August 24, 2020

Warm-up with AFM – Comments on the Sample Case

Today we have published the first sample case “Warm-up with AFM“, similar to one of those that we plan to offer participants at the Financial Modeling World Cup. In this article, I wanted to give some additional comments about the case structure and the general principles of question design at FMWC.

The questions vary in complexity. Typically, more complex questions would bring more points than the easier ones.
However, we believe that a wise approach to choosing questions is important to maximize your result at FMWC.
Most of the questions in the Sample Case 1 are easy enough. They don’t require to have the full 3-way model complete (except for 5, 12 and 13). To solve these questions you need to create just the corresponding schedule which can be done fast enough. At the same time, these questions can bring you as much as 140 points.
Remember that the total length of the round is 2 hours and you can get a maximum of 1,000 points.
On the contrary, questions 5, 11 and 12, require participants not only to create a full-scale model but also to create some scenario variations of it. The total payback for these questions is 60 points, but it would take you much more minutes per point! Probably the best approach would be to leave them for the end of the round if you have time.
Most questions have 6 answer options. We plan to create some of these answer options to “catch” the most common mistakes. So, if your answer is equal to one of the options offered – make sure you are perfectly comfortable with your approach – it might be a pit.
Question 5 is another type of “pit” here. Interest expense goes right after depreciation in the P&L, so it seems very logical to have this question as #5. However, we have purposefully created it complex enough – you can’t solve the question without spending a lot of time creating the full model. So, we also want to test your ability to make a quick workload evaluation before tackling the case.
Question 9 seems to belong somewhere near Q4, right? Here we wanted to make sure that you can easily switch back to the parts of the model that were already created. Another approach would be planning the whole model construction in advance so that you can see which cases are solvable in bundles.
Luck and guess-timation. While we think that it is a bad idea trying to solve questions 11 and 12 during the round (unless you only want to train for the AFM exam – then it’s exactly the right setup for you), more complex questions could be tackled with logic and luck. For example, some answer options could be eliminated, especially for Q5. Then you have higher chances to correctly guess the result. We think that a very small impact of luck could bring more joy to the competition – we don’t want to eliminate it at all.
You are welcome to share your thoughts on the question design. These principles will definitely change over the course of the season, taking into account the feedback of participants.
Write me to andrew@fmworldcup.com if you’d like to contribute ideas or if you’d like to become one of the case authors. We plan to attract a wide number of case authors at the same time NOT restricting their participation in FMWC. I will give more details on how we expect to do it in one of the next blog posts.