Tobias Harris knows the bad things people say about him, and he knows why some teams that considered chasing him in free agency backed off.

“It’s always, ‘Oh, why doesn’t he pass more?’” Harris says, acknowledging he assisted on fewer than 10 percent of Orlando’s buckets while on the floor — one of the lowest assist rates in the league for a high-usage player.¹ “And everybody wants to talk about my defense. I’m not the best defensive player. I’m not gonna tell it to you like that. But if we were winning more games, people would talk about me differently.”

Harris is right. Four years into a strange career, he’s a classic “good numbers/bad team” guy — an archetype that has long flustered general managers who must project how such players would fare in smaller roles on good teams. Those GMs are thinking more than usual about the question today. A ton of losing teams have been in rebuild mode, and they are now burping up free agents with gaudy numbers and dreams of earning big money under a salary cap rising to unprecedented levels. Are guys like Harris, Enes Kanter, Jordan Clarkson,² and Michael Carter-Williams actually good?

“It’s coming up more, with salaries getting so much bigger,” Warriors GM Bob Myers says. “The new numbers are going to take some getting used to.”

“It’s a very tricky thing to figure out,” says Magic GM Rob Hennigan, who just re-signed Harris to a four-year, $64 million contract. “I don’t think anyone has an easy answer on what statistics mean for someone on a team that isn’t exactly excelling in the win column.”

Teams have access to more statistics than ever, but when it comes to the good numbers/bad team dilemma, GMs fall back on the oldest method of talent evaluation: watching the damn games. That’s in part because there is some degree of skepticism over whether numbers compiled on bad teams mean much — even if a player compiles them efficiently.

“Put the best D-League team into the NBA, and someone is going to average 18 points per game,” Celtics GM Danny Ainge says. “Someone is gonna get 10 rebounds. And that’s why you hear the doubts you’re hearing.”

But Ainge cautions: “Those doubts are legitimate, but not every player on a bad team deserves to be penalized.”

Some front-office types give more weight to fat stats, and more sophisticated numbers — some of them based on video — track what GMs see on film. Dig deep, and you can quantify how often a guy closes out on defense, and how many of his rebounds are easy, uncontested jobs. “We are very quick sometimes to blame players for putting up numbers of a team that isn’t successful,” Myers says. “Sometimes the losses aren’t their fault. Sometimes you put them in a better situation, and it clicks.”

A popular example: Monta Ellis, derided as an empty-calories loser before emerging in Dallas as a pick-and-roll destroyer. Ellis’s evolution was overstated in some corners; he had always been a skilled driver whose passes led to layups and 3s, and he remained a woeful defender with the Mavs. But Dallas bet that Ellis would look better attacking within the spacing Dirk Nowitzki provides, and the Mavs were right.

Numbers tell you a lot, but the process behind those numbers — the “how” — is the most important predictor for how a player will fit on a new team. There is no shortcut to making that kind of determination. You have to watch the film, watch it in the right ways, and do deep background work on a player’s character: his work ethic, what he cares about, his relationships with coaches and teammates.

It’s not much different than scouting college players who won’t get to shoot as much in the NBA. “Every NBA player was a star in college,” Rockets GM Daryl Morey says, “and they’re almost all going to be role players here.” Executives with good teams are happy to sit through a reel of Harris jumpers, but they’re really watching to see what Harris contributes when he’s not shooting.

Myers focuses on the last six minutes of close games³ and watches for specific sorts of non-shooting plays. “I’m looking for guys who make winning defensive plays,” he says. “Do they take charges? Do they win 50-50 balls? We want to find players that don’t just want to win, but really hate to lose. Like, if we’re playing Monopoly, and I land on Park Place before you, do you just say, ‘Life goes on, I’m getting a beer’? Or do you get really pissed?”

Roster context also matters. The Bucks knew Carter-Williams was forcing awful shots on a borderline D-League team in Philadelphia, so when they considered trading for him, they isolated his performance alongside the veterans Philadelphia gradually sloughed off: Spencer Hawes, Thaddeus Young, and Evan Turner. Harris’s fans around the league argue he is not operating on that sort of empty roster, and that his numbers — 17 points per game on 47 percent shooting — are more legit. “There aren’t just numbers sitting there to be had on Orlando,” says one GM who wished to remain anonymous, because he was talking about another team. “There are enough scorers and rebounders there. Orlando isn’t Philly.”

But what about the “how” behind Harris’s numbers?

♦♦♦

Here is what Harris’s critics see when they think about him:

That’s Harris jacking a triple outside the corner, an area from which he shot just 31 percent, as Nikola Vucevic rolls free to the rim with 13 seconds left on the shot clock.

