Chris Allen examines how success rate is important to measuring the best offenses and players to target for fantasy football.

I’m skeptical of any product that claims to be “all-purpose.”

On the one hand, the engineering side of my brain applauds the efficiency. One thing to cover a multitude of uses is the type of ingenuity we need. My (semi) lazy side agrees. But on the other hand, it doesn’t leave room for any options.

Think about “one-size-fits-all” clothes. First of all, it should be “one-size-fits-most,” but I’ll leave that alone. Second, there’s no understanding of how I want to wear the item. All-purpose seasoning is usually just salt, pepper, and a few other common herbs and spices. I can’t apply that to everything. I’m looking for something slightly different every meal.

Success rate in football builds on the same idea. Each snap gives the offense a chance to accomplish something. However, not every play will result in a touchdown. Every concept has a separate purpose. In either case, understanding what a positive play is and which teams execute them more often can be a valuable skill in assembling a strong fantasy roster.

The Definition of Success Rate

“Success rate measures how often an offense generates a positive play.” 

Admittedly, this definition deviates from our usual simplistic setup. But, to be fair, I (kind of) did it on purpose.

There’s a level of subjectivity in my explanation. “Positive play” could mean whatever you want it to be. As I mentioned in the intro, some might view success as the touches that put points on the scoreboard. Others could see explosive runs or receptions that flip the field as successful. We could set the bar wherever we want!

But, for the sake of argument, let’s say a success is any snap with at least one yard gained. Technically, a yard forward is a positive gain. Anyway, by our logic (yes, you’re in on this ridiculously low threshold, too), we would grade one of Christian McCaffrey’s third-quarter runs in Week 11 as a successful one. 

We’ve already established I’m not a coach or former player of the sport at any respectable level. But I’ll be the first to say what you’re thinking. That play doesn’t look like a success. Ernest Jones instantly meets McCaffrey at the hole and brings him down. 

Side note: if you see more defenders than offensive players clapping or making encouraging gestures (see Devon Witherspoon clapping after the dust settles), then the offense probably didn’t do their job right.

More importantly, the run had no significant impact on the drive. On the next snap, Brock Purdy scrambled to make third down more manageable. It took nine more plays for the 49ers to score. I doubt anyone would say McCaffrey’s one-yard rush was the key moment of the drive.  

Well, OK then. Maybe having one yard as the minimum was too easy. Any skill player should at least be able to fall forward. I suppose the takeaway is that the ball moving forward a yard isn’t that important. But then again, that same distance can decide a game.

First off, I apologize to Giants fans for digging up the past. After McCaffrey (allegedly) picked up the first down, all Cam Newton had to do was spike the ball. Carolina won. Therefore, if I wanted to determine whether the third-down run was successful or not, I could simply look at what followed. And I’d say securing a win is a positive outcome! But if that one-yard gain doesn’t sway you, I’m sure the most fantasy-relevant three-foot plunge will.

McCaffrey’s goal-line totes, the Tush Push, or anytime a defender commits pass interference in the end zone, we know what’s coming: fantasy points. However, critically extending or cutting into a lead is the clearest form of success. Another way to look at this is that yardage isn’t the only factor at play.

We went through a similar exercise when we broke down expected points added (EPA). Game situation matters. Clearly, a one-yard run on first down isn’t as important as the same type of carry in the red zone or to convert a series. Accordingly, when you apply the same contextual framework to success rate, we can standardize what offenses need to accomplish on a down-to-down basis.

  • First Down: at least 40.0% (of yards needed for first down)
  • Second Down: 60.0%
  • Third or Fourth Down: 100.0% (i.e., a first down) or a touchdown

Let’s revisit the first McCaffrey clip. It was 1st and 10. If he had plowed forward for four yards (40.0% of 10) instead, the play would have been successful. Or, put another way, Purdy might not have had to go into hero mode on the next play to get the offense back on track. Regardless, we’ve got another efficiency metric on our hands, but its importance is slightly different than what we’ve seen before.

