Espn: 99.3% chance that Rockets would win (when there were 5 min left in 3rd)

[CelticsElite]

[azzenfrost]

hohoho

[SCeltic34]

[KGs Knee]

I mean, they're just using a statistical prediction model.  It's not a big deal.  Tonight's result was definitely an outlier.

[saltlover]

Houston hadn’t lost a game in which they led by at least 25 in 20 years, having won 179 such games in a row.  However, it was our first win after trailing by that much in a much shorter period, as we last came back from 26 vs. Miami on the final day of the 2016 regular season.

[GreenWarrior]

I stopped watching espn for their "analysis" 20 yr's ago.

[SparzWizard]

lol BSPN that likes to fabricate stuff and hates anything that's Boston, particularly the winning cultures of Boston.

[chilidawg]

I gave them about the same odds.  The ESPN paranoia here is just sad.

[kraidstar]

That stat ignores a few things:

1. C's are good at staging dramatic comebacks, especially at home.

2. Refs like to even things up in nationally televised games

3. Houston relies too much on 3's, easy to go cold, especially against a good D

4. Houston's interior D is awful

5. Harden is a choker, take away the whistles late and he gets exposed

[tenn_smoothie]

That running percentage is also shown on the "Gamecast" feature on ESPN website. It changes with each basket. It is just a computer-generated number that means about as much as half of these new "metrics" that have taken over the game - in other words, meaningless.

[green_bballers13]

I gave them about the same odds.  The ESPN paranoia here is just sad.

I think its prob b/c many Celtics fans are Patriots fans as well. ESPN is known as a primary suspect in the deflategate hatchet job.

As far as I know, they haven't done anything similar to the Celtics.

[bellerephon]

That running percentage is also shown on the "Gamecast" feature on ESPN website. It changes with each basket. It is just a computer-generated number that means about as much as half of these new "metrics" that have taken over the game - in other words, meaningless.

I agree that the win percentage is meaningless, but I think the reason why so many stats seem useless is that people don't use them correctly. Plus/minus is a good example, for one game, or even a few games it doesn't mean much, but over the course of a whole season it can provide some insight.

[Moranis]

7 out of 1000 seems about right to me