Studying Median in late 2024

Matt’s 2025 beer bet strategy was listed incorrectly on the seasonal calculator from December 2024 until Feb 6th, 2025. The correct strategy number is 255953!

I didn’t write this post until 2/6/2025, but note that I am ignoring all 2025 data for the purpose of writing it. The strategies discussed here do not take into account any 2025 data.

In December 2024 I added median to the Seasonal Builder. We have 2004 – 2024 data available, so I like to split that into two different ranges when using the builder. I used 2004-2015, and best fit the highest median with no other criteria. That kicks out strategy # 255953

Since it was generated using 2004-2015 data, the second step is to “test” it on 2016-2024 data. This will give you an idea of how the strategy has performed without hindsight bias (that is, say you built it with all available data on 1/1/2016, then used it for the next 9 years). Over the 2016-2024 time period, that strategy returned 330.71%, compared to the C fund 237.16% and the S fund 155.85% for the same time frame. Eureka!

I took it a step further, and noted that 255953 appeared to do poorly during election years. So I opened excel and calculated what the return would have been for 2016-2024 if I used 255953 in all non-election years, but just used C fund buy-and-hold for election years:

Wow! Even better. This is why I picked 255953 to be my 2025 strategy.

There are two other strategies that I am going to be watching in 2025:

1. Highest median for all available data, 2004-2024. That strategy is # 274016
2. Highest median for 2004-2015 but only using C,S, and I. That strategy is # 275236 (2016-2024 return is 166.54%, not as promising).

Here’s a link to watch how all 3 of these perform throughout the year.