Embedded Files
It is easy to be excited about a fresh rookie class as an opportunity to turn around the fortunes of your fantasy franchise (assuming you didn’t go Rams circa 2021 and trade away all of your picks). Despite this excitement there is some benefit in looking backward to last year’s rookie class for two reasons.
First, looking at how rookies performed relative to how they were ranked is somewhat instructive of the quality of rankings. Granted, rookie ranks (including mine) are almost always intended to be long-term projections but it is nice when those projections get off to a hot start and you don’t have to hope to see your player’s name on a ‘second-year breakout candidate’ list. There is a chance my rankings were terrible in year one and will turn around in year 2 (or 3 or 4) but the data suggests that year one performance is strongly correlated with year 2 for most positions (but that the relationship between year 2 and 3 is even stronger - see Riske, 2020).
Second, fantasy managers can be an impatient group (for good reason given how commonplace it seems for dynasty leagues to unexpectedly fold). This win-now mentality coupled with the previously mentioned excitement surrounding an incoming rookie class can cause managers to trade away (or cut) players too soon or make suboptimal draft decisions by not understanding the value of second year players based on year 1 performance (good or bad).
With these two ideas in mind, I have conducted an analysis of my 2022 combined OFF/IDP Rookie Rankings and present the full dataset below. I conduct this analysis by position rather than based on overall rankings because the original exercise was meant to provide draft guidance rather than overall predictions. For instance, my top rated 2022 CB was #71 in my combined draft rankings not because I thought he would score the 71st most points among all players but because I thought he was the top CB and relative to the value of other positions and my evaluations of the entire pool, this is where I was comfortable recommending drafting that player. Still, if you are going to use my rankings (or anyone else’s) you want to have a sense of how they do across all of the positions they include so that even if you disagree with overall positioning of positions relative to one another (or your roster composition or scoring system shifts relative value) you can still extract value from the within position rankings.
When examining how well my within position rankings correlated with 2022 fantasy output, I am largely pleased with the results. Correlation scores range from -1 (reality was the opposite of my rankings) to 1 (my rankings perfectly aligned with reality). The unweighted mean correlation of my 2022 rookie rankings by position with 2022 fantasy point rankings by position was .414. When looking at my positional rankings in quartiles, the unweighted mean fantasy points scored by the top quartiles was 115.6, with the second quartile significantly lower at 65.84, the third at a similar average of 57.25, and the bottom quartile averaging only 31.44. In sum, these data points suggest that I did an overall good job predicting 2022 rookie relative output (although I would love to be able to see how these numbers stack up against others but that will be a project for next year).
In terms of positions, I had the most success with DE (correlations with pts, PPG, PFF grade, and snaps were all .475 or greater) and performed similarly well with WR (all correlations were .349 or greater). In terms of struggles, I suspect I was not alone in difficulty deciphering the pecking order in the 2022 TE class but this is also a position where rookies are not typically impactful. My LB rankings also did not meet my expectations with some significant overvaluation (for now) of Nakobe Dean and Channing Tindall and undervaluation of Malcolm Rodriguez and Jack Sanborn.
Full results are available here: 2022 Rookie Rankings Revisited.
Beyond looking at how I did, the data I compiled can be helpful for you as you consider roster moves related to 22 rookies. Hopefully, ideas leap out of that data at you but here are a few ways I use this data in my leagues.
Snaps vs fantasy output. Players who ranked higher in terms of snap count than they did in fantasy output (total or PPG) are in less than ideal positions. If free agency or the draft creates competition which reduces volume, this could spell trouble.
PFF Grade vs fantasy output. People have different feelings about the importance of PFF grades for fantasy. While they are of varying levels of importance for different positions, my own research has shown there tends to be a modest correlation between some sub-grades and fantasy output with weaker relationships between overall grade and output. That said, players who performed well (via PFF) but did not net the expected fantasy results in year 1 are strong holds or buys for me in many cases as candidates for positive regression towards this mean correlation.
Snaps vs PFF grade. Players who performed well with limited opportunities should be given more leeway before writing them off. On the opposite end, players who performed poorly and still played many snaps might be prime candidates for a role reduction and should be viewed with caution.
MyRank. Lastly, I like to go back and see where I had some guys this time last year to look for some buy low opportunities. A player who I had ranked highly and may have underperformed in year 1 can sometimes be acquired from a manager for a round or two less than their previous cost and are equally (if not more) likely to succeed than the completely unseen rookie you would take with that pick in the 2023 draft.
I hope you have found this all interesting or (even better) useful!
For more on the methodology I use to calculate my rookie ranks, see https://www.professoridp.com/rookie-rankings.
For 2023 rankings, stay tuned as I hope to have a pre-draft version available in mid-April.
Page updated
Google Sites
Report abuse