2010 NBA Playoffs Predictions: Conference Finals

posted in NBA

The Celtics’ unexpected victory over Cleveland knocked out our second most likely 2010 NBA Champion. As of the start of the playoffs, our power ratings based analysis gave Cleveland a 27% chance to win it all, second only to Orlando.

The Celtics, on the other hand, projected as having about an 18% chance to make the Eastern conference finals. Not good, but still probably higher than the average fan thought, we’d wager.

Health has become quite the popular topic in terms of the Celtics, and it’s a factor that can potentially wreak a bit of havoc on data-driven prediction systems like ours. If it’s true that the Celtics have achieved a significantly superior state of health (particularly regarding Kevin Garnett) compared to most of the regular season, power ratings based on 2009-10 season data likely understate their playoff performance capabilities.

For now, we’ll leave that to the sportscasters to debate and do the best with the data we’ve got. Our unofficial entry in the TrueHoop Stat Geek Smackdown should now be in third place, one measly point behind John Hollinger and Jeff Ma, so we’re out to get first place back after winning round one.

Here are our simulation numbers for the Eastern Conference finals. In short, the Celtics have been playing well lately, but the Magic have been playing better according to our Last 10 Games power ratings. Our projections for the series indicate the Magic will win it a little over 80% of the time.

Out of 10,000 series simulations:

– The series goes 4 games 1885 times; the Magic win 93.8% of them
– The series goes 5 games 3226 times; the Magic win 89.6% of them
– The series goes 6 games 2556 times; the Magic win 74.3% of them
– The series goes 7 games 2333 times; the Magic win 72.9% of them

Based on this analysis, there’s about a 29% chance of the Magic beating the Celtics in 5 games. The next most likely result (it’s close) is Magic in 6. Orlando has about an 18% chance of sweeping Boston, while Boston has about a 1.2% chance of sweeping Orlando.

Vegas odds indicate implied series win odds for the Magic of about 70%, compared to our ~80%.

At the end of the day, our somewhat injury-blind numbers may be underestimating Boston a bit, but we’ll stick with Orlando over Boston in 5 games, along with our second straight bet-against-the-Lakers contrarian pick, Phoenix over LA in 6 games.

See our earlier post for our breakdown of the Suns-Lakers Western Conference finals series.

  • Arty Blackwell

    This website has lost ALL credibility when you first said l.a would lose to Utah in 7 -_-.
    then you say Phoenix will win in 6? and they have a better chance to sweep l.a than l.a does to sweep them?
    you know 0 about basketball..i cant wait to check back in a minute

  • Tom (TeamRankings)


    OK Arty, thanks for your opinion. Have a great day. You’re welcome to read our other blog posts for deeper explanations of the rationale behind the picks and some possible risks given the underlying data.

    Before you come on our message boards and accuse us of having no credibility though, perhaps you should do a little checking into our long term track record, which speaks for itself.

    P.S. Wonder what you would have said if we picked the Celtics to smoke the Cavs in 6 games before that series started. Because clearly only someone who knows 0 about basketball would have made that prediction, right?

  • Tom (TeamRankings)

    Looks like John Hollinger has no basketball credibility either. Wonder why ESPN pays him all that dough?


  • flotilla

    Hah ha! Ex-gambler Voulgaris is in the last place. Even Henry Abbott’s mom has beaten him so far.