The unknown target

Tom had organised a shooting competition behind the workshop.

He set up four large timber backing boards along the edge of the paddock, pinned a paper target to each one and invited four experienced shooters to take five shots.

Bill watched from the verandah with a mug of tea.

Dog watched from underneath the verandah with the concentrated interest he usually reserved for food, rabbits and things Tom had told him not to touch.

The four shooters took their turns. Five shots each. When the shooting finished, Tom waited for everyone to make their rifles safe and then walked down the paddock to collect the targets.

He stopped at the first board.

The paper target was gone.

Five bullet holes remained in the timber, grouped high and to the left.

Tom frowned and walked to the second board.

That target was gone too.

Its five holes formed a very tight cluster lower down and slightly right.

The third board had a wider spread of holes.

No target.

The fourth had another neat group.

No target there either.

Tom looked along the fence line.

Dog was disappearing beneath the bottom wire with a large sheet of torn paper hanging from his mouth.

“Dog!”

Dog accelerated.

Tom found the remains of the four targets behind the machinery shed. They had been torn, chewed, dragged through the dirt and arranged into something resembling a nest.

The bullseyes were no longer attached to anything useful.

Tom returned to the backing boards and thought for a moment,

Bill had followed him down the paddock.

“Well?” Bill asked.

“Number two won.”

Bill looked at the second board.

“How do you know?”

“Tightest group.”

“It is a tight group.”

“So he wins.”

“Where was the bullseye?”

Tom pointed vaguely at the board.

“Somewhere around there.”

“Where exactly?”

Tom looked at the five holes.

The paper had covered most of the board. Without it, there was no mark showing where the centre had been.

“It doesn’t matter,” he said. “The shots are close together.”

“That tells you he shot consistently.”

“Exactly.”

“It doesn’t tell you what he consistently hit.”

Tom frowned.

“He was aiming at the bullseye.”

“They were all aiming at a bullseye.”

“Yes.”

“And now all four bullseyes are gone.”

Tom walked to the first board.

The holes were closely grouped, but nowhere near those on the second.

Then he looked at the fourth.

Another tidy cluster in a different place.

The third shooter’s holes were more widely scattered.

“Number three clearly lost,” Tom said.

“Maybe.”

Tom turned towards Bill.

“How could he possibly have won?”

“One of his shots might have passed through the middle of the bullseye.”

“But the rest were all over the place.”

“That would make him less consistent. It wouldn’t tell us whether one of them was accurate.”

Tom stepped back and looked at all four boards.

The shot patterns were obvious.

The result was not.

“We had four experienced shooters,” he said. “Good rifles. Same distance. Same conditions.”

Bill nodded.

“And we have four sets of holes.”

“So surely we can judge them.”

“You can judge how tightly each shooter grouped their shots.”

“That’s the same thing.”

“No.”

Tom pointed to the second board.

“That one looks best.”

“It looks most precise.”

“What’s the difference?”

“Precision is putting your shots in the same place. Accuracy is putting them in the right place.”

“And without the targets…”

“We know where the shots landed. We don’t know where they landed compared with the bullseyes.”

Tom stared at the boards for a while.

Then he brightened.

“I could draw new bullseyes around the groups.”

“You could.”

“Put the centre of each target in the middle of the holes.”

“You could do that too.”

“Then we could compare them properly.”

Bill took a sip of tea.

“No. Then you’d be comparing the shots with targets you drew after seeing the results.”

Tom lowered the pencil he had already taken from his pocket.

“That wouldn’t prove where the original targets were.”

“No.”

“It would just make the results look right.”

“Yes.”

Dog returned carrying a damp scrap of paper with half a red circle on it.

Tom took it from him and tried to match it to one of the boards.

It could have belonged to any of them.

Dog sat down and looked pleased with his contribution.

Bill pointed at the four backing boards.

“This is the problem with a lot of expert work.”

Tom looked over.

“What has this got to do with experts?”

“Sometimes the real answer can’t be seen. It might be underground. It might depend on weather that hasn’t happened yet, a machine that hasn’t been built, or an operation that hasn’t started.”

“So the expert estimates it.”

“Yes. They collect information, make decisions, apply a method and produce a result.”

“Like the shot groups.”

“You can see whether the work is detailed. You can see whether the numbers are consistent. You can see whether the report is professionally prepared.”

“But you can’t always see the target.”

Bill nodded.

“And if the target is unknown, you may not know whether the advice is close to the truth, far from it, or very precisely grouped in the wrong place.”

Tom looked at the second board again.

“So a neat result can look convincing without being right.”

“Yes.”

“And another expert might put their shots somewhere completely different.”

“Yes.”

“Which still doesn’t tell us where the bullseye was.”

“No. But it tells you that the answer may not be as certain as either shooter thought.”

Dog began chewing the last surviving piece of target.

Tom reached for it.

Dog swallowed.

Bill finished his tea.

“Well,” he said, “at least we’ve removed any remaining doubt about whether we can check the result.”

What Tom Learned

A tight group shows consistency.

It does not prove that the shots hit the target.

It's possible to be consistently wrong. And it's possible to be inconsistently correct. Without a known target it is hard to tell.

Expert advice can be careful, detailed and professionally presented. But where the true answer cannot be directly observed, one report may not tell us how close the advice sits to the truth.

A second independent result may not reveal the missing bullseye. It can, however, show how much the answer changes when someone else takes the shot.

