The Portland Trail Blazers and the Race Against Time
Plus, what do the creators of the leading player impact metrics think about the MVP race
The main thing I’m paying attention to these next couple of weeks is whether the Portland Trail Blazers can pull off the improbable and capture the 10th seed in the West.
The Blazers are three games back of the Phoenix Suns for the final spot in the Western Conference Play-In Tournament. Their chances of making the postseason are slim, but not zero given the way they’ve played lately.
The Blazers have logged 72 games and if you split their season in half they look like two different teams.
In their first 36 games their record was 13-23. They appeared to be headed for the bottom of the standings and Chauncey Billups seemed like the favorite to be the first coach fired.
But in their last 36 games the Blazers have gone 19-17 with one of the league’s top defenses.
Since mid-January, the Blazers have climbed up the Defensive Rating rankings and now sit at 16th on the season
If you look at just at the last 36 games, they’re 5th in Defensive Rating, trailing only the playoff bound Celtics, Clippers, Pistons, and Thunder.
Usually in situations like this, I’m quick to point out luck as the main driver for a team’s unexpected success. On defense, that often takes the form of opponents shooting worse than average from beyond the arc. But that’s not the case with the Blazers. Opponents have shot 35 percent on threes over the Blazers’ last 36 games — that’s below the league average (36 percent), but not dramatically so.
Instead of relying on luck, the Blazers have done something different. They’ve used time, or the lack thereof, as a sixth defender.
Let me explain.
This year the Blazers lead the league in a niche stat that I couldn’t find tracked anywhere. I had to go into the NBA’s play-by-play data to pull it out. The stat encapsulates what makes the Blazers’ defense pop and why I think Billups might have earned himself contract extension this summer.
The stat I’m talking about is shot clock violations.
This year, the Blazers have forced more shot clock violations than any other team.
The Blazers have forced 69 shot clock violations this year, or about one a game on average. The 2022-23 Cavaliers hold the record for most shot clock violations forced in a season since 20181 with 79. With ten games left, the Blazers have a real chance to surpass that number.
A shot clock violation is a tell-tale sign of a well coached team that’s bought in on the defensive side of the ball. To force a shot clock violation all five defenders have to be in sync and be willing to give 2nd, 3rd, and sometimes 4th efforts. Defenders have to be aggressive, but not too aggressive that their aggression results in a foul. They have to be unrelenting with ball pressure and activity until the offense taps out.
Here’s the Blazers hounding the number one ranked offense until they had no choice but to throw up some bullshit.
Lest anyone think it was a fluke, the Blazers forced another one a few moments later in the same game.
The Blazers would eventually lose to the Cavaliers in overtime, but the message was loud and clear: if the Blazers could cut off the Cavaliers’ oxygen, they could suffocate anyone.
Even the threat of a shot clock violation is enough to put most offenses into a tailspin. League average efficiency falls off a cliff the later you are in the shot clock. So the more shots you can force your opponent into taking against a dwindling shot clock the better your defense will be.
The chart below shows the percentage of opponent shots that come with four or less seconds on the shot clock. The Blazers rank number one. So even when they aren’t forcing violations, the Blazers are forcing opponents to throw up junk more often than anyone else.
Most shot clock violations start 94 feet from the basket. They happen when the defense applies pressure right after the inbound pass and maintains it throughout the possession.
It’s no surprise then that the Blazers have the highest average pickup point in the league, according to Second Spectrum tracking data. Whether it’s Toumani Camara, Scoot Henderson, or Deni Avdija — no team applies pressure to opposing ballhandlers in the backcourt more often or more relentlessly than the Blazers.
Offenses like to get into their first action with 18 seconds or so on the shot clock. But against the Blazers, that number is closer to 17 seconds. That’s one less second for the offense to pivot to a second or third action if the earlier one goes nowhere. A single second might not seem like a big difference, but add it up over the course of a game and you’ve put a real dent in your opponent’s efficiency.
This can be a devastatingly effective way to play defense. It wears down the offense and forces them into a shot they have to take instead of the one they want to take.
I’m sure more teams would love to try what the Blazers are doing if it weren’t also an exhausting way to play. In a good game, a defense might force two shot clock violations. You need the right personnel and the right coach to keep them motivated to go balls to the wall for 48 minutes. Portland appears to have both.
The Blazers are littered with young, indefatigable players up and down their roster. Add in Matisse Thybulle, who just came back from injury, and Billups has his pick of hyperactive defenders that are eager to perform defensive slides up and down the court.
The Blazers have ten games left to capture the 10th seed. Time is not on their side. But if there’s a team that has a finishing kick for the home stretch, it’s the Blazers. Regardless of where they end up, I think this season has proven to be anything other than a waste of time.
Do VORPs and SCHNORPs matter in MVP debates? Here’s what their creators say…
As the regular season winds down, the MVP debate has begun to heat up.
There are only two candidates that are under consideration — Shai Gilgeous-Alexander and Nikola Jokic.
Here’s the quick and overly simplified case for each player:
Jokic is the best player in the world and is currently playing the best basketball of his career
SGA is the best player on the league’s best team and he’s having as good (if not better) an individual season as Jokic
Recently, I’ve seen fans and analysts point to the “advanced metrics” as a tiebreaker between the two candidates. By “advanced metrics,” I’m referring to the all-in-one impact metrics — the VORPs and SCHNORPs.
