Late-Round Best Ball Profiles That Actually Need Spike Weeks
Two players can carry the same July ADP and mean completely different things for your roster. One needs a spike week to matter at all. The other is quietly building a real role that just hasn't caught up to the price yet. Best ball drafters lump both into "late-round sleeper" and move on. That's the mistake this piece is built to interrogate.
What a spike week is actually measuring
RotoGrinders defines a spike week as the probability a player clears the 80th percentile scoring output at his position in a given week. That's a specific bar, not a vibe. A player who hits it once or twice a season on a busted coverage or garbage-time touchdown is not the same as a player whose role supports that outcome on a repeatable basis.
Establish The Run's best ball primer makes the sharper point: the value of boom or bust profiles gets overstated in ADP nearly every year. Drafters chase variance because variance is memorable. The firm's own research frames the better question as role fragility, not scoring distribution. A wide receiver can have a wide range of weekly outcomes and still be a bad roster bet if the underlying role could disappear with one coaching decision.
The tell is route participation, not the highlight
The number that separates a genuine riser from a lottery ticket is target share and route participation trend, tracked over multiple weeks, not a single big gain.
RotoWire's best ball receiver analysis uses Michael Wilson as the template. Wilson's targets per route run sat below 14.6% in each of his first two seasons, a rate that predicted almost nothing. It jumped to 20.6% last year, alongside an 8.6 yards per target average, a 66% catch rate, and an average depth of target north of 12 yards. That combination, efficient volume plus a route share that actually grew, is what turns a late pick into a real weekly contributor. A big single game with none of that underlying trend is a spike week with no follow up.
Profile one: built entirely on the spike
Tyquan Thornton is the clean example of a player who needs the spike to have any value at all. Per Rotoballer's July 2026 best ball sleeper roundup, Thornton posted a 7.1% target share and 38.1% route participation rate last season, against a 27.6 average depth of target. That aDOT is enormous for a role that thin. It means Thornton's fantasy output is concentrated almost entirely in whichever three or four weeks Kansas City's offense hits him deep. Outside of those weeks, the route share isn't there to produce a usable floor. Health notes on Patrick Mahomes II's scrambling ability this offseason only add more variance to an already spike dependent profile.
Profile two: a real role wearing a late price, with a complication
Troy Franklin looked like the seasonal riser story a month ago. Per Fantasy Life's best ball wide receiver piece, Franklin roughly doubled his target share and nearly tripled his yardage and touchdown production in his second season in Denver, building real chemistry with Bo Nix along the way. That's a role trend, not a highlight reel.
Here's the complication. Denver added Jaylen Waddle this offseason, which pushes Franklin behind two established options, Waddle and Courtland Sutton, in the target pecking order. A role that was expanding in isolation now has more competition for the same targets. The chemistry with Nix is real. Whether it survives a crowded receiver room is the actual research question, not a settled outcome.
A similar shape shows up with Josh Downs, whose target competition in Indianapolis has thinned enough that multiple outlets, including Rotoballer, describe him as finally positioned for a featured role rather than a committee share.
What the ADP can't tell you
Draft cost treats Thornton and Franklin as similar names in similar rounds. The underlying role data says they're different bets entirely. One is a low floor variance swing that needs a defense to fall asleep. The other is a target share trend fighting a July roster move that hasn't fully priced into the number yet.
Neither of those facts settles what you should do with your own picks. Which profile fits your roster construction is yours to evaluate.