A Giddey-Caruso Swap + Should We Care About Stat Padding?
An analytical summary of why both teams won't regret the trade and a less serious breakdown of one of the NBA's bigger controversies.
It didn’t take too much time for the NBA Offseason to kick off - we’re not even at draft day yet and we’ve already seen a significant trade. Who won the guard swap?
Before I start, I want to credit Crafted NBA for the stats - highly recommend everyone checks them out - they come up with proprietary stats trying to answer common questions.
Oklahoma City Thunder:
The Thunder received Alex Caruso - a player who many might not know already spent some time in the Blue. Caruso has had an evolution since the beginning of his NBA career, evolving from unwanted to meme to highly effective - and he enjoyed a peak in Chicago, where he shared the court for one season with Lonzo Ball (one of the greatest defensive backcourts ever) and then took on the extra workload in the past two seasons, adding a First Team All-Defense selection in 2023 and Second Team in 2024. Caruso is an effective player - and analytics love him. In his 2022-23 tour de force, he led the league in turnovers forced and posted a defensive rating of 106.5, and then he followed it up with a season where he led the league in Regularized Adjusted Defensive Turnovers (which includes matchup difficulty and then and role of the team) and was 3rd in deflections (minimum 350 minutes played). Clearly, Caruso is a workhorse on defense - and it’s obvious who the Thunder are trying to model their team after (c’mon Presti, be original…we liked you when you drafted Poku!).
Clearly, Caruso takes the matchups and produces (although steals aren’t the greatest metric to use here, the other stats I mentioned should work in conjunction with this).
The question remains though, how much is Caruso going to fit on offense? Is his 40% from behind the arc an accurate result? He had a true shooting percentage of 61.3% and eFG of 59.3%, so is this sustainable? It is a bump over last season, where he shot 58.8% and 55.1%, respectively. And then in 2022, he shot 52% and 48.0%. So, he’s been improving? It’s likely he has just improved as a catch-and-shoot threat.
Having watched enough Bulls games this year (I also have rewatched Attack of the Clones multiple times, so don’t ask), I felt there were moments where Caruso disappeared on offense. Is my “eye test” failing me? Maybe I’m misinterpreting good shot selection for passivity. According to Crafted NBA, Caruso had only 27.4% of the offensive load (significantly less than his former teammate Coby White’s 39.8%), and had more poor games than fair games, but overall, had more good and great games than poor games. My eyes are fine - Caruso isn’t as consistent an offensive player as he is a defensive player - but the Thunder can deal with that. As for shot quality and offensive load, Caruso ranked just down the middle in both.
An important stat I wanted to check out was Portability - I’ll link the formula below. The idea is just to see which players can fit in better on a system (as they bring more universal traits). Caruso ranked 71st in the NBA, to Giddey’s 254th.
Portability:
TS% + (Standardized Shooting Ability x 2.5) + (Standardized CraftedDPM(positional) / 2) + (Standardized Versatility / 3) + (Standardized Passer Rating x 1.5) - (Standardized Load x .25) - (Standardized Box x .15)
16% Scoring Efficiency
40% Shooting Ability
8% Defensive Ability
5% Defensive Versatility
25% Passing Ability
6% Usage (penalize outliers)
All in all, the Thunder swapped a player that wasn’t working for them for someone that should, but in limited minutes. Caruso does have an injury history - but we’ve seen teams deal with that on the way to a title. This is an outstanding trade: A+.
Chicago Bulls:
Josh Giddey as mentioned above, is not a very portable player - but that isn’t a death knell. Some players rank relatively lower on portability: Jaylen Brown and Aaron Gordon, both of which are are playing important roles on championship-level teams. So, clearly, being unportable isn’t a super big deal.
Giddey is supposedly a great playmaker - but wait - let’s think about that again. Is he a good playmaker? Or is he a good passer?
Stephen Curry is a good playmaker because he puts enough pressure on his opponents to MAKE plays. Rajon Rondo was a good passer - he could effectively quarterback a team but didn’t have enough offensive pressure to create the plays himself.
