In this final installment of the Dynasty ADP series, we will look at the performance of players selected in the Early Rounds of the observed dynasty drafts. To review the background, read part 1 here. For the analysis of the Middle and Later Rounds, read part 2 here.
Deeper Dive: Early Rounds (Picks 1-90, corresponding to rounds 1-6)
As mentioned in Part 2, the expectation for the ‘Early Round’ selections was that the data would show very few busts and that most players would essentially retain their value over two years. From personal experience, dynasty owners generally do not like to commit to taking older veteran players in the Early Rounds whose performance window is relatively short (such as Ryan Braun) unless they are generally considered elite; in other words, only if the expectation of elite current year performance drastically outweighs the potential risk of performance falling off a cliff would an older player be selected in the top 90. Simply put, we expected that the analysis would show that players selected in the Early Rounds (defined as picks 1 through 90) would have a bust rate near zero.
Sure enough, from the data set consisting only of players selected in the Early Rounds (n=179), there were zero busts (excluding Jose Fernandez who was removed out of respect, RIP). Note, however, that although there were no busts, there were surprisingly still a few selections whose ADP worsened by over 600 slots (or ~40 rounds) two years later. They were:
- Jonathan Lucroy who went from pick 79 (at age 31) to pick 901, two years later
- Carlos Gonzalez, who went from pick 84 (at age 31) to pick 695, two years later
- Troy Tulowitzki, who went from pick 59 (at age 31) to pick 653, two years later
- A.J. Reed, who went from pick 72 (at age 23) to pick 658, two years later
- Johnny Cueto, who went from 65 (at age 31) to pick 632, two years later
- Addison Russell, who went from pick 54 (at age 23) to pick 609, two years later
The two year change in value of the one hundred seventy nine (179) players selected in the first six (6) rounds was distributed as follows (again, green means improvement, red means worsening, grey is neutral):
Surprisingly, of the 65% (n=117) players who lost more than one round of value two years later, 92 of them fell out of the first 6 rounds entirely. In other words, 92 out of 179 (51%) of players selected in the first 6 rounds were not worth an Early Round selection two years later. Again, this suggests that dynasty owners – even in the Early Rounds where “top” dynasty assets are being selected – appear to be highly discounting the future (and/or potentially over-weighting the current year’s performance), are mispricing what should be the easiest assets to price, or both. The quintile performance of age cohorts taken in the Early Rounds (identical to the analysis performed in Part 2) helps to better illustrate this:
With each age cohort depicted below, we can tell a similar story as with the other rounds:
The youngest age cohort performs reasonably well. The top 60% players in this age cohort essentially retain their value (ie have their ADP stay within ‘21’ picks or so, ie 1.5 rounds). The bottom 20% of this age cohort, however, averaged a drop of 168 picks (or 10 rounds) two years later.
Just as with the age 17-to-22 cohort, if you select an age 23 to 25 year old in the Early Rounds, there is a 60% chance (ie the top 3 quintiles) that the player retains his value (within 25 picks of where they were taken) two years later. Again, the bottom 20% within the cohort degrade significantly: although there were no busts, they averaged a drop in ADP of 275 picks (or 18 rounds) two years later.
Once again, the top 60% (top 3 quintiles) retained value (ADP stayed within 30 picks or ~2 rounds) two years later. But once again, if you were unlucky enough to take players who ended up in the lowest quintile within the cohort, they dropped in ADP by an average of 241 picks (16 rounds) two years later.
Wow! Here – as with all of the other sections of the draft – players who are age 30 or older need to be avoided as much as possible in order to still have value two years later. Only the players who retrospectively fell into the top quintile maintained value; whereas every other player who was age 30+ dropped significantly in two years.
Conclusion (Early Rounds)
Only 35% of players selected in the Early Rounds retain their value or improve. To improve this yield (and avoid busts or vast reductions in value), owners should avoid players who are age 30 or older.
What Does It All Mean??
Firstly, I would be remiss to neglect to re-emphasize that the above analysis is merely a snapshot of the performance of a very narrow collection of Dynasty Leagues. Care should be taken before broadly applying their conclusions across all of your leagues. It is a genuine risk that the insights derived above may only pertain to the specific ‘types’ of owners involved in the drafts that just happened to be examined. However, it must be underscored that within each of the analyzed dynasty drafts, the owners employed the full ‘expected’ range of drafting strategies (i.e. of “going for it”, of “playing for next year”, etc.). Additionally, because much of the analysis is performed on aggregated data points (quartiles and quintiles-within-the-quartiles) – rather than on singular data points – the effect of ‘oddly extreme’ draft outliers would be significantly reduced. For example, if an owner was inexplicably high on Kyle Gibson one year, this should negligibly influence the observed behavior of the larger data set.
Note also that there is inherent selection bias within the data. In general, if an 18 year old prospect was selected in a dynasty draft, it was probably because they are a player who is expected to be a valuable player in the future. The conclusion from the data, therefore, should not be to select a younger player merely because they are young. The data merely suggests that the younger players who were selected are potentially being undervalued (and/or are not having their potential future value being appropriately priced compared to their risk) by owners.
Nonetheless, the above data analysis has a number of striking conclusions and recommendations of which any owner in a new dynasty start-up should be aware. By section of the draft – in order to maximize player value in two years – assuming they are not focusing primarily on year 1 – they should:
- Avoid players who are age 30 or older in the Early Rounds
- Avoid players age 26 or older (and focus on hitters) in the Middle Rounds (discussed in more detail in Part 2) and
- Be aware that > 50% of players over age 30 selected in the Later Rounds will be out of the league in two years (again, see Part 2).
The above analysis could also be applied to help evaluate trade value. If, for example, you are considering trading for a player who is over the age of 29 years old, know that there is a 27% chance (based on the overall bust rate data) that the player will be not good enough to roster in two years.
In closing, remember that everything is cyclical and strategies always shift. The savvy dynasty owner should be constantly re-evaluating their leagues and their opponents’ tendencies toward valuing players or player types. With the recent immediate impact of the Acuñas and Sotos, there has probably been more of a recent leaning toward prospects than what was observed in 2016; similarly, if Vlad or Eloy fail to meet expectations over the next few years, we may see an over-correction the other way. It is all dynamic (and that’s a big part of the fun).
A final note is that, philosophically, valuing players ultimately depends on your own taste, strategy, and approach. If you know that “Player A” will only be worth $2 in year 1, $12 in year 2 and $36 in year 3, do you consider him more or less valuable than a player who will be worth $16 in year 1, $14 in year 2, and $11 in year 3? Ultimately this depends on a number of factors, including how your competitors view his value, whether you think you can trade him at his peak, or if you need a “$12” player this year. And so on and so on.
Hopefully, this series has shone a light on some tendencies that may exist in your dynasty league which could potentially help you build a better team. Now go out there and rip off your fellow owners!
Leave a Reply
Friends don't let friends talk to themselves.
You must be logged in to post a comment.