In 2 previous posts, I looked at what draft strategies worked in snake drafts for 2020 and how different league settings affect positional value. Today I am combining these to understand how different league settings affect draft strategy. Specifically, we will be looking at how different draft strategies fared as league size, points per reception, number of playoff teams, and passing TD value varied. Should you focus on an early QB or TE when there are only 8 teams? Should you go RB heavy early even in a full PPR? Let's find out.
The Data
As with previous posts, we will be looking at the results of the 2020 season and analyzing how different decisions affected the odds of winning the championship. In the end, we will be comparing 8, 10, and 12 team leagues, standard, half-point, and full point PPR, leagues with 4, 6, or 8 playoff teams, and 4pts and 6 pts per passing TD. We can see the total number of teams in each category below. Remember that ESPN defaults to 10 teams, 4 team playoff, full PPR, and 4 pts per pass TD so that will be the majority of public leagues (a reminder of how your chosen platform will shape your experience).
A note, clearly some of these combinations have pretty low representation. I will be covering all 4 of these settings, but recognize that the results will be more robust as we move towards ESPN's default settings.
As far as metrics go, I will be focusing on a team's odds of winning their fantasy championship. In order to compare across settings, I have normalized all championship odds relative to change from random chance. So in a 10 person league, you have a 10% chance of winning at random, so if a certain strategy gave you a 12% chance, that will be shown as 2% in the following graphs. In general, any positive value is increasing your odds of a championship, while any negative value decreases your odds.
When to draft a QB
League Size
Line plot of your change in championship odds based on when you first draft a QB, split by league size. Red line represents the random chance for a given league size.
For QBs, we see the same shape for the curve for all 3 league sizes, but the effects are larger in the 8 team league vs a 10 or 12 team league. As we saw with the overall draft strategy there is a big bump in the middle rounds that is really driven by Josh Allen. But in general, you should be drafting a QB in the middle rounds (say between rounds 4 and 9) regardless of the size of your league.
PPR
Line plot of your change in championship odds based on when you first draft a QB, split by PPR setting. The red line represents the random chance for the average league size of that sample.
We see a little more noise with PPR, with half-point PPR suggesting a slightly narrower range of when to target a QB (rounds 5-8) compared to full point PPR. Overall though this is a negligible change and you should stick with grabbing a QB in the middle rounds, likely a mid-level QB1 with a chance to make a jump.
4 vs 6 Points per TD Pass
Line plot of your change in championship odds based on when you first draft a QB, split by Points per TD Pass. The red line represents the random chance for the average league size of that sample.
Interestingly, having passing TDs worth 6 points narrows the range of when to target a QB to 4th through 8th round roughly. Comparatively, having 4 points per TD pass widens the range, with positive odds from 3rd through 12th round. My guess is that by reducing passing TD value, you allow for more qbs to be viable, particularly at the end of the draft where a QB who throws for a lot of yards (like Matt Ryan) can make up for not throwing as many TDs. This does not change the overall message of targeting a QB in the middle rounds, but does mean you have more wiggle room in 4pt pass TD leagues.
Number of Playoff Teams
Line plot of your change in championship odds based on when you first draft a QB, split by the number of playoff teams. The red line represents the random chance for the average league size of that sample.
I think more interesting is how the number of playoff teams changes the draft strategy. In particular, as soon as the number of playoff teams increases from 4 to 6 or 8, you should wait on QB, with positive odds only starting around round 7 or 8 in the larger playoff formats. My guess is that you don't have to win as many weeks during the regular season to make the playoffs when more teams make the playoffs. This provides a cushion to figure out the QB position and rewards teams that emphasize other positions earlier.
When to draft a TE
League Size
Line plot of your change in championship odds based on when you first draft a TE, split by league size. The red line represents the random chance for a given league size.
For TE, just like with the overall strategy, we again see the benefit of drafting a top TE last year. Across the board grabbing a TE in the 2nd round was beneficial, with that extending to the 3rd round for 8 and 10 team leagues (and that includes many owners who drafted George Kittle only for him to miss most of the season). After that, you kinda just want to draft a TE before the 10th round in 8 and 10 team leagues, or really at any round in 12 team leagues. My guess is that missing on a top TE hurts you less in the 12 team league simply because you face the top TEs less often. I will be really interested to return to this graph after the season to see if Travis Kelce really is worth a first-round pick.
