Our Blog


Predicting Multi-Touchdown Games in NFL DFS, Part 3

This is Part 3 of a series on predicting multi-touchdown games in DFS. In Part 1 and Part 2 we looked at every instance of a multi-TD game in the last two years and how those performances related to salaries, projections, Bargain Ratings, Opponent Plus/Minus values, and a variety of Vegas data.

In this final part, we’re going to look at good ‘ol football metrics. Of note, we’ll look at the following statistics of multi-TD performers and how they compare to league-average rates:

•  Average Rushes per Game
•  Average Rush Yards per Game
•  Average Yards per Carry
•  Average Rush Touchdowns per Game
•  Average Receiving Targets per Game
•  Average Receptions per Game
•  Average Receiving Yards per Game
•  Average Yards per Reception
•  Average Receiving Touchdowns per Game
•  Average Yards per Receiving Target

Since the first several of those are applicable to only running backs, let’s start there.

Running Backs

For the rushing metrics listed above, here’s the data on our multi-TD RB sample compared to 2015 league-average RB rates:

Avg Rushes Rush Yds Unweighted Yards/Carry Weighted Yards/Carry Rush TDs
NFL 8.29 34.50 4.05 4.16 0.22
RB 10.55 48.52 4.31 4.60 0.34

In this first table, there isn’t groundbreaking data. Players in the multi-TD sample were ones who saw more volume going into their multi-TD game than the average RB. One interesting thing to note before we get into some actual actionable RB data is that efficiency for these backs was up compared to league average. There’s a current trend in the NFL DFS community to highlight just how much more opportunity matters than efficiency or even talent. And that’s still true. However, for guys with extreme, multi-TD upside, perhaps we’re missing the middle ground; perhaps for the truly ‘upside’ plays we need to find players who are efficient and get opportunity.

Here’s the RB data for the passing metrics:

Targets Receptions Rec Yards  Unweighted Yards/Reception Weighted Yards/Reception Receiving TD Unweighted Yards/Target Weighted Yards/Target
NFL 2.51 1.88 15.98 8.32 8.50 0.08 6.20 6.37
RB 2.31 1.62 16.93 8.06 10.45 0.09 4.65 7.33

It seems that the data is pointing toward running backs who get opportunities and are efficient in the run game. Those are the backs who have typically had multi-TD games in the past couple of years — not backs with upside in the passing game. Those guys are certainly valuable on DraftKings because of the PPR-style scoring; in fact, of viable running backs last year, pass-catching specialists Danny WoodheadBilal Powell, and Theo Riddick all ranked top-eight in total DraftKings Plus/Minus.The RBs with multi-TD games were far less reliant on the passing game than the whole sample of RBs from 2015. They averaged fewer receptions, yards per reception, and yards per target on average than the overall sample of RBs.

passcatchers1

Woodhead is a bit of an exception because of his red-zone work in San Diego’s offense, but these guys are valuable because of their low salaries and DK’s scoring, but they aren’t huge-upside plays. Knowing the difference between those two things is huge and should affect how you create rosters in large-field GPPs.

Tight Ends

Here’s the data on the multi-TD tight end sample compared to league-average tight end rates.

Targets Receptions Rec Yards  Unweighted Yards/Reception Weighted Yards/Reception Receiving TD Unweighted Yards/Target Weighted Yards/Target
NFL 4.21 2.76 31.02 11.12 11.24 0.25 7.20 7.37
TE 4.94 3.16 42.06 11.63 13.31 0.43 7.27 8.51

I ran a trend and found that only 187 of 2,199 tight ends in the last two years went into a game with a TD/G mark of at least .43, which is what the TEs in the multi-TD sample averaged. Put simply: Paying up for tight ends isn’t sexy, but those guys — regardless of matchup or whether they’re a huge favorite or dog — are the ones who score multiple touchdowns in a game. The data here matches up with our major findings in Part 2: Tight ends are so touchdown-dependent that other variables — Vegas lines and spreads, for example — should be ignored. Tight ends in our multi-TD sample on average went into their multi-TD games with more targets, receptions, receiving yards, and almost double the amount of touchdowns per game compared to the overall sample of TEs. They were also superior in the ‘efficiency’ metrics — yards per reception and yards per target.

