Nate Silver’s FiveThirtyEight under fire for wrongful predictions

Silver under fire

In the early stages of the NBA playoffs in April, FiveThirtyEight had the Boston Celtics as its favorite to win the championship. The Celtics are the Eastern Conference champions, and just as Nate Silver’s projections said, they had a 59 per cent chance of winning. They will play in the NBA Finals this week. Check out the best bookmakers for betting on the NBA Finals!

Check out the top sportsbooks for betting on the Celtics/Warriors

A lot of people mocked the FiveThirtyEight numbers at first, saying they were bizarre because they fluctuated so much, depending on the outcome of individual games, and erroneously rated the Celtics’ chances higher than all the other top contenders combined. Yet, the Celtics made it out of the Eastern Conference.

[mytwitter url=”https://twitter.com/NateSilver538/status/1520028222280376320″]

On the other hand, FiveThirtyEight surprisingly gave the Golden State Warriors only a 22 per cent chance of making the finals and a 9 per cent chance of winning compared to the Boston Celtics’ 39 per cent.

The algorithm was not bothered by Golden State’s success or the fact that the Celtics had to win two games in a row to get out of the Eastern Conference. According to FiveThirtyEight, the Celtics have an 80 per cent chance of winning the NBA Finals, while the Warriors have a 20 per cent chance.

This does not imply that Silver’s computer model arrived at this point through a process of constancy or consistency.

“Forecasts are based on 50,000 simulations of the rest of the season,” FiveThirtyEight said. “Our NBA player projections, which estimate each player’s future performance based on the trajectory of similar NBA players,” the site wrote, explaining the system disregards teams’ victories and defeats in favor of athlete’s predictions.

That huge set of data, however, miraculously puts the machine more volatile by readily dumping all 50,000 simulations of its previously held opinions as soon as it sees who wins Game One. It is still not clear why. Check out the top new sportsbooks for betting on the NBA Finals!

Inconsistent, wrongful projections

Early in May, a week after giving the Celtics a 39 per cent chance of winning the championship, FiveThirtyEight panicked and cut Boston’s chances to 18 per cent, which was half of what it gave to the Phoenix Suns. The Suns failed to advance to the Western Conference Finals, much less the NBA Finals.

In spite of this, the computer was adamant about dismissing the Warriors. FiveThirtyEight kept Golden State’s championship odds in the single digits or low double digits, despite the Warriors’ 12-4 playoff record. The computer predicted that the Warriors would lose to the Dallas Mavericks in the Western Conference finals by a margin of 73 per cent to 27 per cent before the series began. The Warriors defeated Dallas by a score of 115.2 to 106.6 over the course of five games.

[mytwitter url=”https://twitter.com/tomscocca/status/1532017296868315138″]

Until now, the FiveThirtyEight machine continues to insist that the Warriors still do not belong. According to its numeric rating system, they are still the eighth-best team in the NBA this year, trailing six teams that have already been eliminated — including the Brooklyn Nets, who were sent home in the early round, and the Los Angeles Clippers who failed to advance past the play-in games.

As of right now, Nate Silver’s model thinks that the Celtics will beat the Warriors even more than it thought the Mavericks would. Silver dismissed that fact, just like he did not answer when asked to defend his numbers back in April. He refused to bet $400 on the Celtics against $100 on the Warriors. On the other hand, the DraftKings Sportsbook picks the Warriors as the clear favorites.

Check out the top sportsbook bonuses for betting on the NBA!