Why Does The Optimal Path Tool Not Recommend The Top Ranked Pick For My Pool?
Sometimes, My Pool Picks rankings and the Optimal Path Tool can differ. We explain why that is, and how you should treat that information.
We provide several different Survivor tools that help you research your pick each week, and they all attack the data from somewhat different angles, based on the purpose of the tool.
Occasionally, this can lead to conflicting pick suggestions for a specific week from different tools. In particular, when two or more teams look like very attractive picks based on the current week data (win% and pick%), the Optimal Path tool and the My Pool Picks custom pick rankings may end up suggesting different picks as the “best” for your pool.
It’s important to remember that this is occasionally just splitting hairs — sometimes very little difference separates the top few teams listed in the My Pool Picks custom pick rankings, for example.
Here’s an explanation of the differences between those two tools, and why their picks might differ.
They Treat Future Value Differently
The Optimal Path tool tries to find a single path through the rest of the year that maximizes your expected value (Immediate EV from the My Pool Picks custom pick rankings) each week. This means that if a team is projected to be the second-best option in several future weeks, but the absolute best option in none of them, the Optimal Path tool won’t care at all about saving them for the future, and might go ahead and suggest using them now.
The My Pool Picks custom pick rankings, on the other hand, take a more holistic view of a team’s future value. If a team looks like the second-best option in several future weeks, then the My Pool Picks custom pick rankings will assign them quite a bit of future value, and will lower their current-week grade. This is because:
- Our projections for future weeks aren’t perfect. The My Pool Picks custom pick rankings effectively account for this. If we’re slightly off on our public pick% projection for a game 8 weeks from now, for example, then maybe the team in question will actually end up being the best pick that week, instead of just the second-best like we think they will be now.
- It’s always good to have a backup plan. Even if our current projections are great, what if the QB of the best option in some future week gets injured before then? It’s good to be able to plug in the second-best pick.
The result of these differences is that the My Pool Picks custom pick rankings are more robust and future-proof than the Optimal Path tool recommendations. A single bad future prediction only makes a marginal difference in the My Pool Picks custom pick rankings, but can completely change the Optimal path tool output.
An Example With Numbers
For those of you with a more mathematical mind, here is a simplified example that illustrates how the Optimal path and the My Pool Picks custom pick rankings could produce different suggested picks. If you don’t care about the numbers, go ahead and skip this section.
Let’s assume there are only 3 weeks left in the season, and all we care about is win odds (so we’re ignoring public pick%, not worrying about your opponents’ remaining teams, and paying no attention to your pool’s rules and size, in order to simplify the example).
Here are the only teams you have left to choose from, and their win odds in each of the last 3 weeks:
- Team A: 89%, 50%, 50%
- Team B: 50%, 90%, 50%
- Team C: 50%, 50%, 90%
- Team D: 90%, 89%, 89%
The Optimal Path tool would suggest you choose Team D this week, since the path with the maximum win odds is D/B/C with win odds of 90/90/90.
The My Pool Picks custom pick rankings would rank Team A ahead of Team D this week, because Team D looks like a great option in the last two weeks, and you only need to sacrifice 1% in win probability this week in order to preserve some flexibility the last two weeks.
As you can see, the My Pool Picks suggestion (Team A) leaves you less vulnerable to a freak injury in the last two weeks. In addition, if our win projections change, and Team D becomes, say, 93% to win in the last week, you’ll have them available.
So What Should You Use?
Early in the season, we suggest relying on the My Pool Picks custom pick rankings for current week pick analysis, and using the Optimal Path tool only to help with future week planning.
Later in the season, when we have a very good idea of the relative strengths of teams, and when the exact teams you have remaining becomes a bigger factor, the Optimal Path tool can be useful to help plan out endgame strategy. Plugging your top few available options from the My Pool Picks rankings into the Optimal Path tool as this week’s pick, and examining the resulting future paths, can help illuminate any future weeks where it’s vital to save a specific team.
To summarize, we recommend relying on the My Pool Picks custom pick rankings to generate an initial ranking of teams, because it more properly weights the uncertainty involved in our future week projections. Later in the season, the Optimal Path tool can be used to highlight cases where your particular mix of available teams means that a specific team is more important for you to save, in which case you may want to skip past them to your next available My Pool Picks team.