I thought I put this out to see if anyone is interested. I'll be doing a World Cup prediction games as a learning opportunity for myself (and yourself if you participate). The reason is this: We make so many stock predictions in i3 everyday but how many of us actually learn from our own prediction? Little. This is understandable given the nature of stock prediction with long feedback loop. Say, you might predict a stock to do well in the mid-term 3 to 5 years, which means it takes that long before you find out how well is your prediction. Throw in the element of luck, it becomes even harder to tell. Not to mention our prediction is not done probabilistically.

This is where game comes in handy. Any game, in this case, World Cup games, have short feedback loop: You'll know if you're right or wrong in 90 mins. Short feedback loop = Faster learning opportunity; Game is discrete, there is an end point whereas stock is continuous, which makes it easier for measurement purpose. And in just one month time, you can make 64 predictions. Not a lot, but a great way to give youself insights on how to improve. 

The plan

So how are we going to measure the accuracy of our prediction? Brier score. Mauboussin explains it this way:

"Psychologists commonly use the Brier score as a method for gauging the accuracy of probabilistic forecasts...In its simplest form, the Brier score measures the square of the forecast error, or (forecast − outcome)2 . For binary events, the value of the outcome is 1 if the event occurs and 0 if it does not. As in golf, a lower score is better...When calculating the Brier score this way, you consider the squared forecast error for both the event and the non-event.

Exhibit 5 shows a meteorologist’s probabilistic forecasts for whether it will rain over the next four days. For example, on Day 2, she forecasts an 80 percent probability that it will rain. Likewise, we can say she forecasts a 20 percent probability that it will not rain. Because it did rain, we place a 1 in the outcome column below “Rain” and a 0 in the “No Rain” column. Her Brier score for that day was 0.08. For multiple forecasts, the overall Brier score is the mean of the scores for each forecast. The meteorologist’s overall Brier score comes to 0.25. "

Brier score will be the 'ruler' where you measure how well you do with your prediction. In Philip Tetlock's Superforecasting, a Superforecaster - someone who is good at making prediction on all kind of world events i.e politics, economics, sports etc has a Brier score of 0.2 - 0.25, or even less. In other words, the lower the score, the better you're at making prediction. 

Knowing how well you can predict is crucial in investing, as Buffett says you want to swing "when the odds is in your favor". But many people swing thinking the odds is in their favor when it is not, a calibration error between subjective confidence and objective truth. 

For the sack of simplicity, this is how it will work. Russia vs Saudi's game took place yesterday. I'll be using Betfair and Bet365 as the benchmark - aka the market. Few hours prior to the kick-off for the first game, both sites gave a 67-69% probability of Russia winning, while I assign a 60% probability, so as a result, based on the Brier formula, my brier score is higher than the market - the market is better than my prediction in this case. I'll do this for 64 games and compare the score.  

Of course, you can do this by yourself, the formula is easy to compute on spreadsheet, or I'm happy to do it for you, you just need to email me if you're interested - jia.yeo@gmail.com. While the World Cup mataches is not binary i.e Rain/No Rain, but trinary i.e Win/Lose/Draw. But to make it simple, I'll just pick the country that's the most likely to win and assign a probability on that. 

A few things I expect we can learn when this finishes is:

1. Market expectation - It is hard to outsmart the wisdom of the crowd (maybe I'll be wrong, we will find out), but this will definitely be an exercise of humility. To put your ego to the test. If you dare to take the challenge.

2. Learn probability - You'll learn how to think in probability and being comfortable using it. 

3. Measure yourself - You'll learn a new tool on how to measure your prediction and apply it onto your investing. 

I'll put up my results as well as anyone who participate after the World Cup ends. Thanks.