish Laws
h_l...@postoffice.utas.edu.au
h_l...@tassie.net.au
Bob Dubery
Hamish) wrote:
>I don't trust all technology but I do trust well designed technology. I am
>pretty confident that this message will make it around the world without any
>real hassles.
Well, NNTP is not a 100% reliable protocol - I have access to 2 news
servers in SA and your message got to only one of them.
There's another angle to this debate that I heard aired by Cyril
Mitchley...
it's generally considered to be a good idea for umpires to be
ex-players as they understand the game, the way that is played, the
noises that a ball makes coming off a pad etc. etc.
But Mitchley feels that the technology could get to a point where the
umpire's control of the proceedings will be substantially lessened,
and that the job will consequently become less appealing to players
nearing the end of their careers.
Another thing: what happens when the contraptions - as they surely
will one day - break down? One suggestion I've heard is that some kind
of cyclops-style device be used to detect no balls arising from
over-stepping. Mitchley says that having to watch the bowler's feet
and then get one's eyes up quickly enough to watch the flight of the
ball is a skill that all experienced umpires have and that the cyclops
will not really reduce the umpire's work load.
But say it's employed. Before long we will have umpires who do not
have the skills that current top umpires have because they won't need
them...
...until the cyclops breaks down or starts misfiring. Then the umpire
has to do the foot calls himself, he will be too slow in getting his
eyes up to watch the progress of the ball and other details such as
whether the batsman hit it or not and there will be a rash of bad
decisions.
Bob Dubery
Mike Holmans
>On Mon, 24 Aug 1998 04:59:45 GMT, h_l...@postoffice.utas.edu.au (Mad
>Hamish) wrote:
>
>>I don't trust all technology but I do trust well designed technology. I am
>>pretty confident that this message will make it around the world without any
>>real hassles.
>
>Well, NNTP is not a 100% reliable protocol - I have access to 2 news
>servers in SA and your message got to only one of them.
Is that why you posted this message twice (both copies reached this
server, anyway)?
Cheers,
Mike
Bob Dubery
<pos...@jackalope.demon.co.uk> wrote:
>Is that why you posted this message twice (both copies reached this
>server, anyway)?
No.
That's down to what we in the trade call a PEBCAK problem.
Mad Hamish
>On Mon, 24 Aug 1998 04:59:45 GMT, h_l...@postoffice.utas.edu.au (Mad
>Hamish) wrote:
>
>>I don't trust all technology but I do trust well designed technology. I am
>>pretty confident that this message will make it around the world without any
>>real hassles.
>
>Well, NNTP is not a 100% reliable protocol - I have access to 2 news
>servers in SA and your message got to only one of them.
>
it evens out, I've received 2 copies of this reply <g>