NBA Trend Testing: Vegas Line Movement

With our Trends tool, you can see current and historical matches for players in matchups that meet the specified criteria. This makes it pretty easy to track performance within the result set. In this series, I thought it might be cool to take it one step further. I will be creating a Trend early in the week, playing the “Current Matches” in my lineups throughout the week on FanDuel and then reviewing the Trend at the end of the week.

This Monday, I created the following Trend:

Description

Vegas Line Movement is a tool that a lot of players use in NFL DFS.   There are some key differences between NFL and NBA scoring, however, that could make a difference in how effective it is in NBA DFS. In NFL, the scoring is such that some scores are going to be much more common than others – 21 is a much more likely score than 22 for an NFL team to land on, for example.

So when a Vegas line moves across a key number like this, there is significance. If a line moves from 19 to 20, that doesn’t mean as much as one that moves from 20.5 or 21.5. In NBA though, for the most part, each score is as likely as the next because the scoring system is much more linear. With this trend, I wanted to track NBA teams whose projected point total had moved up by 1.5 points or more to see how helpful this would be on the NBA side.

vegasmov1

The filters I used were:

  • The player’s salary is over $6,000
  • The Vegas line change is over +1.5

Results

The screenshots below are from teams I entered into FanDuel’s $5 Layup.

12/7

vegasmov2

 

Okay, one note before we get started. This trend provided very uneven matches – one night, there were seven matches and for the next two nights, there were zero matches. In the interest of this article, I took the players from the team that saw the largest positive Vegas line adjustment each night. Please note that this does differ slightly from the trend’s parameters.

On this night, Chris Paul was questionable (and by my interpretation of the news that day, closer to doubtful) at lineup lock, so I did not roster him. He did end up playing, but had a pretty unimpressive line.

The Clippers came into this game having won five of six games while the Wolves were trending in the opposite direction. When lines opened, the Clippers were slim favorites, but expanded to four-point favorites by tipoff.

This was a nine-game slate and Blake ended up being the most highly-owned power forward in this GPP. Truthfully, I think Blake’s ownership was elevated more so due to Chris Paul’s anticipated absence, leading to increased usage than anything else. Blake slightly underperformed, while Pau Gasol vs Phoenix (12.3% owned, 50 points) and Jon Leuer @ Chicago (13.6% owned, 25.6 points on $4k salary) were better plays.

DeAndre Jordan, however, was rostered on each of the top four finishing teams in this contest. Minnesota ranks in the bottom 10 in the league in terms of blocks and rebounds allowed to their opponents and DeAndre feasted on this night. Andre Drummond was owned by 30.1% of the teams and posted a relative dud of 26.3 fantasy points against the Hornets. Interestingly enough, the Pistons has a +1.7 Vegas line change in their contest this night.

12/8

On this six-game night, the Warriors had the highest positive Vegas line change prior to tipoff of just +1, which brought their team total up to 110.3 points. As it turned out, Vegas could have gone about 10 points higher.

vegasmov3

 

There were a few cheaper point guards who out-produced Steph from a points per dollar standpoint, but even so, he was rostered on the majority of the teams that finished in the top-10 teams. Klay was on every team that finished in the top 10.

The spread was a somewhat surprising 5.5 points for Warriors @ Pacers and Golden State ended up winning in a 131-123 barn burner. This game actually looked like it was going to turn into a blowout and Steph and Klay were rested for much of the fourth quarter. Indiana, however, outscored Golden State 40-20 in the fourth quarter and both Curry and Klay came back into the game for the final couple of minutes to rack up a few extra points.

The Vegas line movement was only one of many positive factors the Warriors had going for them in this game, so it’s hard to give a whole lot of credit here. In fact, Steph was rated as the #3 point guard and Klay was the #1 rated shooting guard using Phan’s Model on 12/8.

12/9

vegasmov5

 

The Pistons added 1.3 points to their Vegas projection after lines initially opened and were favored against the Grizzlies this night. It feels weird going hard against the Grizzlies in DFS and I’m guessing the line movement on this night had a lot to do with uncertainty regarding whether Marc Gasol was going to be able to play. He had injured his ankle in Memphis’ previous game, but he did end up taking the court against Detroit after all.

Gasol wasn’t enough to slow down Drummond, who exceeded his implied point total by about 10.5 points. The matchup did cut his ownership in half – he was over 30% for Monday’s contest against the Hornets. Unfortunately, Karl-Anthony Towns (vs LAL) and Alex Len (vs ORL) also went over 40 fantasy points at much lower salaries.