For what it’s worth, there are four TEs who are going into Week 1 with a TD/G rate of .43 or higher: Rob Gronkowski, Jordan Reed, Gary Barnidge, and Richard Rodgers.

Wide Receivers

The WR data is informative: The WRs in the multi-TD sample outperformed the league-average WR in terms of targets and receptions but not significantly. However, the WRs in our sample drastically outperformed the average WR in terms of receiving yards per game, receiving touchdowns per game, and yards per target.

Targets Receptions Rec Yards  Unweighted Yards/Reception Weighted Yards/Reception Receiving TD Unweighted Yards/Target Weighted Yards/Target
NFL 5.43 3.28 43.05 13.31 13.13 0.27 7.94 7.93
WR 6.86 3.99 61.15 13.89 15.33 0.43 8.27 8.91

Conclusion: This probably isn’t groundbreaking, but if you’re looking for WRs with multiple-touchdown upside, you’re better off investing in players who do a lot with their targets and not necessarily ones who simply receive a lot of them. Think of the difference between Jarvis Landry and Eric Decker. The former averaged nearly seven receptions per game last year compared to Decker’s five. However, Decker had a higher yards-per-target mark — 7.78 versus Landry’s 6.97 — and was much more of a red-zone threat. The Landry-Decker example is extreme, as Landry is an elite receptions and Decker is an elite red-zone guy, but the point still stands: Target WRs who can actually . . . you know . . . score touchdowns.

The Summation

I hope this study has been informative. To conclude, I’ll list some actionable takeaways from our three parts. If you want players with multi-TD upside in your lineup, you should . . .

• Emphasize volume over price or matchup for running backs.
• Emphasize talent over all else for wide receivers.
• Potentially fade running backs and wide receivers with great matchups. Or at least don’t be scared to roster those positions even when they don’t have great matchups.
• Target tight ends with a high TD rate regardless of other variables, specifically ‘poor’ Vegas data.
• Focus on traditional running backs as opposed to pass-catching specialists.
• Emphasize yards-per-target and red-zone work for wide receivers as opposed to opportunity metrics like receptions and targets.

This is Part 3 of a series on predicting multi-touchdown games in DFS. In Part 1 and Part 2 we looked at every instance of a multi-TD game in the last two years and how those performances related to salaries, projections, Bargain Ratings, Opponent Plus/Minus values, and a variety of Vegas data.

In this final part, we’re going to look at good ‘ol football metrics. Of note, we’ll look at the following statistics of multi-TD performers and how they compare to league-average rates:

•  Average Rushes per Game
•  Average Rush Yards per Game
•  Average Yards per Carry
•  Average Rush Touchdowns per Game
•  Average Receiving Targets per Game
•  Average Receptions per Game
•  Average Receiving Yards per Game
•  Average Yards per Reception
•  Average Receiving Touchdowns per Game
•  Average Yards per Receiving Target

Since the first several of those are applicable to only running backs, let’s start there.

Running Backs

For the rushing metrics listed above, here’s the data on our multi-TD RB sample compared to 2015 league-average RB rates:

Avg Rushes Rush Yds Unweighted Yards/Carry Weighted Yards/Carry Rush TDs
NFL 8.29 34.50 4.05 4.16 0.22
RB 10.55 48.52 4.31 4.60 0.34

In this first table, there isn’t groundbreaking data. Players in the multi-TD sample were ones who saw more volume going into their multi-TD game than the average RB. One interesting thing to note before we get into some actual actionable RB data is that efficiency for these backs was up compared to league average. There’s a current trend in the NFL DFS community to highlight just how much more opportunity matters than efficiency or even talent. And that’s still true. However, for guys with extreme, multi-TD upside, perhaps we’re missing the middle ground; perhaps for the truly ‘upside’ plays we need to find players who are efficient and get opportunity.

