Its almost draft night, you have your plan, your target players and your favorite sleepers. but then its the 3rd round and a QB comes off the board, then another, then before you know it, a run has begun, 5 QBs go and you are worried if you don't take your favorite now, there won't be any left when you have your next pick.
This is a classic example of the herd mentality, the tendency of individuals to mimic the trades, selections, or behaviors of a larger group, often bypassing their independent analysis or intuition. Many people have talked about avoiding herd mentality in fantasy football, but how much does herd mentality actually effect outcomes?
In the QB run described above, you may be tempted to reach for a QB like Jordan Love in the 3rd round rather than lose out. Such a pick is driven not by the value of the player, but by FOMO. While there is an obvious problem with reaching for players, I've previously shown that losing some draft value is not as big of an issue as we may think (within reason).
So should you participate in positional runs? Or avoid the herd mentality all together?
The Data
For this we are using a dataset of over 10,000 snake drafts from the 2023 season, all collected from sleeper leagues. These span leagues of different sizes and scoring settings. For each draft, I defined a positional run as anytime 4 players at the same position were taken within 6 picks. I then merged consecutive runs together (so a RB run from 5-11 and 8-14 were merged to be 1 run from 5-14. I also ignored the first pick of a run, since its hard to argue they are following a herd mentality when they are setting the trend, not following it.
Number of Runs vs Final Standing
Line plot showing how the average finish of a team based on the number of positional runs participated in. Error bars represent standard error.
And right off the bat, a surprise. When we take the simplest approach, across all leagues, when the number of runs you participate in increases, your final standing improves! We see the final standing improving (getting closer to 1st) as the number of runs increases. While there is a little leveling off from a total of 6-11 runs, the effect seems to hold as the number of runs increases further.
Such a result runs counter to the standard anti-herd mentality and actually suggests that you should be participating in positional runs. Now, it's important to note that I am combining all of the different league types here. Next let's compare the effect of different league types.
League type vs value of participating in a positional run
To assess the effect of league type, I will look at the slope of the best fit line between the number of runs and final standing. For instance, the linear correlation of ppr leagues has a slope of -0.1 and an r-squared value of 0.03. The negative slope means that participating in more runs leads to an improved final standing. The more negative, the larger the effect (each run improves your final standing more). R-Squared measures how strong or reliable the relationship is, in this case an R-squared value of 0.03 is pretty weak. Because we are using 10000 drafts, almost any correlation will be statistically significant, but the question is whether the relationship is meaningful (and therefore something to consider during your draft). With that in mind, let's look at the slopes in each league type.
Scatter plot of correlations between draft run participation and league type.
Again, we see a pretty consistent relationship. Regardless of league type, 1 or 2 QB, PPR or Standard, Dynasty or Redraft, all of them have a weak negative relationship. Standard and IDP seem to have a steeper slope, possibly because there are only so many RBs and Defensive Players that matter and you can seriously miss out. But every league has a negative relationship here. That said, these are all pretty week relationships and I am not sure there is anything conclusive here.
Does the Position of the Run Matter?
Scatter plot of correlations between draft run participation and position of the run.
If we go back to looking across all league types and examine the position of the run, we actually see the largest effect based on the Defense and QB. I think that is because these are the most likely to be susceptible to runs and where there is the strongest drop off if you miss out on the top players. Being a part of a run ensures that you are not left with a truly terrible QB or Def. That said, all of these relationships are still weak. That doesn't mean it doesn't matter, but this is just one piece of what makes a successful fantasy draft.
If position does not matter, does the timing of the run matter?
Does the timing of the Run Matter?
Scatter plot of correlations between draft participation and league type.
When looking at the round that a draft run occurs in, again we basically see no difference between whether you are participating in a run early or late in the draft. Across the board, participating in drafts improves your final standing (remember a lower final standing is better so we want a negative slope).
If the timing of the run doesn't matter, maybe the length of the run does?
Does the length of the run matter?
Scatter plot of correlations between draft participation and final standing split by the length of the draft runs.
Here we see a definitive trend that the impact of participating in a run increases with longer runs (a more negative slope). These are all still weak relationships in terms of the overall effect on final standing, but it's interesting to see such a consistent increase with the length of the run. I wonder if this is more of a punishment, such that teams that don't participate in the long runs are punished more (due to such an extreme drop off at the position).
These are all still weak relationships thought, maybe it's about when you join the run, rather than details about the run itself that matters?
Pick in the Run
Here we see that it is likely better to be early, but not to start the run. 3-5th in the run seems to be the sweet spot. Basically, don't be copying every position the guy in front of you chooses, but if a run is starting, you should jump in early. I wonder if 3-5 players that is a normal cutoff for tier breaks at a position? If you are joining the run then, you may be grabbing players at a similar tier, ensuring you get a top level guy. If you join the run later, you are likely reaching for a lower value players, rather than pivoting after the tier is over.
We can try to measure this by comparing the value of the pick vs adp in each run.
Value Gained/Lost in the Run
Line plot of the pick value accumulated during a draft run and the team's final standing. We see a negative slope (with more value gained leading to a better final standing), though the overall effect is rather small.
In line with the last graph, participating in a run makes more sense when you are accumulating value relative to ADP. Sometimes an entier tier of a position may drop in the draft and you should participate in the rush to take advantage of that drop. That said, you should not be reaching to be a part of the run. If you see an entire positional tier dropping, you should get in on that. If a run has started and crossed the tier break, you had better avoid it.
Limitations
As always, there are limitations. For simplicities sake, I tended to combine all of the leagues together here. I also did not control for league skill, and maybe runs tend to happen in lower or higher quality leagues. Finally, all of the correlations here qualify as weak correlations. While the results should therefore be viewed cautiously, it does leave us with a couple of takeaway points.
Conclusions
Surprisingly, I did not find that participating in runs is a bad idea, instead the timing of when you jump in the run matters. You should be joining runs, but you should join them early, while the run is still starting, and especially if you can grab a player at value still. Try to avoid reaching on a player just because a run is happening, if there have already been 5 or 6 players grabbed at the position, joining the run now is likely a bad idea. While all of the relationships described are pretty week, the takeaway is that runs should not be avoided all together. Instead, there is some value in joining the herd, just make sure you are joining the right one.
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 or by signing up for fantasyleaguereport.com
Commenti