DoDo on the rise: Who's better than RoCo?

[vfromlmf]

I like tacos

TaCo > DoDo > RoCo ?

[ClutchDirk]

Thread map:

X/Y => DFS vs RoCo => Not either/or but both/and => Gaslighting => Anger/Apology/Forgiveness => more DFS stats => Triggered/Everything that's wrong with the internet => Tacos => Curry mention => Fantasy Trades

All paths lead to fantasy trades.

---

I like tacos

Its not Tuesday...LeChina hate detected...

[dirkfansince1998]

Why did you choose to use the x/y approach with the selected stats to start the thread?

Because I was surprised that DoDo's numbers looked so good compared to RoCo. Thought others would be surprised too. I think DoDo is a steal @ $4 million.

The numbers aren't cherry picked. They're in every boxscore. You know, height, weight, minutes, points, assists, rebounds. Salary is the salary. Nothing disingenuous. If you don't like it, make your case. 

In the meantime I'm going to go get some tacos.

Why not add steals, blocks or a non boxscore defensive metic? Why not mention the role they have on their team? Instead we get the assumption that both are equally good on defense.

[WildArkieBoy75]

Lacking better judgement, I will weigh in on RoCo vs DoDo. . . . . 

Stats only, no names player comparisons are a common gimmick.  I've learned to simply pic the player making the fewest bucks and say "good observation" and move on down the proverbial road, but," Scottie, a certified devil's advocate and board contrarian, quite possibly in that that "time of the month" or "red flag" ,as my filipina wife refers to it, chose to escalate the discussion to one of those evolution vs creation mini board wars, a veritable monkey trial.

Fifteenth, playing the part of William Jennings Bryan, came back with righteous vengeance and indignation.  

What I learned from the discussion is that I have no idea how to compare players.  I'm an "eye test" guy who uses the box score when I can't watch. (I read Luka's stat line first, then look at KP's rebounds, then it's on to Maxi vs Powell.) Charles Barkley told me to ignore analytics, let alone the dreaded 'advanced analytics." 

Now I've learned I can't compare players without factoring salary, so it's not "RoCo vis DoDo," but "Roco vs DoDo plus Seth."

Thanks guys!

[dirkfansince1998]

For the mentioned stats from greekgoodofstats.com.
The author tries to build a defensive all around stat based on play-by-play data. He uses the same data to determine a players defensive load.
LOAD is based on an opposing players PPG and the amount of defended possessions. Not including the efficiency.
He combines the LOAD with play-by-play shot defense data to determine the shooting defense.
Non-shooting defense is only focusing on forced turnovers. Again using play-by-play data to determine how many times the matchup turns the ball over.

It´s an interesting concept because the author basically says that the kind of shot the opponent takes is not important. It´s more important who is shooting.
Not really sure if I agree. Big question mark that remains is the algorithm he uses to determin a players role/position. Seems like it is unavailable unless someone is willing to buy a 15$ ebook.

I have nothing against the numbers but we need to know what we are talking about. LOAD + play-play-data for shooting and forced turnovers are not everything that matters for defense. I really don´t like the forced turnovers/ changes of possession --> non shooting defense.
Obviously not possible to account for everything that happens on the basketball court but that is a really shallow approach.

[vfromlmf]

I'm an "eye test" guy who uses the box score when I can't watch. (I read Luka's stat line first, then look at KP's rebounds, then it's on to Maxi vs Powell.) Charles Barkley told me to ignore analytics, let alone the dreaded 'advanced analytics."

Love your post. And generally speaking, I'm in the same camp. I watch the games, then I check the boxscore. Not so much for "player evaluation" though. Boxscores are fun. Sort of a game within a game. I like Luka triple doubles. I like KP rebounding well. I like when our role players shoot well. I like when we out-rebound the opponent and get more than 25 assists.

In fact -- and not to derail this thread further -- but one of the best parlor games in sport is to guess which version of THJ will show up on any given night. It's a total crap shoot. Try it.

But back to the point, I do occasionally like to dive into the advanced analytics ...though I would never rely on stats alone to do player evaluation. 

The idea that analytics can be used to capture value from data is not controversial. It's well proven across every industry. New sources of data and new methods like ML/DL are enabling insights-based value creation models to drive top-line revenue & enhance activities like pricing, churn-prevention, cross- and up-selling; bottom-line activies like predictive maintenance, supply chain optimization, fraud prevention; and wholly new business models. Of course Moneyball taught us analytics can be applied to help identified undervalued assets in sports, for example players who a likely to outplay their contract.

So there's some value in checking to see what analytics has to say about DoDo. Is he scoring efficiently with his limited chances? Is he really rebounding well or is he just getting boards because he's playing the four in a small lineup? The Mavs switch a ton so is DoDo really defending the opponent's best scorer more often than, say THJ? The advanced stats can help with this.