Ben Simmons and Lonzo Ball were great playmakers in transition(!), where they could put pressure on their opponents through power or speed, but in the half-court, their lack of shooting failed to make them good enough PLAYMAKERS.
So where does Giddey go? I fear he’s more in the Simmons, Rondo, and Lonzo camp as of now - but those players were all great (more for their defense than offense, but I’ll try not to be too annoying), so that’s not a bad sign. Giddey does better with the ball in HIS hands - he can’t reliably make things happen without that - so how much will the Bulls improve with him handling the rock?
Why was Giddey so bad this year? His on-off metrics were terrible, his relative true shooting was below average (mostly at the beginning of the season), and he seemed utterly uncomfortable most of the time he was on the floor.
Giddey had an offensive load of 45.1% in 2022-23, and 27.1% in 2023-24. The Thunder made a tactical decision to take the ball out of his hands - they clearly had better options. His shooting quality went up though, most likely because of his taking more efficient shots and training with Chip Engelland.
As for his passer rating and box creation, they both remained similar - so he still is an effective passer. Cutting off the oxygen hasn’t yet regressed other parts of his game. Giddey can help the Bulls - and even if he doesn’t, Chicago will be happy keeping their pick away from the Spurs (other than me and the San Antonio locals, no one really wants to see Cooper Flagg and Wemby on the same team). Giddey is more of a floor raiser than anything. The Bulls were #28 in pace last season, averaging around 97 possessions per 48 minutes, and Giddey will be an improvement, with his 102 possessions per 48. If the Bulls want him to improve his trade value before being shipped out again, this seems like a worthy gamble.
As for the actual trade package, I feel like Chicago could’ve gotten more. Caruso has a perception of being super valuable, even if he does play limited minutes. No picks back from OKC’s treasure chest feels disappointing. I’ll give this one a C.
A Brief History of NBA Stat Padding:
An investigation by Tom Haberstroh of Yahoo Sports has uncovered some intriguing details about Michael Jordan’s 1987-88 Defensive Player of the Year (DPOY) season. Notably, some of the stats from that season appear to have been fabricated.
I think I just heard a LeBron fan yell, “I knew it!” as I typed that sentence.
But I’m not here for silly Twitter discourse; this is serious! Let’s talk about how many times Michael stopped Larry from shooting into the peach baskets (just kidding).
First, what is stat padding? Was statistical fabrication exclusive to the '90s? I believe the NBA was more likely to fabricate stats before games were widely available for replay and before stat recording became computerized. However, we’ve seen other ways players’ counting stats have been misinterpreted or misrepresented.
What if I told you Fred VanVleet (all 5'10" of him) had more blocks than Kawhi Leonard this season? Would you believe me? Would you start reassessing Fred as more than just the stout guard he is? Well, I looked through some of Fred VanVleet’s blocks, and they look more like steals/deflections on rip-throughs. Suddenly, Fred isn’t the hero we thought he was. Boom. Stats suck.
What about Russell Westbrook in 2017? Does it count as stat padding if his teammates let him grab all the defensive rebounds in sight? Does Hassan Whiteside count, because he jumped for every shot contest rather than actually playing defense? Once again, stats suck!
The lesson from the Jordan story is that while it’s problematic the league didn’t force accurate recording of steals and blocks, relying solely on counting stats to determine awards has always been flawed. Yes, box scores aren’t entirely useless—a player with a higher usage rate usually has the ball in his hands or is shooting for a reason, and humans are generally good at assessing who is a good basketball player. But when we start reverse engineering our decisions using counting stats, we get stuck in limbo.
It’s unfortunate that some of these advanced metrics weren’t around back then—all they had was a weird version of PER called the IBM Award (and I think that came much later than 1988 anyways). But now that we do have advanced metrics, we can more accurately identify the best defenders. And yes, Michael Jordan was one of them.