I think people were on Reggie Jackson mostly because he was cheap here. Obviously, it was a pace-down game for Detroit and Mike Conley has been a more or less neutral matchup for opponents this season. Due to excellent salary ratings and maybe a little help from the Vegas line, Reggie was the top-rated point guard in Phan’s Model and ended up exceeding his expectation by over four points. None of that helped me with this team though, as John Wall was owned by nearly 40% of the field (vs HOU) and scored 58.8.

This was a matchup I was not crazy about at all, but I actually came out a bit ahead from a value perspective with these two plays.

12/10

vegasmov4

 

This game went sideways in just about every way possible. The Clippers’ projected total rose 2.6 points up to 102.5 before the game, so the 80 total points they scored against Chicago were very disappointing. Additionally, Blake Griffin was ejected midway through the third quarter and Chris Paul laid an egg against Derrick Rose.

There were only four games on the docket on 12/10, so seeing Blake at 22.2% was actually a little low compared to what I was expecting. He was the top-rated PF play in Phan’s Model and the second-rated play overall. If he had played the whole game, he would likely have crushed his implied total of 36.5.

The Bulls have a +2.0 Opponent Plus/Minus Rating against centers this year. That, in combination with the high Vegas total, was not enough for DeAndre to exceed his 34.8 implied total. Meanwhile, Chris Paul is clearly still bothered by his recent rib injury and in fact will not be playing on 12/11. Vegas was way off in this game and that left me holding a couple big ticket items with minimal return on some of my teams.

All three of the Clippers missed value and this was a pretty low scoring DFS night in general. The first place finisher in this GPP put up just 323 points.

Review

 

Performance this week was very up and down. Overall, five of nine plays identified by the trend exceeded their implied point total. This trend is very limiting on its own and you can get yourself in trouble sometimes by relying too much on Vegas as shown by the Clippers/Bulls game.

I tried revising this trend to add a minimum threshold on the Vegas point total. With the above criteria, we are matching ANY movement of +2, even if the total moves from 80 to 82. I added a filter to only find games with Vegas movement where the Vegas score was over 102 and the results were still not very good:

vegasmov6

 

Vegas line movement doesn’t appear to be a very good predictive tool for DFS purposes in NBA – at least not compared to NFL DFS. I’m definitely not saying you should totally ignore the category, but instead use it to “like a little bit more” matchups that you already like rather than as a high level tool to identify players to roster.

With our Trends tool, you can see current and historical matches for players in matchups that meet the specified criteria. This makes it pretty easy to track performance within the result set. In this series, I thought it might be cool to take it one step further. I will be creating a Trend early in the week, playing the “Current Matches” in my lineups throughout the week on FanDuel and then reviewing the Trend at the end of the week.

This Monday, I created the following Trend:

Description

Vegas Line Movement is a tool that a lot of players use in NFL DFS.   There are some key differences between NFL and NBA scoring, however, that could make a difference in how effective it is in NBA DFS. In NFL, the scoring is such that some scores are going to be much more common than others – 21 is a much more likely score than 22 for an NFL team to land on, for example.

So when a Vegas line moves across a key number like this, there is significance. If a line moves from 19 to 20, that doesn’t mean as much as one that moves from 20.5 or 21.5. In NBA though, for the most part, each score is as likely as the next because the scoring system is much more linear. With this trend, I wanted to track NBA teams whose projected point total had moved up by 1.5 points or more to see how helpful this would be on the NBA side.

vegasmov1

The filters I used were:

  • The player’s salary is over $6,000
  • The Vegas line change is over +1.5

Results

The screenshots below are from teams I entered into FanDuel’s $5 Layup.

12/7

vegasmov2

 

Okay, one note before we get started. This trend provided very uneven matches – one night, there were seven matches and for the next two nights, there were zero matches. In the interest of this article, I took the players from the team that saw the largest positive Vegas line adjustment each night. Please note that this does differ slightly from the trend’s parameters.

On this night, Chris Paul was questionable (and by my interpretation of the news that day, closer to doubtful) at lineup lock, so I did not roster him. He did end up playing, but had a pretty unimpressive line.

The Clippers came into this game having won five of six games while the Wolves were trending in the opposite direction. When lines opened, the Clippers were slim favorites, but expanded to four-point favorites by tipoff.

This was a nine-game slate and Blake ended up being the most highly-owned power forward in this GPP. Truthfully, I think Blake’s ownership was elevated more so due to Chris Paul’s anticipated absence, leading to increased usage than anything else. Blake slightly underperformed, while Pau Gasol vs Phoenix (12.3% owned, 50 points) and Jon Leuer @ Chicago (13.6% owned, 25.6 points on $4k salary) were better plays.