None of these metrics are perfect, but when most of them rank SGA and Jokic 1st and 2nd in some order it’s fair to say they do a good job of defining the pool of MVP candidates.
The question I have though is, Should these metrics help us differentiate between the best and second best?
Is the gap between, a +8.0 player and a +7.5 player big enough for us to say that one player should be the MVP over the other?
I’ve spent the last few weeks thinking about what role advanced metrics play in the MVP race. None of these metrics were created with the intent to help us settle MVP and other awards debates. But that’s what they’re used for this time of year.
So last week I reached out to the creators of four of the most popular all-in-one impact metrics to get their take.
I asked them whether it was appropriate for people to use their metrics when debating the MVP and their thoughts about the race more broadly.
Here’s what they said:
Kostya Medvedovsky — creator of DARKO Daily Plus Minus (DPM)
Short answer, no. I think I even have this on the explainer somewhere, but
generally DARKO is predictive and forward looking at all times, so if a
player plays great, but somehow in a fashion that’s really fluky and
completely unsustainable, their DPM rating may remain quite low. Now, in
practice, if you play well, you will probably have a good DPM, and if you
play poorly, you will have a bad one, but conceptually, DPM is always trying
to be forward looking, not backward looking, and so it’s not a great tool for
the MVP.
Taylor Snarr — creator of Estimated Plus Minus (EPM)
Yes! I think it's appropriate for people to use EPM when arguing who should be MVP, but perhaps not as an absolute rule. The goal of EPM is to capture player value, so ideally, the most valuable player would be at the top. EPM has aligned fairly closely with MVP winners over the last 15 years, but it is an *estimate* of true impact, meaning it can't capture everything.
I don't align my player evaluations perfectly with EPM, but feel EPM had it right in 2022-23 and Jokic should've won. I also think EPM, and other impact metrics, can help serve as a check on our obsession with counting stats (especially when undervaluing efficiency), like perhaps with Westbrook winning in 2016-17 and D. Rose in 2010-11.
A tip when using EPM in the MVP debate: there's a bit more leeway with defensive EPM, meaning if there is a good scouting case for defense to be higher or lower than DEPM, then feel free to qualitatively factor that in.
For this season, SGA and Jokic have clearly separated themselves both in EPM and in the MVP discussion. Right now SGA has a slight lead in EPM, and I'd probably lean that way personally, but I think choosing either one wouldn't be a betrayal of EPM.
Krishna Narsu — creator of Luck-adjusted player Estimate using a Box prior Regularized ON-off (LEBRON)
I think it's fair to use LEBRON when arguing who should win MVP. After all, the award is called Most Valuable Player which imo, is close to impact. We're not voting on who is the best player (I say we as if I have a vote which I don't). And I think that's an important distinction because with something like "best player", that's where you run into issues with the team context such as what role a player is in, are they optimized etc. And most of the impact stats can't account for that super well. But for MVP, the context of your surroundings has always mattered and so I think LEBRON (and other impact metrics) are worth looking at for MVP.
One last thing I wanted to mention- we have a few brothers of LEBRON (no Bronny though). One is called multi-year LEBRON which is like DARKO in that it's the best estimate of what a players' impact is right now and includes multiple years of data. The other is EBRON which is LEBRON without the luck adjustment. I think there's an argument that EBRON might be worth looking at more than LEBRON since it's more reflective and less predictive. But then LEBRON better accounts for teammates' hot shooting when you're on/off the court so maybe not. I go back and forth on this.
Also, since you asked for my take on the MVP race, I think it's as close to 50-50 as you can get. I'd probably lean Shai right now but ask me in a week and it might be Jokic.
Jeremias Engelmann — creator of xRAPM
I think the main take-away from the metrics should be that no one needs to spend a split-second pondering whether anyone but the top-2 should even be considered for MVP
Regarding Jokic vs SGA, the error bars of all metrics are too large to make a definite statement on who was the better player this season with more than ~60% certainty.
What's funny is that it seems so ambiguous no matter what stats you prefer. More into plus-minus? Jokic is #1 in On-Off, but SGA is far ahead in single-season RAPM.Traditional stats are more your thing? Jokic has the most triple doubles, but SGA has scored more points in just the first 3 quarters than anyone else has scored in total
What I'd love to see is one of them winning the award, and the other getting revenge in some later playoff round, reminiscent of Hakeem Olajuwon's battle against David Robinson in '95
For more info on these metrics and their creators, visit The F5’s Q&A archive. I’ve done interviews with the creators of DARKO, EPM, LEBRON, xRAPM, and more.
In 2018, the league changed the rules so that the shot clock resets to 14 seconds after an offensive rebound.
Toumani Camara, y'all.
Fun information. Two pieces of context that I think are interesting to add:
1) Miami runs the most zone in the league by far; Portland and Miami are both top-10 in possessions where they press, per Synergy
2) Portland and Miami both are rather middling defenses. Again, we're dealing with small samples here, but I would've expected the best defenses to rank better here. Might be related to point 1, but I don't have a good way to disentangle those.