CW decides the greatest test spinner ever. 43 names: Countdown/Rankings thread

[the big bambino]

NotMckenzie identified the issue when he said the object is to make clear evidence inadmissible by requiring an unnecessary level of complexity.

 

[CricAddict]

Link? I'm missing a fair bit. Something to do with that CasMg guy who was pretending to be a girl?

And Steve Waugh got banned for stating to give pleasure to that guy.

 

[Starfighter]

It's funny how this discussion started over Murali who is hardly as clear cut as some other cases, as certainly could bowl all his deliveries legally (and unlike, say Griffin, finished with his arm bent rather than straight), although whether he did all the time is an interesting issue and one that will never be resolved.

You've identified the issue when you said the object is to make clear evidence inadmissible by requiring an unnecessary level of complexity.

It's sort of like if I had an object that consisted of a flat plate of steel, about 30x40 cm, slightly bowed along the long axis, and attached in line with this axis in roughly the same plane as the plate it had a piece of wood, rounded, about 5 cm thick by 160 cm long and had a T-shaped piece on the far end of the piece of wood, also of similar thickness but only about 15 cm long and set at right angles to the large piece, but in the same orientation as the flat plate, and I didn't call it a spade.

 

[StephenZA]

In the final image shown is his arm bent away from the camera or possibly in towards the camera? In the second image shown is there no possibility his arm is still bent and his body has twisted? The only way you can be certain is by having another camera angle of a photo at the same time. You can call that overly complex. All I'm saying is that single camera shots do not give good 3D images and can be misleading. That is not a controversial statement, this is a known phenomenon. Does that mean I think that you ignore the photos taken, no I don't. All I have said is that it needs to be taken with caution.

 

[the big bambino]

How about you just have a photograph showing a bent arm?

 

[vic_orthdox]

How about you just have a photograph showing a bent arm?

Why, having a bent arm the whole way through isn't a throw.

 

[the big bambino]

Look at the next frame.

 

[vic_orthdox]

Look at the next frame.

Sorry, thought you were making a general point, rather than in that instance.

 

[the big bambino]

I agree if there’s no straightening it’s ok but I’m guessing if anyone saw a bent arm they’d be interested enough to follow it through to release,

 

[OverratedSanity]

The reason why I still think a 2d photograph is a decent enough indicator is that if at some point early in your delivery swing if you have a bent elbow, keeping it bent all the way through is ridiculously difficult. Murali had a deformity which prevented his elbow from straightening past a particular angle. Unless you have a defect like this in your arm, having a visibly bent arm early in your action should automatically be a bit of a red flag that something might be wrong.

 

[NotMcKenzie]

In the final image shown is his arm bent away from the camera or possibly in towards the camera? In the second image shown is there no possibility his arm is still bent and his body has twisted? The only way you can be certain is by having another camera angle of a photo at the same time. You can call that overly complex. All I'm saying is that single camera shots do not give good 3D images and can be misleading. That is not a controversial statement, this is a known phenomenon. Does that mean I think that you ignore the photos taken, no I don't. All I have said is that it needs to be taken with caution.

And again, I would argue that your level of caution is too strict. We don't need angles in degrees quoted to five significant figures: all we need is to see that there was a straightening after his arm. The frames are taken to coincide with release, but an offset of a few milliseconds will not make that much difference.

The angles are the only possiblity of concealing a still-bent arm, so we can take a broader view:

In both cases, his arm is straight at release. One could argue whether the difference in angle is enough, perhaps 40–60 degrees, but it certainly eliminates a region in which a bent arm at release could be concealed. Perhaps a camera on a line between third man and mid on or mid wicket, but a bent-all-the-way-around arm can only conceal itself in so many places.

So again, I think 2D analysis is adequate to show this.

 

[Starfighter]

In the final image shown is his arm bent away from the camera or possibly in towards the camera? In the second image shown is there no possibility his arm is still bent and his body has twisted? The only way you can be certain is by having another camera angle of a photo at the same time. You can call that overly complex. All I'm saying is that single camera shots do not give good 3D images and can be misleading. That is not a controversial statement, this is a known phenomenon. Does that mean I think that you ignore the photos taken, no I don't. All I have said is that it needs to be taken with caution.

The Big Bambino's point is that you seem to be going out of your way to look for confounding factors to avoid drawing an obvious conclusion. See my 'spade rule' above. Another case, Saeed Ajmal. Plenty of photos showing him with his arm markedly bent as it's coming up but dead straight at delivery. Not all at the same angle or in the same sequence though. Are they a 3D projection? No. Does they, collectively make a strong case? Yes. Not necessarily one that would pass a court of law or at a 99% confidence interval, but satisfactory enough.

 

[StephenZA]

And again, I would argue that your level of caution is too strict. We don't need angles in degrees quoted to five significant figures: all we need is to see that there was a straightening after his arm. The frames are taken to coincide with release, but an offset of a few milliseconds will not make that much difference.

The angles are the only possiblity of concealing a still-bent arm, so we can take a broader view:

View attachment 24266View attachment 24267

In both cases, his arm is straight at release. One could argue whether the difference in angle is enough, perhaps 40–60 degrees, but it certainly eliminates a region in which a bent arm at release could be concealed. Perhaps a camera on a line between third man and mid on or mid wicket, but a bent-all-the-way-around arm can only conceal itself in so many places.

So again, I think 2D analysis is adequate to show this.

