I don't think that's a fair assumption. Their front court is worse than it when they were good, and they're starting Andre Roberson after failing to plug the Harden and Sefolosha holes. They're overhyped for sure. I think they're more likely to have issues this year and finish in the bottom of the top eight.
except they weren't http://fivethirtyeight.com/features...fseason-pickups-should-give-the-cavs-a-boost/ Got almost half the east wrong and they did that preview AFTER PG's injury. Predicting who makes the playoffs in the west with perfect health is easy if that program had merit it would prove its worth at the EAST where there is constant bloodbath to make the pos. (except the cavs).
I thought the write-up was positive and fair. It clearly says that with good health, the Rockets will be contenders. It acknowledges that Dwight is still elite when healthy - it just correctly points out that he isn't always healthy. And the author explicitly disagrees with the model about Lawson, thinking the PG will exceed expectations. And it says Harden will likely cut down on turnovers. The math is the math - it reflects the injury history from last year. In the qualitative interpretation, the author gave the team plenty of respect while only really questioning health. Not every article needs to be a cheering section.
didn't know stats mattered all that much considering the MVP was inferior in stats to both Harden and Westbrook. Couldn't care less what some stats nerd says I watch games
Did they take into account our record last year despite the glut of injuries? Seems to me the drop off they predict is too severe. I question their methodology.
Read up on how they came up with the analysis. Capela didn't meet the criteria. He didn't have 100 NBA minutes and he didn't play in college. So they didn't have data. They know that the data is still wonky. It's a work in progress.
This is comparing who he is similar to in the future. With Howard's rate of decline it makes some sense. Miller had game back in the day.
Okay okay I read it. And it did shed more light. But you know who else they don't have the data for? 2 of the top 5 draft picks this year. 40% of the top 5. So if they nailed it with the other 3 guys, A++++ hitting the nail on the head 100% accurate, then at best their CARMELO accuracy of the top 5 picks of last year's draft receives an F in accuracy. They miss 3 guys if you extend it to the top 10 drafted last year. Wonky as in so much significant missing data that the next progression shouldn't be coming up with W-L predictions based off CARMELO. What's equally worse is then using bad data to also come up with wonky usage ratings & wonky rotations, and then bake that into the W-L predictions. Why can't they just use international data subbed in for college data? That's how international guys got WARP ratings, & CARMELO is based off PECOTA which as I understand is an extenuation of WARP.
It goes to show how complex predicting sports is when these nerds can nail election and after election yet still can't understand how basketball works.
Well that's because elections are based on who people will vote for. It's not hard to poll people in different areas and figure out how the election will go based on that data. They are essentially tallying votes before the actual "vote". So it's easy to see who is going to win.
why so logical Sam, leave that at the door of the GARM and be a homer, didn't you get the memo, everyone outside of houston hates the rockets!!!