That’s Harris pulling the same thing, looking off a wide-open Victor Oladipo at the top of the arc with 14 seconds left on the shot clock.

Harris has a quick trigger, but he’s not accurate enough to justify jacking long jumpers early in the shot clock. His shot selection rankled some within the Magic organization. Harris knows he should make the extra pass more often — assuming he sees them in the first place. “I have to get better at reading that,” Harris says. “But more times than not, I feel I can make my shots. That’s just the confidence I have in my game.”

Harris shot an outstanding 44 percent on corner 3s last season, but he wants to extend his range to longer triples outside the corner. Harris spends a lot of time spotting up around Oladipo and Elfrid Payton pick-and-rolls, and defenders drift far from him to clog driving lanes.⁴

That will slowly change if Harris’s uptick from deep last season proves lasting; he already felt some defenders paying closer attention. “Some teams definitely hugged up on me more,” Harris says. “And some teams still backed way off.” Harris toggles between both forward spots, and improved shooting would be especially useful when Scott Skiles slots him at power forward in small-ball groups. If Harris becomes a real threat from long range, he could drag one opposing big-man defender away from the paint — uncluttering it for Payton and Oladipo.

Harris spends a lot of practice time shooting 3s off the dribble, even though he’ll rarely do that in games. Shooting off the bounce is harder, and if he can master that, shooting off the catch “should be money,” Harris explains.

Harris and Skiles have already spent time together at the team’s practice facility, and Harris says Skiles has promised to let him run more pick-and-rolls — a change that could juice up his assist numbers. Harris laughed off the idea that he holds any grudge toward Skiles for mostly nailing him to the bench during their time together in Milwaukee. “We never had any problems,” Harris says. “He pushes me hard. I’m ready for that. When I heard the news we hired him, my first reaction was, ‘I’m excited.’”

But giving Harris the reins won’t help if he keeps doing this:

Harris has a chance to move right into a quick-hitting pick-and-roll with Vucevic, but he pauses and gives the defense time to set itself. Those moments of hesitation dot Harris’s game,⁵ and those blips can kill possessions in a league that is increasingly about flinging the ball from side to side.

Instead of using Vucevic’s screen and slicing down the middle, Harris drives away from it and into the open pocket of space along the baseline — right where Indiana can fence him in. That’s a bad habit Harris leans on too often; he veered away from the screen on about 25 percent of his pick-and-rolls, one of the league’s 10 highest rates, per Synergy Sports. That’s not all on Harris; defenders stray far from Harris and Oladipo, turning the paint into a forest of swiping arms. The good passing lanes get clogged, and the open guys are mostly awful shooters. Still: Harris has to mix things up. “I need to be more unpredictable,” he says.

Meanwhile, executives who are bullish on Harris see things like this:

Kick him the ball against a scrambled defense and Harris is deadly. He’s also a quick decision-maker. He has the reputation of a ball-stopper, but Harris mostly keeps the offense moving. He’s not Carmelo Anthony, stalling out the offense with endless jab steps, and that bodes well for his ability to fit in a modern NBA offense:

Harris can score when he gets going downhill on the pick-and-roll:

He has a nice change-of-pace game, and he rises up for nifty floaters.

And the dude can get you a bucket:

That’s Kawhi freaking Leonard. Critics wonder what position Harris plays, but his versatility is a big part of his value. He can drive past power forwards and bully wings; Harris shot 51 percent on post-ups last season, per Synergy, one of the best marks in the league.

He can do a little bit of everything on offense. He just needs to do less of the bad things and more of the good things — to trade a few bad jumpers for passes that keep the offense chugging.

Whether he can do that is the biggest question surrounding his game. Harris just got paid, and there is a school of thought among league sources that with money in the bank, he might be a little less greedy on offense. He only turned 23 last week. If he passed more, perhaps Harris could be a score-first version of Draymond Green — a jack-of-all-trades who spaces the floor and works mismatches.

Others are skeptical that a good numbers/bad team player can change on a dime. “Guys don’t change,” one GM says. “You either have that [unselfishness] in you, or you don’t. You don’t care about contract years.” If he can’t refine his passing skills, Harris may top out as an elite bench scorer who can dominate the ball on second units — and someone the Magic might eventually try to trade.

A team can’t know the future, but a deep background check can at least provide a sense of whether a player cares enough to change his game. Even Harris’s critics laud his work ethic. He’s an avid film-watcher, he’s in the gym constantly, and he pays careful attention to his diet. “We wholeheartedly believe in his work ethic and his makeup,” Hennigan says. “We’re always going to bet on those things.”