The Big Bambino's point is that you seem to be going out of your way to look for confounding factors to avoid drawing an obvious conclusion. See my 'spade rule' above. Another case, Saeed Ajmal. Plenty of photos showing him with his arm markedly bent as it's coming up but dead straight at delivery. Not all at the same angle or in the same sequence though. Are they a 3D projection? No. Does they, collectively make a strong case? Yes. Not necessarily one that would pass a court of law or at a 99% confidence interval, but satisfactory enough.

Where did I ever say that 2D analysis was not good enough? I said be cautious of using single photos. "As an aside (not that I'm saying Griffin did not throw) but be careful of using 2D pictures to try determine this sort of thing." That is the statement that I made, nothing controversial, nothing against known good advice.

I think everybody is missing the point here. I never said you could not use photo's all I said is that using them needs to be done with caution. That's it. You give me 2 photos that appear to show a bent to straight arm I will go huh, not enough info. You give me 100 photos (hyperbole!) from varying directions over a number of deliveries, no probs. If you want to say that a few photos is good enough so be it I believe that is insufficient because I know that, particularly, single frame photos can be misleading. I have never argued that it could or should not be used.

 

[Migara]

I think you can use 2D images to screen bowlers who need urgent testing. And there should be a given number of bowlers with so called "clean" actions are needed to be tested to keep as control values. Currently, the latter "controls" are non existent.

 

[Starfighter]

Where did I ever say that 2D analysis was not good enough? I said be cautious of using single photos. "As an aside (not that I'm saying Griffin did not throw) but be careful of using 2D pictures to try determine this sort of thing." That is the statement that I made, nothing controversial, nothing against known good advice.

I think everybody is missing the point here. I never said you could not use photo's all I said is that using them needs to be done with caution. That's it. You give me 2 photos that appear to show a bent to straight arm I will go huh, not enough info. You give me 100 photos (hyperbole!) from varying directions over a number of deliveries, no probs. If you want to say that a few photos is good enough so be it I believe that is insufficient because I know that, particularly, single frame photos can be misleading. I have never argued that it could or should not be used.

The problem is your definition of 'good enough' seems to be based on too high a bar and I'm sure you'd still manage to introduce complicating factors. Call a spade a spade. You have two photos there on widely separated angles - and I can get one taken from roughly square leg from the same documentary which shows the same thing, though it is the least clear due to the shape of the elbow joint.

 

[Starfighter]

I think you can use 2D images to screen bowlers who need urgent testing. And there should be a given number of bowlers with so called "clean" actions are needed to be tested to keep as control values. Currently, the latter "controls" are non existent.

I've certainly thought that a few clean bowlers looked a bit interesting. The current standard is based on that 15° is roughly the point where a) the extension becomes really noticeable and b) begins to confer an advantage in terms of pace etc. I'd be interested to see where some bowlers stand with respect to this. And obvious ones should be pointed out, I think the earlier that such actions are caught the better the chance of making them better. although that's something that really needs to happen at the grassroots level. My own club has 2-3 bowlers who throw, one quite badly, and about every 2nd-3rd team I play during the season has one, because people are too afraid to call spades spades. This is in Australia. If you can fix it while a bowler is still in their formative years they'll probably never have problems again.

 

[StephenZA]

The problem is your definition of 'good enough' seems to be based on too high a bar and I'm sure you'd still manage to introduce complicating factors. Call a spade a spade. You have two photos there on widely separated angles - and I can get one taken from roughly square leg from the same documentary which shows the same thing, though it is the least clear due to the shape of the elbow joint.

I don't have to introduce any complicating factors; the complicating factors are there. If you think that 2 or 3 still photos are good enough and show enough to say somebody was chucking, so be it, I don't.

 

[NotMcKenzie]

....

Problem is, I provided more than one photo; I first two from one angle showing that there was actually a straightening and not just that he had a bent arm. And then after providing more, you naturally provide a whole bunch more reasons as to why it might be misleading, deceptive, requiring caution , needing frames exactly simultaneously or whatever. I could provide all of the frames between, but they'd merely show what I've already shown: an arm that is straight at release but bent when back behind the head: i.e., it has straightened. Furthermore, multiple angles suggest that it is not a case of the arm changing orientation with respect to the camera.

The fact that one can concieve of an objection does not completely or automatically preclude the hypothesis or observation from being true. Saying, 'you know, single frame photos can be misleading,' borders on a thought-terminating cliche: a way to dismiss anything objected to straight away. You admit that the pictures show the arm straight [at release] when it was bent earlier in the same action, but you say it is 'not enough info' based specious reasoning that is in effect always demanding more sophistication or something even though it is not necessary to use anything more sophisticated means when multiple photos from different angles show the same thing. Your caution appears unwarranted: if I was to try measure the amount of extension then yes, we would need a more precise means, but saying that 'extension occured' is less precise and arguably requires less precision.

 

[Starfighter]

I don't have to introduce any complicating factors; the complicating factors are there. If you think that 2 or 3 still photos are good enough and show enough to say somebody was chucking, so be it, I don't.

I don't think your complicating factors are as significant as you say, especially as we do have angular separation. This isn't, say, particle physics and we don't need an absurdly high bar to draw a reasonable conclusion.

 

[StephenZA]

I never argued whether he was chucking or not.... I said be careful of using still shot 2D photos. Which you then provided more and more of to prove he was chucking. You will not convince me that 2 or 3 photos is enough to show that the player was chucking; because those images can be misleading. If you start showing me multiple photos, particularly from multiple angles yes that begins to provide enough evidence. And that reasoning is not specious in anyway.