DeAndre Jordan, however, was rostered on each of the top four finishing teams in this contest. Minnesota ranks in the bottom 10 in the league in terms of blocks and rebounds allowed to their opponents and DeAndre feasted on this night. Andre Drummond was owned by 30.1% of the teams and posted a relative dud of 26.3 fantasy points against the Hornets. Interestingly enough, the Pistons has a +1.7 Vegas line change in their contest this night.

12/8

On this six-game night, the Warriors had the highest positive Vegas line change prior to tipoff of just +1, which brought their team total up to 110.3 points. As it turned out, Vegas could have gone about 10 points higher.

vegasmov3

 

There were a few cheaper point guards who out-produced Steph from a points per dollar standpoint, but even so, he was rostered on the majority of the teams that finished in the top-10 teams. Klay was on every team that finished in the top 10.

The spread was a somewhat surprising 5.5 points for Warriors @ Pacers and Golden State ended up winning in a 131-123 barn burner. This game actually looked like it was going to turn into a blowout and Steph and Klay were rested for much of the fourth quarter. Indiana, however, outscored Golden State 40-20 in the fourth quarter and both Curry and Klay came back into the game for the final couple of minutes to rack up a few extra points.

The Vegas line movement was only one of many positive factors the Warriors had going for them in this game, so it’s hard to give a whole lot of credit here. In fact, Steph was rated as the #3 point guard and Klay was the #1 rated shooting guard using Phan’s Model on 12/8.

12/9

vegasmov5

 

The Pistons added 1.3 points to their Vegas projection after lines initially opened and were favored against the Grizzlies this night. It feels weird going hard against the Grizzlies in DFS and I’m guessing the line movement on this night had a lot to do with uncertainty regarding whether Marc Gasol was going to be able to play. He had injured his ankle in Memphis’ previous game, but he did end up taking the court against Detroit after all.

Gasol wasn’t enough to slow down Drummond, who exceeded his implied point total by about 10.5 points. The matchup did cut his ownership in half – he was over 30% for Monday’s contest against the Hornets. Unfortunately, Karl-Anthony Towns (vs LAL) and Alex Len (vs ORL) also went over 40 fantasy points at much lower salaries.

I think people were on Reggie Jackson mostly because he was cheap here. Obviously, it was a pace-down game for Detroit and Mike Conley has been a more or less neutral matchup for opponents this season. Due to excellent salary ratings and maybe a little help from the Vegas line, Reggie was the top-rated point guard in Phan’s Model and ended up exceeding his expectation by over four points. None of that helped me with this team though, as John Wall was owned by nearly 40% of the field (vs HOU) and scored 58.8.

This was a matchup I was not crazy about at all, but I actually came out a bit ahead from a value perspective with these two plays.

12/10

vegasmov4

 

This game went sideways in just about every way possible. The Clippers’ projected total rose 2.6 points up to 102.5 before the game, so the 80 total points they scored against Chicago were very disappointing. Additionally, Blake Griffin was ejected midway through the third quarter and Chris Paul laid an egg against Derrick Rose.

There were only four games on the docket on 12/10, so seeing Blake at 22.2% was actually a little low compared to what I was expecting. He was the top-rated PF play in Phan’s Model and the second-rated play overall. If he had played the whole game, he would likely have crushed his implied total of 36.5.

The Bulls have a +2.0 Opponent Plus/Minus Rating against centers this year. That, in combination with the high Vegas total, was not enough for DeAndre to exceed his 34.8 implied total. Meanwhile, Chris Paul is clearly still bothered by his recent rib injury and in fact will not be playing on 12/11. Vegas was way off in this game and that left me holding a couple big ticket items with minimal return on some of my teams.

All three of the Clippers missed value and this was a pretty low scoring DFS night in general. The first place finisher in this GPP put up just 323 points.

Review

 

Performance this week was very up and down. Overall, five of nine plays identified by the trend exceeded their implied point total. This trend is very limiting on its own and you can get yourself in trouble sometimes by relying too much on Vegas as shown by the Clippers/Bulls game.

I tried revising this trend to add a minimum threshold on the Vegas point total. With the above criteria, we are matching ANY movement of +2, even if the total moves from 80 to 82. I added a filter to only find games with Vegas movement where the Vegas score was over 102 and the results were still not very good:

vegasmov6

 

Vegas line movement doesn’t appear to be a very good predictive tool for DFS purposes in NBA – at least not compared to NFL DFS. I’m definitely not saying you should totally ignore the category, but instead use it to “like a little bit more” matchups that you already like rather than as a high level tool to identify players to roster.