The Theory Of Risk

“When you are eager for a project to work, you are more inclined to be overconfident…” (Plous 1933)

Risk is thought of as the possibility of something going wrong and the cost if it does. If you put it into an equation, the equation would look like this:

RISK = PROBABILITY OF OUTCOME x COST

You like to think that things will go well and we will meet the best-case scenario and reap maximum benefit, you like to think that the probability of something going wrong is less than .01, so you put a low likelihood on that event.

So you get: Bad Outcome = .01 (1%) / Good Outcome = .99 (99%)

This is what ‘you like to think’ in an ideal world. But the world is less than ideal.

- It’s that Simple – or is it?

You will already be over-estimating that things will go right due to the ‘overconfidence principle’. Because your expectation is optimistic you don’t see the true risk you take.

You are also typically overconfident about your returns too…

To endure risk is low, you typically use, ‘the best of breed technology’. This is usually at high initial outlay, but you expect a high return on investment, so that’s OK. – or is it?

An Example of Risk

Say you invest £500,000 on a technical specification for the project. The tech is brilliant and does way more than you need, but that extra tech you don’t need but have it ‘just in case’ costs you. So you have already spent probably an extra 20% on tech you simply did not need. Even though you don’t realise it, speculating funds on something that is superfluous – is risk. Likely returns on this cost are zero.

You have now invested £500,000 on a project where your expected return exceeds £2,000,000 so you feel confident that this is money well spent. You have a high expectation of good returns, but you ‘know’ you will have a payback of ‘at least’  £80,000 – which covers costs and gives you a little more besides. With good marketing, press coverage etc your outlay will reap your anticipated higher returns too.

Using the above equation, the less than the covering costs scenario is not even given much consideration.

 

So here is a chart of your prediction:

• £200,000 success                       50% probability
• £80,000 moderate                     49% probability
• £-50,000 failure                            1% probability

Looks good doesn’t it?

 

Let’s Go Topsy-Turvy

What if, we turn this equation on its head?

Instead of looking at risk as an effort of ensuring high probability of success, we look at risk as an exercise of reducing cost?

COST = PROBABILITY OF OUTCOME x RISK

What we are saying is, “OK, you know what you want – eventually – but instead of listening to all the hype of what these great solution can do (for you and your competitors equally), you focus on the actual cost of the solution (in monetary terms as well as labour, time and other factors) and decide, instead of going for the ‘best solution’ (reducing the risk by raising the probability of things going well), you go for the best-fit and least costly solution before committing to a huge expense.”

As an example, a camera or phone. You get lots of functionality on your equipment these days, but you don’t typically use it all. Lower cost cameras will do the job you need perfectly adequately. Typically you don’t use all the filters – some look so awful we would ‘never’ use them, even though you have paid for them ‘ as a package’. Do you see where I am going now?

The same goes for high-ticket solutions to a problem, much of what is paid for is ‘insurance money’ when you could have gone for a lower cost solution that would have done the job adequately.

 

Recalculating and Rethinking

Now back to the initial (more realistic) scenario – you thought that the project had a risk of .01 – yet it is still a risk and factoring overconfidence into the equation would make the ‘failure’ option a lot more likely. In fact it is well documented that, certainly in IT, 30% of projects fail – so take that .01 and move it up to .3. Why put a he investment into something that has 30% chance of failing? Not only that, you lose the 50,000 overinvestment you initially put into the project that you didn’t need to.

 

Standish reports that only 9% of projects are actually successful, therefore this equation looks like this instead:

• £200,000 success                      9% probability
• £80,000 moderate                   61% probability
• -£50,000 failure                       30% probability

This is a very different scenario than you looked at initially, but more realistic and less over confident.

“When you are eager for a project to work, you are more inclined to be overconfident, yet oddly enough overconfidence decreases as knowledge increases, we become less sure as we get better at something.” (Plous 1933)

This is why it is important to always look at a best-fit solution (usually a hybrid), to minimise risk by reducing the cost of a project. Bearing in mind the likelihood of failure. It is also wise to limit initial outlay.

New technology is prey to overconfidence. The more proven a technology is, the more we can rely on the probability of it working because we have sorted out the teething problems and have experience in that technology so can use a best-of-breed solution.

With new technology, don’t overlook the ‘little guy’ – he may just be the market leader of the future!