Data Driven DFS: Plays and Visualizations of the Night (11/1)
Welcome back to Data Driven DFS, your most inconsistent source for NBA DFS analysis!
Not enough time write out my game-by-game analysis tonight, so I am just going to accentuate some spots. We are going to get real data-sciencey in this one.
A new way to judge defenses:
I have a script that loops through every game that has been played this year and spits out a number for each team about how the stats they are allowing compare to their opponent’s average output on the year. So if the Bulls give up 250 fantasy points to the Celtics and the Celtics season average to that point was 240, that’s a +10 for the Bulls. It looks at every game each team has played and averages those numbers out. It does this for fantasy points, possessions, points, assists, blocks, and rebounds. Here are the numbers on tonight’s teams:
Keep in mind that the small data sample does still matter even when looking at this. Each game has about a 25% weight with most of the league playing four games so far this year. However I do believe this is a much better way to look at defenses instead of just seeing what they are allowing on average as a whole. This approach judges an opponent’s output against their own average, not the league average — giving us a more accurate idea. So you can see from the above that the team’s to target against are the Warriors, Rockets, Kings, Bulls, Nets, and Spurs. The Warriors have been egregiously bad on defense allowing team’s to score 28 more fantasy points per game than their season average. On the other end is Utah who is taking away 33 fantasy points from their opponent’s average. Just to drive that home, here are the fantasy point totals the Warriors and Jazz have allowed this year:
Warriors: 264, 241, 268, 257
Jazz: 194, 219, 155, 198, 178
Just an insane disparity there.
Spots to Target
Nets vs. Rockets:
Doesn’t take an actual smart person to point this game out with its ridiculously high total. I will, however, do more than just tell you that the total is 241.5 points right now. I will show you how high that really is! Every morning I have scheduled Python scripts to go out and collect the Vegas line information every morning and puts them all together in one CSV file so we can really get a picture of this. We can build a quick histogram with every projected over/under Vegas has spit out this year:
You can see from that the vast majority of over/under’s have fallen between 211 and 228, with just 22 of the 152 games played being above that range (14%). This game is only the second game to be projected above 241 points (with the other being another Rockets game against the Pelicans which actually went for 249 points scored). Both of these teams rate terribly on defense by the metric we used above, so yeah there are going to be a caboose load of fantasy points scored here.
It’s not enough to know that the teams will score fantasy points, we have to pinpoint who is going to score them. Here are each team’s fantasy points distributions in pie form:
You can see that Russ and Harden have accounted for 46% of the Rockets points scored and 45% of the fantasy points. Their combined fantasy point output has been 113, 105, 109, and 136. Playing them both together costs $20,800 tonight, so if you are looking for a 6x return on that salary you would be hoping for 125 fantasy points. That is clearly very doable, and this would appear to be their best matchup of the year.
On the Nets side, Irving has one of the highest shares of team fantasy points in the league, scoring 24% of the Nets Draftkings points so far this year. He has a 29% share of their points scored, 27% of their field goal attempts, and 28% of their assists. He is even averaging six rebounds a game which we did not see from him as a Celtic, which raises his floor a bit. You could get a bad shooting game from Kyrie and not necessarily miss on this play at $9,000.
The rest of the Nets is guesswork as you can see a very spread out distribution after Kyrie’s chunk of the pie. I do think I like the Jarrett Allen #deandre-jordan)DeAndre Jordan”:/players/deandre-jordan-1486 combo tonight though. They have been playing a much more even minutes split since Allen played 20 more minutes than Jordan on opening night, and Jordan played one more minute than Allen in their last game (although neither played even 23 minutes).
You could play them both for $9,800 and guarantee yourself 45+ minutes in a high scoring game environment. They both average just under one fantasy point per minute, and with the great matchup we can reasonably project them for 1.2 FPPM, which gives us a base projection of 54 fantasy points, which is solid enough at the price. This is even more viable with Allen having PF eligibility so you would not have to use your UTIL spot while playing both of these guys.
The Warriors seem to be the most exploitable defense in the league right now, which gives the Spurs a huge boost. The only Spurs getting consistent minutes and production are Demar DeRozan and LaMarcus Aldridge, who certainly project very well here. They are both priced just above $7,000 and if they both play minutes in the thirties you would expect one or both of them to really exceed that salary expectation. There is the risk of a minutes downturn here as the Spurs are on a back-to-back and the Warriors might just get their doors blown off here without Curry. Those two are great tournament plays if you want to bet on the Warriors keeping it competitive.
Prices to Exploit:
We have seen the price on Rudy Gobert do nothing but fall since the season began:
It’s true he hasn’t been shooting the ball very much with Mike Conley in town, but now is the time to buy on Gobert, and he should be in for a big rebounding game against the Kings. I also think Mike Conley is a nice value tonight:
He finally figured out his shot in that game and made 11 of his 17 attempts, scoring 29 points. Prior to that game he has made just 9 shots on 45 attempts (20%), so it was a rough start to his Jazz career to say the least. The price is down $800 from where it started and if he shoots an average percentage it will continue to climb, so jump on now:
Derrick Rose remains much too cheap tonight as he has been one of the highest FPPM producers on the year and he sits below $6,000 on Draftkings. This is mainly because he had to leave the game early on Wednesday with an injury and only played 20 minutes in that (he still put up 31.5 fantasy points). Here’s a scatter plot of all the players on tonight’s slate that have played 20+ minutes per game with their average MP and average FPPM rates:
So the guys clustered in the top right of the graph are the best fantasy players, with high minutes and high FPPM rates. They are all quite expensive. You can see Rose’s dot highlighted red there. He is low on the x-axis because of that 20 minute performance bringing him down, but he is very high up on the y-axis, averaging 1.28 FPPM. If you would change his MP to 30, he would be clustered right in with guys like Love, Drummond, Porzingis, and LeBron — and those guys are all $3,000+ more than him. Rose is a super play until his price gets closer to $7,000 and/or Blake Griffin comes back.
There is obviously a lot more to get into, but those are the things that really stood out to me tonight, and I hope you enjoyed the visuals! Thanks for reading!