Sticky vs. Unsticky Stats: Which Numbers Actually Predict?
Sticky vs. Unsticky Stats
Some stats are “sticky” — they repeat year to year and predict future production — while others are “unsticky,” mostly noise that regresses to the mean. Opportunity metrics (targets, carries, snaps) are sticky; efficiency and touchdown-rate metrics are largely unsticky. Knowing which is which lets you tell a real breakout from a mirage and trade ahead of the regression everyone else gets fooled by.
| Stat | Sticky (predictive)? | What it tells you |
|---|---|---|
| Target share / carries / snaps | Yes — very sticky | Opportunity; the foundation of production |
| Route participation | Yes | Pass-game involvement |
| Touchdown rate (TDs per opportunity) | No — unsticky | Regresses hard; luck-heavy |
| Yards per carry / per catch | Mostly no | Noisy; efficiency swings year to year |
| Red-zone conversion rate | No | Small samples, high variance |
Not all stats are created equal. Some describe things a player reliably does and will keep doing; others describe things that happened but won’t necessarily repeat. The difference between these two kinds of numbers — sticky and unsticky — is one of the most important concepts in fantasy analytics, because it separates the stats you can bet on from the ones that will fool you. Get this right and you’ll consistently buy low and sell high; get it wrong and you’ll chase noise.
What “Sticky” Means
A sticky stat is one that’s stable and repeatable — a player who posts it one year tends to post something similar the next, because it reflects a durable underlying reality (his role, his usage, his skill). Sticky stats are predictive: they tell you something about the future, not just the past.
An unsticky stat is volatile and largely random from year to year — a player who posts an extreme value is likely to regress toward the average next time, because the number was driven more by luck, variance, or a small sample than by a repeatable skill. Unsticky stats are descriptive: they tell you what happened, but not what will happen.
The trap is that both kinds of stats look equally real in a box score. A great touchdown total and a great target share both show up as production — but one will likely repeat and one will likely evaporate.
The Sticky Stats: Opportunity
The stickiest, most predictive stats are opportunity metrics — the volume a player is given:
- Target share and target volume (for receivers and pass-catching backs). How much of the offense flows through a player is remarkably stable, because it reflects his role and his coaching staff’s trust. High target share tends to repeat.
- Carries and snap share (for running backs). Workload is a role, and roles are sticky. A back getting 18 touches a game is likely to keep getting them barring an injury or depth-chart change.
- Route participation. How often a player runs a route on pass plays is a durable measure of pass-game involvement.
These are sticky because they’re decisions the team keeps making, not outcomes of chance. That’s why “opportunity is king” — it’s the foundation of production and it’s predictable.
The Unsticky Stats: Efficiency and Touchdowns
The least predictive stats are the ones driven by variance and small samples:
- Touchdown rate (touchdowns per opportunity). This is the classic unsticky stat. A player who scored on an unusually high share of his touches is very likely to regress — touchdowns are relatively rare and luck-heavy, so an extreme rate almost never repeats. A player’s opportunity near the goal line is stickier than his conversion of it.
- Yards per carry / yards per reception. Efficiency bounces around year to year far more than volume does. A career year in yards per touch is often a red flag for regression, not a new baseline.
- Red-zone conversion rate. Small samples plus high variance make this noisy.
The danger: a player can post a monster fantasy season built largely on unsticky stats — a spike in touchdown rate and efficiency on ordinary volume — and look like a new star, when he’s actually a regression candidate whose production won’t hold.
How to Use Stickiness
This concept is the engine behind buying low and selling high:
- Sell players whose production is built on unsticky stats. A back coming off a career year fueled by an unsustainable touchdown rate is a sell-high — his name value is peaking, but the numbers that created it will regress. Trade him while the market prices the career year as the new normal.
- Buy players whose sticky stats are strong but whose unsticky stats were unlucky. A receiver with an elite target share who scored on a low percentage of his looks is a buy-low — the sticky foundation (volume) is there, and positive touchdown regression is coming. His down year was bad luck on a good role.
- Trust breakouts backed by sticky stats. A hot start built on a genuine jump in targets or carries is real; one built on touchdowns and efficiency is a mirage. (This is exactly why opportunity confirms a breakout faster than points do.)
The market constantly overreacts to unsticky production and underreacts to sticky opportunity — which is your edge.
Worked Example: Two Career Years
Two running backs post career-best fantasy seasons. Back A did it on a huge workload — 320 touches, a locked-in featured role — with an ordinary touchdown rate. Back B did it on modest volume (200 touches) but a career-high touchdown rate and a fluky yards-per-carry spike.
Their fantasy finishes look similar, but their outlooks are opposite. Back A’s production rests on a sticky stat (volume), so it’s likely to repeat — hold or even buy. Back B’s rests on unsticky stats (touchdown rate, efficiency) that almost always regress, so his career year is a sell-high — the market will price him as a new stud, and you should trade him to that market before the touchdowns dry up. A manager who can’t tell sticky from unsticky sees two similar seasons; one who can sees a hold and a sell.
Common mistake: treating all production as equally predictive and buying (or holding) players coming off career years without asking what drove it. Production built on sticky opportunity stats — targets, carries, snaps — tends to repeat and is worth paying for; production built on unsticky touchdown rate and efficiency almost always regresses and is a sell-high. Look under the fantasy total: strong sticky stats plus unlucky unsticky stats is a buy-low; weak sticky stats propped up by lucky unsticky stats is a sell-high.
The Bottom Line
Fantasy stats split into sticky (predictive, repeatable) and unsticky (noisy, regression-prone). Opportunity metrics — target share, carries, snaps, routes — are sticky and are the foundation you can bet on. Efficiency and touchdown-rate metrics are unsticky and will fool you if you treat a lucky spike as a new baseline. Use the distinction to buy players with strong sticky stats and unlucky unsticky ones, and sell players whose big seasons rest on unsticky production about to regress. The market keeps chasing the noise — knowing which stats stick is how you get ahead of it.
For more on opportunity, regression, and evaluating stats, continue with when to trust a breakout, target share, and sell high strategy.