Here’s the RB data for the passing metrics:

Targets Receptions Rec Yards  Unweighted Yards/Reception Weighted Yards/Reception Receiving TD Unweighted Yards/Target Weighted Yards/Target
NFL 2.51 1.88 15.98 8.32 8.50 0.08 6.20 6.37
RB 2.31 1.62 16.93 8.06 10.45 0.09 4.65 7.33

It seems that the data is pointing toward running backs who get opportunities and are efficient in the run game. Those are the backs who have typically had multi-TD games in the past couple of years — not backs with upside in the passing game. Those guys are certainly valuable on DraftKings because of the PPR-style scoring; in fact, of viable running backs last year, pass-catching specialists Danny WoodheadBilal Powell, and Theo Riddick all ranked top-eight in total DraftKings Plus/Minus.The RBs with multi-TD games were far less reliant on the passing game than the whole sample of RBs from 2015. They averaged fewer receptions, yards per reception, and yards per target on average than the overall sample of RBs.

passcatchers1

Woodhead is a bit of an exception because of his red-zone work in San Diego’s offense, but these guys are valuable because of their low salaries and DK’s scoring, but they aren’t huge-upside plays. Knowing the difference between those two things is huge and should affect how you create rosters in large-field GPPs.

Tight Ends

Here’s the data on the multi-TD tight end sample compared to league-average tight end rates.

Targets Receptions Rec Yards  Unweighted Yards/Reception Weighted Yards/Reception Receiving TD Unweighted Yards/Target Weighted Yards/Target
NFL 4.21 2.76 31.02 11.12 11.24 0.25 7.20 7.37
TE 4.94 3.16 42.06 11.63 13.31 0.43 7.27 8.51

I ran a trend and found that only 187 of 2,199 tight ends in the last two years went into a game with a TD/G mark of at least .43, which is what the TEs in the multi-TD sample averaged. Put simply: Paying up for tight ends isn’t sexy, but those guys — regardless of matchup or whether they’re a huge favorite or dog — are the ones who score multiple touchdowns in a game. The data here matches up with our major findings in Part 2: Tight ends are so touchdown-dependent that other variables — Vegas lines and spreads, for example — should be ignored. Tight ends in our multi-TD sample on average went into their multi-TD games with more targets, receptions, receiving yards, and almost double the amount of touchdowns per game compared to the overall sample of TEs. They were also superior in the ‘efficiency’ metrics — yards per reception and yards per target.

For what it’s worth, there are four TEs who are going into Week 1 with a TD/G rate of .43 or higher: Rob Gronkowski, Jordan Reed, Gary Barnidge, and Richard Rodgers.

Wide Receivers

The WR data is informative: The WRs in the multi-TD sample outperformed the league-average WR in terms of targets and receptions but not significantly. However, the WRs in our sample drastically outperformed the average WR in terms of receiving yards per game, receiving touchdowns per game, and yards per target.

Targets Receptions Rec Yards  Unweighted Yards/Reception Weighted Yards/Reception Receiving TD Unweighted Yards/Target Weighted Yards/Target
NFL 5.43 3.28 43.05 13.31 13.13 0.27 7.94 7.93
WR 6.86 3.99 61.15 13.89 15.33 0.43 8.27 8.91

Conclusion: This probably isn’t groundbreaking, but if you’re looking for WRs with multiple-touchdown upside, you’re better off investing in players who do a lot with their targets and not necessarily ones who simply receive a lot of them. Think of the difference between Jarvis Landry and Eric Decker. The former averaged nearly seven receptions per game last year compared to Decker’s five. However, Decker had a higher yards-per-target mark — 7.78 versus Landry’s 6.97 — and was much more of a red-zone threat. The Landry-Decker example is extreme, as Landry is an elite receptions and Decker is an elite red-zone guy, but the point still stands: Target WRs who can actually . . . you know . . . score touchdowns.

The Summation

I hope this study has been informative. To conclude, I’ll list some actionable takeaways from our three parts. If you want players with multi-TD upside in your lineup, you should . . .

• Emphasize volume over price or matchup for running backs.
• Emphasize talent over all else for wide receivers.
• Potentially fade running backs and wide receivers with great matchups. Or at least don’t be scared to roster those positions even when they don’t have great matchups.
• Target tight ends with a high TD rate regardless of other variables, specifically ‘poor’ Vegas data.
• Focus on traditional running backs as opposed to pass-catching specialists.
• Emphasize yards-per-target and red-zone work for wide receivers as opposed to opportunity metrics like receptions and targets.