Photo: Ian Johnson/Icon Sportswire

A new football season brings a fresh start and a reset on our perceptions of teams and players. This year, our favorite metric, Total Points, gets a reset as well, with a lot of enhancements that affect the past and future.

In short, Total Points is our answer to the question, “How much was each player worth in terms of points on the scoreboard?” It uses our charting data to attribute Expected Points Added to every player on nearly every play.

We last made substantive updates to Total Points in 2020. Over the last few years, we’ve learned a lot about our own data and added several new data points to our collection operation, especially in support of evaluating the passing game. Naturally, we want to imbue those learnings into our metrics, so every couple years we aim to update the Total Points methodology to account for this new information.

These updates aren’t just going forward, either. We’re applying as many of these changes as we can retroactively, even if most of the updates can’t be backfilled completely because they involve data that was only collected within the past few years. 

The nitty gritty details are integrated into our living primer that details the metric, but here are the highlights of the new and improved version:

Passing Game

• Building granular throw accuracy and openness information into our evaluation of each throw. This not only improves our estimates of expected performance, but also allows us to more accurately assign responsibility for the actual result.
• Incorporating metrics we’ve created over the last few years, like Expected On Target Rate and Expected Pressure Rate.
• Using quarterback drop type and snap-to-throw timing to evaluate pass blocking with more specificity.
• Because play action isn’t dictated by the players themselves and has an impact somewhat outside of the players’ performance, we are making an adjustment to pull back the offensive value on those plays. 

Running Game

• The tracking of Forced Blown Blocks allows us to precisely assign responsibility to defenders for blowing up a play, instead of using their initial alignment to make guided assumptions.
• Pre-snap safety depth affords more precision in evaluating how likely they are to make a tackle.
• Better measurement of the value of broken or missed tackles thanks to our tracking of the depth at which they occur.
• Adjusting yards before contact when there’s a missed tackle before first contact, leaning the credit for that run towards the rusher.

General

• Beyond the Total Points impact of this, we are also updating our Expected Points framework in one specific way. In general, that model doesn’t take into account time when judging the expected points scored from a particular game state, but to prevent situations where we overestimate in a late-half situation, we are considering plays inside the two minute warning as a separate category. 

Some examples of how our play-level evaluation changes

We’ll start with an example of a receiver getting more credit because of the tough play he had to make. On this play, Jalen Hurts heaves the ball downfield but doesn’t step into it, underthrowing the ball and forcing DeVonta Smith to come back towards the coverage to make the catch.

In the previous version of Total Points, Hurts was credited with 2.2 points compared to 1.8 for Smith. With the accuracy and openness info brought in, we’d now credit Hurts with 0.4—still a positive play—but give Smith a much more robust 2.7 for making this play.

Next, an example where the offensive line value changes as well, which is more subtle on a play-to-play basis. This is a deep dropback that Justin Herbert throws downfield with good accuracy but the receiver isn’t able to make the catch. 

Herbert’s value on this play increases by 0.3 points—a direct credit for the accurate throw, and an indirect credit because the ball came out quickly, earning the line less credit than we gave them previously. 

In the running game, we’re doing a lot better assigning responsibility when a player blows up a play. 

For example, on this play, Kwity Paye (51)—lined up way out wide as a pass rusher—blows right through the tackle’s outside shoulder and forces the running back to pause and adjust, resulting in no gain. 

In the past, we used each player’s initial alignment to try to split responsibility for the right tackle’s blown block, which resulted in some credit going to Julian Blackmon (32), who was lined up opposite the tackle but got caught in traffic early in the play. Now, we are assigning the appropriate credit for forcing the blown block to Paye.

Typically we use yards before contact to inform how well the offensive line did in setting up a clean lane for the rusher. Now, when a defender gets into the backfield and the rusher is able to elude the tackle without being contacted, we’re considering that the point of contact for the purpose of evaluating the offensive line. 

On this play where Joe Mixon is able to make an edge defender miss behind the line and get to the outside for a first down, we used to be giving about 50/50 credit to the line and to Mixon. But with this adjustment in place, Mixon is actually getting more than 100% of the credit for the first down and the line is being debited, because he could have easily been brought down for a loss on third down.

Our new NFL leaderboards

A lot of things changed here, both in college and the NFL, so we’re not going to go through everything. But it’s useful to know where we stand in terms of valuing players. So here are a handful of updated leaderboards for the 2023 season.

Quarterback Leaders

Jordan Love is getting a lot of, er, love heading into the 2024 season. After the Packers’ Week 6 bye, Love was a top five quarterback by almost any measure you can come up with, but Total Points is giving him as much credit as anyone just for the throws he attempted (which is calculated before even knowing whether the throw was catchable). We take a mix of the route, throw depth, and receiver openness for this, so he was making good decisions within the context of the scheme. He finished 4th overall in Total Points.

Dak Prescott falls off a bit for the opposite reason, although he’s kept in the top five because he gets great marks for his accuracy relative to the difficulty of his throws. He dropped from 2nd to 5th on this leaderboard.

Top Non-QB Offensive Players in 2023

We wrote not that long ago about the somewhat controversial conclusion by Total Points that Derrick Henry is still one of the most valuable non-quarterbacks in the league, but James Conner is assuredly the more surprising presence above, just 2 points from the top spot. 

Conner’s calling card is that he tied for the NFL lead in yards after contact per carry with De’Von Achane (with twice as many carries). Total Points splits rushing performance among the line and rusher, but after-contact productivity is mostly owned by the running back, so Conner (and Henry) more than make up for poor pre-contact performance.

Puka Nacua and CeeDee Lamb go in opposite directions with the update. We mentioned Dak Prescott’s accuracy before; that ends up pulling Lamb’s contribution down, while Matthew Stafford’s relative inaccuracy buoys Nacua.

Top Defenders in 2023

Big changes on the offensive side in the passing game lead to big changes on the defensive side as well, with safeties affected the most. Total Points has generally favored playmaking safeties quite a bit, so it might be a justifiable tempering of enthusiasm for that position. 

The big drops for coverage players are often coming from better assessment of throw accuracy and correspondingly how difficult the defender’s job was. Picture a key third down stop where the passer throws behind the receiver, making it very difficult to convert. The defender makes the tackle short of a first, but the off-target throw made that work a lot easier.

For pass rushers, employing the Pressures Above Expectation framework gives us a much better assessment of how well players get to the passer. Khalil Mack and Bradley Chubb are not alone among big gainers from this addition.

Where can I learn more?

We’ll be posting more “How Total Points Works” articles over the coming weeks, with specific breakdowns of why players are valued the way they are. And let us know @football_SIS if you have any requests!

If you want to check out individual players and positional leaderboards, the SIS DataHub has you covered. If you’re looking for more in-depth analysis, the SIS DataHub Pro could work even better for youl! We offer free trials when you sign up.

If you’re interested in more of the specifics of how the metric works, make sure to check out our primer, which is updated whenever the calculation changes.