Quite honestly it surprised me ... DoDo looks really good according to advanced analytics.

[fifteenth]

TaCo > DoDo > RoCo ?

Ta > (DoDo)/Co > Ro

---

Why not add steals, blocks or a non boxscore defensive metic? Why not mention the role they have on their team? Instead we get the assumption that both are equally good on defense.

I don't think we have to require that every OP is comprehensive. It can be a conversation starter. DF98 if you think an OP is tilted one way or leaving out some relative info, why not just come along and add it?

[SleepingHero]

An interesting question I think should be posed,

If DFS is performing so well in our system, how much better would a guy like RoCo be, given that he has a much better knack to score than Dodo? 

Only reason why I say that is because RoCo has 6 seasons of 36%+ from 3pt percentage, a sample size of 326 games. While this is DoDo's first year (so far) ever getting higher than 31% from 3. And unfortunately DoDo has shown in the past that he starts hot and by January his 3pt shooting dies. Last year Dodo was averaging 39.8% from 3, and then by 12-14-2018, his 3pt shooting took a nosedive. From 12/14 on, he shot 26% on 3 attempts a game. It is undeniable that Robert Covington is a more consistent 3pt shooter. 

So wouldn't RoCo fit in even better than DFS? The analytics show that they are remarkably similar on the defensive end, so they aren't giving up anything there. Plus the Mavs as a team take the most open 3's more than anyone in the league, and Dorian receives a lot of these looks. Given that RoCo is a better shooter, wouldn't he be that much better than Dorian?

Again I'm in the camp of DFS AND RoCo, not either/or.

[vfromlmf]

So wouldn't RoCo fit in even better than DFS?

Possibly. Maybe DoDo is improving, maybe not. I hope he doesn't hit a slump, but he might. Interestingly, the WP48 stats I posted earlier actually show DoDo's career Wins Per 48 Minutes, are better than RoCo's career numbers. And those career numbers were posted with guys like Deron Williams and DSJ running point. So, there's that.

DoDo certainly seems to be a better value than RoCo, at least from a Moneyball standpoint. 

There's also a great series of quotes from Carlisle and DoDo in The Athletic article. This regarding offensive rebounds: 

“He’s got a real gift,” Mavericks head coach Rick Carlisle says. “It’s something that you can teach guys — you can tell guys to crash the boards at certain times and things like that — but you can’t teach the feel and the anticipation and the tenacity he has for going after offensive rebounds. It’s really remarkable. He’s made uncanny plays all four years he’s been here.”

And this from DoDo, which relates offensive rebounding to defensive transition:

“It depends on my matchup, too,” Finney-Smith says. “If I’m guarding a guy who likes to leak out or a team that pushes a lot, my offensive rebounds probably won’t be there that game.

“We also do a good job of picking up each other in transition so if I crash, Tim will pick up my (man), Luka will pick up my guy and we communicate.”

“Crazy part is, I feel like I can get more, man,” Finney-Smith says. “Like I said, it be matchups. Some matchups, I can’t really crash like I want to because I’m sprinting back. I feel like if I wasn’t too worried about getting to a guy, I feel like I could get more.”

It’s part of the challenge, though, and something Carlisle coaches all of the players on in practices and the film room.

“(Offensive rebounds) involves the ability to regain possession of the ball so if you can do that, it’s a very big advantage,” Carlisle says. “But if you’re throwing a bunch of guys at the offensive boards, you’re not going to have numbers back, and that’s a challenge. So there is a balance to it.”

This goes back to a point I made earlier. To properly evaluate defensive performance, you have to have knowledge of what each player is asked to do. There's no way of knowing this from the boxscore. Analytics can help - if you consider who does what on every possession (this is part of the advanced defensive metric above) - but there's no way to really know unless you're in the locker room listening to the game plan.

Again not to derail this thread but this is why you have to pay attention to what DoDo's coaching staff and teammates say about him. Some players are selfish and get stats, despite the game plan. Not saying this is what RoCo does ... but I suspect a guy like Julius Randle might be guilty of this kind of thing. 

So stats alone don't cut it. Box scores are fun but not for deep player evaluation. Analytics can be better, but they're not to be relied upon in a vacuum. You have to look at the whole picture of a player including size, position, role, boxscore, advanced metrics, minutes, statements from coach & teammates, wins & losses, contract. It's very complex. 

One thing for sure, this thread has helped me develop a new level of appreciation for what DoDo brings to this team relative to what he's paid. He's an incredible value and the Mavs have to be ecstatic about what they're getting for the money. 

And if Seth Curry continues to look like the next JJ Reddick, DoDo + Curry for the price of a single MLE player has to be among the best value combos in the league.

[fifteenth]

Given that RoCo is a better shooter, wouldn't he be that much better than Dorian?

