We outlined the full set of rules for the Newfound 2018 March Madness Brackets here.
After two more rounds, we are down to the Final Four in our Newfound 2018 March Madness Bracket Challenge, and while the right hand side of the bracket looks predictable with two number 1 seeds facing off (Kansas and the current tournament favorite, Villanova) the left hand side has 11 seed Loyola facing off against 3 seed Michigan.
If Loyola pulls off a win – FiveThirtyEight is giving them a 31% chance of doing so – then they will be the first team seeded lower than 8 to ever make the National Championship Game.
In our Sweet 16 update, we published the current standings along with the expected final standings based on the win probabilities at that time. However, expectations and reality rarely line up (see our post on the Lie of Averages).
The expected number of wins for each team has changed dramatically after two 9 seeds (Kansas State and Florida State) and 11 seed Loyola made it through to the Elite 8. Kentucky (5) didn’t capitalize on its “easy” route to the Final Four.
These deviations from expectations were great for many of the brackets. Relative to the expected ranks at the Sweet 16, the average absolute change rank based on the current path of the tournament is 4. For brackets like justinkl, this led to an expected shift of 25 ranks in just two rounds. The 10% allocation to Loyola certainly helped!
Unfortunately, with how much the value-tilted and equal weight benchmarks crushed it in Round 1, they are still sitting in first and second place.
The good news is that we have had one bracket emerge as a contender to knock out the equal weight benchmark! It may be a longshot requiring Loyola to win it all, but Movaltrend could do it.
At Newfound, we like to focus on risk in all its forms. When we are discussing investments that are exposed to a continuum of outcomes given the infinite ways the future can unfold, this quantification of risk can be a very involved exercise.
Thankfully, with March Madness when we only have four teams left, this isn’t quite so difficult; there are only 8 ways the future can unfold. Using this knowledge, we can look at our current rankings and not only see the expected final rank of each participant but also see the best and worst possible outcomes in all states of the bracket.
WARNING: The graph below may not make you happy. There are some unfortunate truths presented in it that you may not want to know.
You can see a few things in this chart:
- Some final ranks are already determined (a preemptive congrats to mike.p and sharonp4 for your strong finish – great minds must think alike).
- Some brackets will move up in rank. That’s good news for jmdishman and jcwest.
- Some brackets will move down in rank. That’s bad news for Scarton Seed Ranking and galluci.
- There is still a decent amount of mobility possible. Daniel.M, Retirement Planning Center, Pete_N, jeff.k, cappysweets, justinkl, athrasher, and tylander could all move up to 5 ranks.
If there is overlap between your bar and another one, there is a possibility that you could overtake that bracket. However, since there is no mention of correlation, it is possible that the states of the tournament that increase your rank may also increase their rank.
While we can specify with certainty that the tournament will end in one of 8 ways and have somewhat of a handle on the likelihood of each scenario, which path the true bracket takes is up in the air.
We’ll check back at the end of the tournament to see what lessons we can take away from the brackets, be they good, bad, or ugly.