That’s a defensible bet, and given Harris’s tantalizing skill set, it’s a better bet than the opposite kind: Overpaying a guy in part because he played on a winning team. Harris will earn about as much per season as Aron Baynes, Cory Joseph, and Matthew Dellavedova combined, but Harris is the one guy among them who could change a franchise.

♦♦♦

Harris’s boosters recognize he has more ground to cover on defense, where his “tweener” status hurts most. When he’s at small forward, he faces a speed deficit against opposing wings. And he hasn’t shown the killer ferocity any small-ball type needs to bang with power forwards. If Harris can’t defend either position, what exactly is he? “The next phase of his development is to make a tangible imprint answering that question,” Hennigan says. “Right now, he’s more productive on offense, but I think he’s motivated to do it.”

Harris has an upright defensive stance that makes him look flat-footed, stuck in cement, against quicker wings. He falls behind when chasing guys over screens, and he’s vulnerable to blow-bys when lurching toward shooters; he can’t stay in front of Marcus Morris after recovering toward the right corner here:

“I watch film of myself, and I’m like, ‘Damn, I’m up pretty high,’” Harris says of his stance. He needs to crouch down so he can slide faster side-to-side, and he’s hired a yoga instructor to help loosen his hips. “They are really tight,” Harris says. “But the yoga has helped. I can already get a little lower.”

At the same time, he’s working with Orlando’s strength coach to get stronger so he can battle with bulkier power forwards. Extra power will help, but some of those scrums come down to simple effort. If you’re at a size disadvantage, you have to work your ass off boxing out Tristan Thompson–types on every possession. “The biggest difference playing power forward is the toll it takes on your body,” Harris says. “That’s something I’m preparing my body for now.”

You have to bring the fire every second you’re on the floor. Harris hasn’t always done that; he’s taken just two charges over his four seasons, per data provided by Stats LLC. A lot of guys have flopped their way into more charges than that.

Harris got better last season. He was more diligent getting back in transition after years of admiring his own wayward jumpers while his guy leaked out. As the league get smaller, Harris will face more power forwards his size. And when Harris dials up the intensity, he can hang with crafty wing ball handlers; watch him switch onto Dwyane Wade and stick step-for-step with him on a crunch-time possession:

That right there — that fluid bit of positional versatility — is why Harris almost got a max contract. It is the “how” you’d bet big on — if you think it’s something Harris can replicate. He swears he can bring that kind of defense more often. “You look at my on-ball defense, and it’s not bad,” Harris says. “If you say I don’t play defense, that’s an insult. I put in my effort every night. Guys are not straight dogging me out there. It doesn’t happen.”

Harris generally gets where he needs to be, and when, within Orlando’s help scheme, though he can get turned around if an enemy offense moves the ball fast enough. He has been a minus defender so far, but he hasn’t exactly had a stable NBA career. He started as a benchwarmer on Milwaukee teams lacking veteran leadership, and he fell in and out of favor with two different Orlando coaches. He has flipped almost haphazardly between positions, in all sorts of crazy lineups. Provide a bit more stability, and Harris might flourish.

Skiles will demand accountability on defense, and his last Milwaukee teams flashed the kind of side-to-side ball movement that became even more in vogue after the Bucks fired him. The ball got sticky in Orlando last season, and the team hopes that will change under Skiles.

The Magic will be an interesting team this season, but they are most intriguing with the Aaron Gordon–Harris pairing of positionally ambiguous hybrid forwards. Gordon is a bouncy defender learning to shoot more, and Harris is a score-first guy working to refine his defense.

They could switch almost anything on defense and use their speed to punish teams on the other end. The Magic will be hurting for rim protection with Vucevic at center, but Dewayne Dedmon might be the only traditional big on the roster who brings that skill. Dedmon has hooves for hands,⁶ and pairing him with Vucevic was the final death blow for Orlando’s already shaky spacing.

Getting funky with the Harris-Gordon duo might be Orlando’s best path toward solid two-way play. “Teams would have to pick their poison,” Harris says. It would certainly be the most entertaining.

“This team could be really good,” Harris says. “It’s only a matter of time.”

That’s in part up to Harris. He has his warts, but look closely, and you can see the profile of a player who could flourish in the modern NBA. The Magic were smart to wager on that upside. Now they and Harris must work together to bring it out.

1. And especially one who spends a lot of time on the perimeter.

2. A restricted free agent next summer, but already drawing the good numbers/bad team scrutiny.

3. And he’s not talking directly about Harris here, to be clear.

4. Harris ranked about average in “gravity score,” a SportVU metric provided to Grantland that measures how closely defenders stick to individual offensive players. He ranked well below average in “distraction score,” which measures how far defenders roam off players.

5. He would drive Gregg Popovich crazy.

6. Trademark Haralabos Voulgaris.