Does our rotation need the scoring of RoCo at his price tag or the offensive rebounding of DFS at his price tag?

[dirkfansince1998]

TaCo > DoDo > RoCo ?

Ta > (DoDo)/Co > Ro

---

Why not add steals, blocks or a non boxscore defensive metic? Why not mention the role they have on their team? Instead we get the assumption that both are equally good on defense.

I don't think we have to require that every OP is comprehensive. It can be a conversation starter. DF98 if you think an OP is tilted one way or leaving out some relative info, why not just come along and add it?

I did. But that´s not my point. I just dislike the player x/y concept. It fakes objectivity and takes away the context. If we learned anything about stats in the last few years it is that they are meaningless without context.

Example: Who would you rather have?

A:  35 minutes, 25/5/3, 46/34/75, 27mio, 24 years old
B:  34 minutes, 25/7/4, 49/39/84, 7mio, 22 years old
C:  31 minutes, 25/8/5, 46/36/87, 33mio, 28 years old

In the following posts the OP added something like this.

Player A is still on his original team
Player B has been traded once
Player C is a known diva and forced his way out of his original team


Without context those numbers are useless. It can be a fun exercise to find the matching players but it makes no sense to compare the players based on those numbers.

[vfromlmf]

Example: Who would you rather have?

A:  35 minutes, 25/5/3, 46/34/75, 27mio, 24 years old
B:  34 minutes, 25/7/4, 49/39/84, 7mio, 22 years old
C:  31 minutes, 25/8/5, 46/36/87, 33mio, 28 years old

In the following posts the OP added something like this.

Player A is still on his original team
Player B has been traded once
Player C is a known diva and forced his way out of his original team

 

I’d rather have Player B. Starter, good shooter, rookie contract. Who is it?

[dirkfansince1998]

Example: Who would you rather have?

A:  35 minutes, 25/5/3, 46/34/75, 27mio, 24 years old
B:  34 minutes, 25/7/4, 49/39/84, 7mio, 22 years old
C:  31 minutes, 25/8/5, 46/36/87, 33mio, 28 years old

In the following posts the OP added something like this.

Player A is still on his original team
Player B has been traded once
Player C is a known diva and forced his way out of his original team

 

I’d rather have Player B. Starter, good shooter, rookie contract. Who is it?

Brandon Ingram.

[vfromlmf]

Brandon Ingram.

 

Interesting. That guy has turned into a straight up bucket getter. Really tough to stop with his length.

[dirkfansince1998]

Brandon Ingram.

 

Interesting. That guy has turned into a straight up bucket getter. Really tough to stop with his length.

Other options were Andrew Wiggins for A and Kawhi Leonard for C.

[vfromlmf]

Other options were Andrew Wiggins for A and Kawhi Leonard for C.

Figured that, based on the max deal at nearly 5x more money than Player B. It proves stats can be deceiving and don't tell the whole story. Ingram's numbers are on bad team while Kawhi is an MVP, with titles on two different teams. A true superstar.

See, X/Y constructs can be fun. 

Of course, superstars on max deals are totally different animal than role players. You wouldn't really make the argument that Ingram is on Kawhi's level, would you. You know five good $7 million role players don't equal one $35 million MVP. Just doesn't work that way. I wouldn't classify Ingram as a superstar. Booker isn't a superstar either. At this stage Ingram - like Booker - is looking like one of the great young scorers in this league though. He has star potential but does he make players around him better? I don't know. Maybe a better comparison would be Doncic's rookie deal versus Kawhi.

Another interesting comparison of top-flight G and F scorers might be Booker vs. Ingram vs. Beal vs. healthy Oladipo. Maybe throw Trae Young in there. Donovan Mitchell. Maybe some others. Beal & Oladipo are the most accomplished but also cost the most money. Beal is a star but certainly not an MVP candidate, at least with his current team.

[vfromlmf]

Bump :-)

[dirkfansince1998]

Bump :-)

If he continues to hit corner 3s like he currently does he is extremly underpaid. That said...at his best RoCo was a defensive anchor. DFS is playing good onball D most of the time. Not so much today but DeRozan is a tough matchup
Really hope he continues to shoot like he does. In the last two years his shooting declined post allstar break.

[Jym]

Dorian is exceeding all of my expectations
I would have considered it crazy talk to suggest he was going to hit 3's at a higher rate than Jackson. On a much higher volume too. What a steal at 4 mil a year.
And I like his clutchiness. I have more confidence in a last minute shot from Dodo than I do KP

[ballsrchr]

I've got to find a crow to eat I guess.  I originally thought DFS had a middling low ceiling.  But if he keeps improving a little every year--like he has been doing--he might actually turn out to have a fairly high ceiling.  I'm happy to say I was wrong.  Yes...quite a steal at 4 million.