Why is Success Rate Important?

Wait, now let me get this straight.

We have a stat that gauges how efficient an offense is, which also takes game-situational context into account. I hinted at it earlier, but this sounds like EPA. That means we’ve already done this. We’ve shown that the number of attempts left to convert a series, as well as the ball's position on the field, matters. Subsequently, you’d think that if the same logic went into the metric, it’d have a similar (or better) connection to fantasy points.

 

Seeing relatively close R-squared values of 0.61 and 0.70 was a positive sanity check. But it simultaneously provides a common takeaway:

“If EPA has a stronger correlation to fantasy points, success rate isn’t as useful.”

That’s fair. I’ve consistently advocated for using stats that matter more for our purposes. However, fantasy football isn’t about a single season, like the data in the charts. In some cases, we make decisions now with the hope that they’ll pay off down the road. And that’s where success rate plays a larger role. I’ll let my quarterback explain.

Sidenote: I knew Cincinnati was going to lose midway through the second quarter. Their defense (or lack thereof) took years off my life.

Anyway, EPA and success rate will tell us the same thing. It was a good play! A touchdown, especially one on fourth down from 40-plus yards away, defies expectations, given the situational context. Accordingly, the result would make the pass a successful one. So, let’s expand the analysis to the entire drive.

Two short gains. An incomplete pass. And then, boom, magic. Of course, I loved seeing Joe Burrow and Tee Higgins do their thing. But the nerd in me knows the volatility in a sequence like this. And so does just one of the stats in the two right columns.

If I averaged all four plays, I could say the Bengals averaged 1.08 EPA per drive. Yes, it’s from one series. Even better, it’s more than their seasonal average (0.5). But, more importantly, you can see how they outkicked their normal. The TD throw swings the average. Meanwhile, success rate tells a different story.

Chase Brown’s reception didn’t generate at least four yards (or 40.0%) to be successful. The incomplete on third was a dud, too. So, we’ve got two good plays and two bad, resulting in a 50.0% success rate for the drive. Coincidentally, their season-long success rate was 50.2%. 

There are two takeaways here. First, EPA is like grading with extra credit. Big plays are a bonus, which the stat awards. However, there’s a reason why the average over 18 weeks aligns with that single drive for success rate, rather than the EPA. Success rate is a stronger indicator of future efficiency.

Going all the way back to the beginning, playcallers don’t design every concept with the idea that it’ll get their team into the end zone. Sometimes, just four yards (on first down) will do. Accordingly, any player or offense that can consistently generate successful plays should continue to do so, unless there is an injury or a scheme shift. To test this idea, I reviewed the last five years and examined the year-over-year stability between EPA and success rate.

  • Success Rate: 0.39 (R-squared)
  • EPA per Dropback: 0.27

Ultimately, we’ve got two offensive efficiency metrics. But they’re telling us two distinct things. EPA measures how much each play contributes. Success rate evaluates how often good plays occur. With that understanding, you can use it to your advantage.

How Can You Use Success Rate?

Using the two stats together, you can weed out the fakes and spot underrated players. Let’s look at an example of fool’s gold from last season.

HC Mike Tomlin’s decision to bench Justin Fields in favor of Russell Wilson looked like a stroke of genius for the first month. However, plays like the above were a primary part of the Wilson experience. Nearly 40.0% of his passing yards were on throws of 15 air yards or more. He was sixth in TDs from a similar distance. You’d get a similar sense of unsustainability looking at his EPA and success rate:

  • EPA per dropback (Weeks 7-13): 7th (out of 28 qualifying QBs – min. 150 dropbacks) 
  • Success Rate: 19th 

The down-to-down consistency was lacking. Explosives, like the many against my Bengals, were papering over the flaws in the passing game. And with Wilson’s deficiencies and Pittsburgh’s supporting cast, his downfall wasn’t surprising. But it’s through the use of success rate that we can see these types of landmines before we step on them.