The Parker Challenge

A few years ago I helped organise an industry conference in my main area of expertise, Mineral Resource estimation (or MRE). The conference brought together over 500 Australian and International experts over two days to discuss new trends, old practices and generally share knowledge. For those who have never heard of Mineral Resource estimation, it’s a niche discipline in the mining industry and, every publicly listed, producing mining company relies on the work. In a nutshell it involves estimating the tonnes and quality (grade) of the mineralisation to be mined and judging the confidence associated with the estimate. Practitioners take widely spaced samples, often more than 20-50m apart and, from those limited samples determine the size, shape, location of the orebody - the basic information that determines its potential value. The models underpin mine plans, mining schedules, corporate production forecasts, stock market valuations.

It’s a central foundation of the modern mining industry.

My part of the conference organising committee involved a strange concept. We arranged a competition. We offered anyone who was interested a set of data and asked them to run and document their estimate. They had to use the data to:

• Consider the geology

• Identify the extent of the mineralisation or orebody

• Estimate the copper grade and

• Apply industry standard confidence classifications (Measured, Indicated or Inferred)

The idea was simple. We were testing how different the outcomes could be.

MRE is a mature discipline. Its beginnings date back over 70 years, the mathematics is well established. It involves standard practices using established techniques. It incorporates ideas around recognising uncertainty and variability. It has reporting codes mandated by ASIC the ASX and several professional bodies.

And it involves professional judgement.

That’s where the competition idea comes in. The competition (The Parker Challenge) was testing an idea that few people consider. How much does the person doing the modelling matter? How much influence do their decisions and judgements make on the final result?

I presented the results in a Keynote presentation on the last day of the conference. The first thing I did was ask the audience how wide they thought the results would be, how different these experts in the discipline would be. After all, they started with the same data.

Most people considered +/-10% the likely outcome.

The results were a bit different to expectation. Even after removing the two highest and two lowest results, the estimated copper metal classified as Measured and Indicated ranged from 66 per cent below to 91 per cent above the average of the entries.

For a discipline that is used by investors, businesses and even governments.

We shouldn’t be surprised by the results. Anyone who has asked two builders to diagnose the same flaw, two doctors to interpret an unusual set of symptoms, or two valuers to assess the same property knows that trained people can examine the same evidence and reach different conclusions.

Problems arise when these human judgement factors are not tested or conceived. When we start to see an expert’s report as a definite answer with low error. And that happens a lot with consultant reports. A consultant can sound confident and authoritative. They may have professional credentials, education, experience. Those are good properties for a career as a consultant, but they also affect the way people read or interpret consulting reports. Detailed reports with supporting figures and charts, technical language and firmly stated outcome can make it hard to understand that the work is based on human judgements and thus has potential flaws or large error margins.

I should know, I’ve made big mistakes myself. Then I’ve packaged those mistakes into a structured consulting report and stated my opinion as if it was fact. I remember once being wrong by a factor of 100! All because the clock was ticking and I was in a rush. An all too common consulting trap.

Bluntly, consultants and experts are human. Our work involves judgements and opinions and therefore, even with the best of intentions, the best data, the best processes, we cannot guarantee that the results are ‘the truth’. We can’t even guarantee the results lie in the middle of the likely outcomes.

Which takes me back to the Parker Challenge competition. One of the better ways to assess an expert report is to have an independent review of the work and an independent check of the results. Without that type of check the level of uncertainty around an expert’s opinion (yes even my own) is unknown.

What happens if a ‘second opinion’ doesn’t exist? There are some tools and techniques everyone can use to evaluate how much confidence they should place on an expert report. As you read or review the report ask these questions

• What question was the consultant actually instructed to answer?

• What matters were outside the scope?

• Which information came from the client?

• Which assumptions were independently checked?

• Does the evidence support the conclusion, or merely remain consistent with it?

• How sensitive is the result to different assumptions?

• Were uncertainty and operating variability examined?

• Has anyone independently reviewed the work?

• What changes in data or circumstance would make the work invalid?

• What necessary assumptions or conditions does the report depend on?

• Did the consultant assess the proposal, or a simplified representation of the proposal?

What does this have to do with the proposed Biosolids and non-recyclable waste incinerator at Glan Devon? The development application contains a series of specialist reports. Those reports should be read seriously. They should not be dismissed merely because the applicant commissioned them, but neither should they be accepted merely because qualified consultants prepared them. Each specialist report should be examined on its own terms. The reports should then be considered together to determine whether their assumptions, inputs and conclusions form a coherent assessment of the proposal.

What conclusions are drawn, and what is the basis for those conclusions? Do the conclusions from one report flow naturally into the assumptions in subsequent reports? Has each specialist answered a narrow question or a broad one, and how has that answer been presented in the application? What is missing or overlooked?

I am not asking anyone to substitute my opinion for theirs. My own analysis should be tested in exactly the same way. Check my sources, examine my reasoning and point out my errors. I could be wrong. My analysis is often based on the work of others.

Checks, reviews, audits are not a critique of any individual expert or company. They are practices aimed at addressing the challenges of human judgement. Professional expertise is necessary, but professional presentation must not be mistaken for proof.

The question for Council is not whether a consultant has signed a report. It is whether the evidence, assumptions, treatment of uncertainty and reasoning are sufficient to support the decision being requested—and whether enforceable conditions can keep the actual development within the circumstances assessed.

It is not enough to record that the application includes “Report XYZ” and therefore treat the relevant assessment matter as addressed. The real questions concern the report’s relevance, reliability, scope and reasoning.

© 2026. All rights reserved.