When the NBA season started, the Miami Heat had 60-to-1 odds to win the championship, making them the 14th-most-likely team to win it all. Needless to say, a lot has happened since then. The Sonic the Hedgehog movie was delayed. Donald Trump was impeached. The Sonic the Hedgehog movie was released. But after nearly a full calendar year of Sonic- and non-Sonic-related tumult, the long shots from South Beach are within striking distance of winning the NBA Finals.
On Monday, the William Hill sportsbook listed the Heat as +320 underdogs while the Lakers were -400 favorites. That means you’d make $320 on a $100 bet on the Heat to win it all, while you’d have to wager $400 on the Lakers to win $100. Miami star Jimmy Butler disagrees with those odds.
Not to be the fly in Butler’s motivational ointment, but his us-against-the-world assumption is not true. According to ESPN, 17 of the first 18 NBA Finals bets placed at the Westgate in Las Vegas were on Miami. And it’s not just extremely punctual gamblers who like the Heat. There’s an advanced computer model that favors them, too.
FiveThirtyEight’s 2019-20 NBA forecast model, code-named RAPTOR, gives the Heat a 73 percent chance of winning the Finals. How is it possible that the Lakers, a team with LeBron James and Anthony Davis, have just a 27 percent chance to win a seven-game series?
“Basically, the ‘predictive’ version of RAPTOR is wildly high on the Heat and down on the Lakers in the playoffs so far,” FiveThirtyEight’s Neil Paine told me. RAPTOR, Paine explains, comes in two flavors. The standard version credits players for things they’ve done in the past, even if they are considered to be lucky in retrospect. Predictive RAPTOR, by contrast, “gives more weight to things that are considered more persistent and downweights stats that are more luck-based,” Paine says. For the curious, FiveThirtyEight’s Nate Silver wrote up a long methodological explainer back in October. For the marginally curious, consider this one example: The Lakers have been making more field goals than expected based on where those shots were taken. RAPTOR takes that (among lots of other things) into account, and forecasts the Lakers won’t be as good as their past performances might lead you believe. (Though, the Denver Nuggets, the Lakers’ opponents in the previous round, out-shot expectations even more than Los Angeles did.)
Normally, there is a strong correlation between the two RAPTOR models, but the Lakers’ postseason performances have caused a rift. According to Paine, the gap between the Lakers’ standard and predictive RAPTOR scores is the fifth-widest that FiveThirtyEight has seen since 2014. (To be precise, the Heat have a +23.6 predictive RAPTOR in the postseason compared to +1.5, for the Lakers, while they’re relatively even—+12.1 for the Heat, +11.0 for the Lakers—in regular RAPTOR.)
RAPTOR has been a certified Lakers-hater throughout the playoffs. Paine and FiveThirtyEight sports editor Sara Ziegler discussed this on their Hot Takedown podcast last week. At that point, when Los Angeles was up two games to none against the Denver Nuggets in the Western Conference Finals, RAPTOR still gave LeBron and co. just a 23 percent chance of winning the Finals.
That their odds increased only 4 percentage points after the field was reduced in half is a testament to predictive RAPTOR’s disdain for the Lakers, even if the normal, “backwards-looking” version of the model thinks they’re just swell.
One might recall that FiveThirtyEight’s presidential forecast gave Hillary Clinton a 71 percent chance to beat Donald Trump on Election Day in 2016. Are the Heat the new Hillary? Paine says that his “own personal intuition is that [RAPTOR] is underrating the Lakers.” But don’t underrate the scrappy Heat, either. They’re both favorites and underdogs, just like those gritty Lakers.