We Might Be Overrating The Celtics, But You're Probably Underrating Them
Ahead of today’s Game 1 of the NBA Finals, let’s get one thing out of the way: Our forecast model loves, loves, loves the Boston Celtics. (Or maybe it just hates the Golden State Warriors.) Either way, the model gives Boston an 80 percent chance of winning the championship over Golden State, in very stark contrast to the betting markets — which immediately installed the Warriors as pre-series favorites last week. Based on the odds from Caesars Sportsbook, which list Golden State as -160 (and Boston as +140), we can infer that the bookmakers consider the Warriors a 60 percent favorite to win the title.
So something has to give between the two predictions. And if you ask for my opinion, yes, I think our forecast is too bullish on the Celtics. However, it also seems like the markets are too bearish on them — or, again, too bullish on Golden State. Maybe the conventional wisdom is just stuck in the mid-to-late-2010s Warriors dynasty era. Who knows? But in 2022, most indicators from throughout the season suggest that the Celtics are genuinely a better team than the Warriors.
During the regular season, Boston had a much better point differential than Golden State on both a per-game (+7.3 versus +5.5) and per-100-possession (+7.5 versus +5.6) basis. While the Warriors won two more games than the Celtics, that happened because Boston undershot its Pythagorean record by eight games — making it the unluckiest team in the league by that measure. And none of this was simply an artifact of schedule strength: According to Elo ratings, Golden State’s opponents were just 0.3 points per game better than Boston’s in the regular season. (Besides, Eastern Conference teams went 226-224 against the West anyway; the weak East is a thing of the past.)
NBA Finals tale of the tape: regular season
Values and rankings in key statistical categories for the Boston Celtics and Golden State Warriors during the 2021-22 NBA regular season
|Points per game (PPG) differential||+7.3||2||+5.5||5|
|SOS-adjusted PPG differential||+7.1||2||+5.6||4|
Now, we know that the NBA has arguably the least meaningful regular season in all of professional sports, so perhaps those stats should count for relatively little when comparing the two teams. But the Celtics have also been the better team during the playoffs. While the Warriors have improved their scoring attack (which ranked an uncharacteristic-for-the-dynasty 17th during the regular season) in the playoffs and surpassed Boston offensively, the Celtics have the superior postseason numbers in every other category despite playing a more difficult schedule.
NBA Finals tale of the tape: playoffs
Values and rankings in key statistical categories for the Boston Celtics and Golden State Warriors during the 2021-22 NBA playoffs
|Points per game (PPG) differential||+6.1||1||+5.4||2|
|SOS-adjusted PPG differential||+11.0||1||+9.3||3|
I realize this is sort of a radical concept, but if a team is better during the regular season and better during the playoffs, it might just be … better.
(And that’s without directly considering the health factor going into the series. While both teams have had well-documented injuries throughout the postseason, Boston’s most important banged-up players — Marcus Smart and Robert Williams — seem like they’re in better shape than their Golden State equivalents — Otto Porter Jr., Gary Payton II and Andre Iguodala.)
This is not to say there aren’t any measurable factors breaking in Golden State’s favor with this matchup. The Warriors do have more postseason experience, an important factor when looking at historical Finals success. Weighted by the minutes played in the 2022 playoffs, the average Golden State player had 2,031 previous career postseason minutes, second among playoff teams behind only the Brooklyn Nets.1 But Boston ranks fourth in that metric among playoff teams, with an average of 1,573 previous playoff minutes, so it’s not exactly a landslide edge for the Warriors.
Similarly, the Warriors have home-court advantage for the Finals by virtue of their superior record (despite being a lower seed within their own conference). History tells us that the team with home-court advantage in the NBA Finals wins about 72 percent of the time, which would seem to confer a massive advantage on Golden State in this series. However, since that stat also captures the effect of having the better team — which usually (but not always) correlates with a better record — the true value of Finals home court is smaller.
How much smaller? I performed a logit regression on NBA Finals results since 2000 while controlling for the pre-series Elo ratings of each team, and I found that a team with home-court advantage over an evenly matched opponent would be expected to win the series 57 percent of the time — meaning most of the perceived advantage of home court in the historical Finals records was just an artifact of the better team also having the home court more often than not.
And in this case, there’s plenty of evidence that the better team doesn’t have home-court advantage. Even if you don’t like or believe our default forecast based on RAPTOR player ratings (which, again, I don’t necessarily blame you), the Celtics are also 68 percent favorites according to our classic Elo-based predictions. Even that might be a tad on the high side, but given how much evidence there is that Boston is better than Golden State on a neutral court, it’s not exactly unreasonable that the Celtics are favored here. Because while Golden State may still win, there isn’t much objective support for the Warriors being the heavy favorites in this Finals matchup that they’re currently perceived to be by the wider NBA world.
Check out our latest NBA predictions.