PPR
Line plot of your change in championship odds based on when you first draft a TE, split by PPR. The red line represents the random chance for the average league size of that sample.
In PPR leagues we actually have an interesting result. Again we see the benefits of an early TE, though the benefit is largely erased in standard non-PPR leagues. Instead, there was a big spike in the 10th rd for standard and a second spike in the 11th round for half-point PPR. Not really sure what to make of that but maybe a later TE like Robert Tonyan really hit on TDs without a lot of receptions. The other point is that full point PPR (which is most leagues) basically follows a trend of the earlier you take a TE the better (outside of round 1). This seems to support the idea of securing Travis Kelce at the end of round 1 this year.
Number of Playoff Teams
Line plot of your change in championship odds based on when you first draft a TE, split by the number of playoff teams. The red line represents the random chance for the average league size of that sample.
Similar to the PPR leagues, as you move away from the standard 4 team playoff, you start adding a benefit to both early and late TE. This may be a small sample size effect, but I also think it has to do with how wins are more important from the get-go in leagues with a smaller playoff format. The larger playoff formats just gives you time to figure out these one starter positions.
What Positions To Target With Your First 2 Picks
League Size
Bar plot of your change in championship odds based on what positions you take with your First/Second pick split by league size. The red line represents random chance.
I find this graph fascinating. The clear advantage of going RB/TE followed by RB/WR shows up across all league sizes. Interestingly, WR/TE was actually beneficial in 10 team leagues while being a hindrance in 8 or 12 team leagues. In general, this suggests trying to maintain a balanced approach to your roster, likely because it allows you to grab the best player available more easily later when you have options in your starting lineup. As with the general strategy and with the QB draft rounds above, we see that you should not draft a QB in the second round (though the fact that Mahomes and Lamar got hurt or underperformed respectively last year does not help this analysis). Finally, zero-RB seems like a poor choice given the overall difficulty in a WR/TE or WR/WR start.
PPR
Bar plot of your change in championship odds based on what positions you take with your First/Second pick split by PPR setting. The red line represents random chance.
When we look at the splits on PPR, the most glaring (and unsurprising) result is how important drafting a top-end RB is in non-PPR leagues. While RB/WR and RB/TE were the top 2 again, RB/RB comes in third and any combination that does not include an RB significantly hurting their odds of a championship (cutting them in half). Non-PPR is also the only version that has positive odds to the RB/RB start. I also think it's interesting that half-point PPR had better odds with the WR/TE combo than full-point PPR, though I doubt that is a significant effect and may just be some small sample size noisiness.
Number of Playoff Teams
Bar plot of your change in championship odds based on what positions you take with your First/Second pick split by PPR setting. The red line represents random chance.
Overall, I don't think much of interest is occurring here. There is obviously the massive drop for the WR/TE championship odds for an 8 team playoff, but I worry that is more small sample noise than anything of actual substance. In that sense, I don't think that you should change your strategy at the top of the draft based on the number of playoff teams
Limitations
The biggest limitation will always be the underlying data, particularly how many teams are 10-team full-point PPR. That combined with all of these results being a reflection of 2020 definitely adds a couple of grains of salt to this analysis. That said there are some interesting trends that come out here that we will certainly be watching to see if they continue into 2021. Overall, the limitation that bothered me more is that I could not explore other setting combinations (think 2 vs 3 WRs or 1 vs 2 starting QBs) simply because I did not have enough leagues to feel confident in any conclusions.
Takeaways
Overall, we can see that some league settings do seem to impact draft strategy. You should wait on QB longer if more than 4 teams are making the playoffs. You should target an early TE if there are PPR points, but don't worry about it if there you are in a standard league. Importantly though, a lot of strategies (like starting RB/WR) seem to work across all setting combinations, though which players are valued likely changes more than the positions overall.
Questions? Comments? Let me know at ac@alexcates.com. Want to read more breakdowns like this? sign up for my newsletter. Finally, like what I do? Consider supporting me on buy me a coffee.
Comments