Rummpd
03-21-2011, 10:58 AM
The NBA playoffs are just around the corner, and that means it's time for another spring tradition as well: Accuscore's Playoff Predictor.
http://insider.espn.go.com/nba/insider/columns/story?columnist=hollinger_john&page=PERDiem-110315
This neat little tool allows you to play out the postseason as many times as you like and simulate what might happen if different teams met ... or what the odds are of different outcomes in later rounds ... or what might happen if the bracket changed ... or countless other possibilities.
First, the nitty-gritty stuff. There was some confusion about this a year ago, so just to be clear: This is Accuscore's baby, not mine. My Power Rankings have nothing to do with it, although you can fill in the bracket using the current Playoff Odds.
Instead, their simulation tool looks at every matchup and plays out the series, showing you the percentage odds of each side winning and then advancing the winner into the next round.
Just looking at the early rounds, it immediately becomes clear that for anyone in the East, avoiding New York is the key. Boston, for instance, has an 88 percent chance of beating projected No. 7 Philadelphia and a 90 percent chance of beating projected No. 8 Indiana ... but shows up only as a 70-30 proposition against the Knicks. Similar odds face the Bulls and Heat in their prospective matchups.
Meanwhile, it appears the 4-5 series will be a lot more competitive in the West than the East. Orlando has an 88 percent chance of vanquishing No. 5 Atlanta in the East, belying the closeness of their seedings, but the Thunder have only a 65 percent chance of advancing against Denver in the West. (This assumes home-court advantage for the Thunder, which we might not want to assume so cavalierly: Denver is three games behind and still plays Oklahoma City twice. If the Nuggets win both, they're a game back and own the tiebreak.)
We can also set up our own brackets using the "Custom Rankings" feature at the bottom right, which is useful for two reasons. First, we can see what happens if home-court advantage flips in a series; it appears it changes the odds by about 8 percent to 12 percent, depending on the series. For the Denver-Oklahoma City case, for example, the Thunder's odds of advancing go from 65 percent with home-court advantage to 53 percent without it.
These results imply that home-court advantage in a series increases the odds of winning by about 5 percent compared to a neutral-site seventh game, which sounds about right; after all, the series first has to be close enough to warrant a seventh game before the advantage becomes a factor.
But I digress. The real fun of the "custom bracket" tool is that it allows you to rig the bracket to see what would happen if two teams met in the Finals, rather than replaying the matchups a hundred times before you get what you want. If, for instance, you hanker to see what would happen if Miami played the Lakers, no problem: Just insert a series of Clevelands and Sacramentos into the bracket with them, and voila ... after sweating out Detroit, Toronto and Milwaukee in the Eastern playoffs, the Heat make the Finals and face a Lakers team that outlasted the Clippers, Phoenix and Utah.
When they meet in the Finals, the Lakers beat Miami in 62 percent of simulations, assuming home-court advantage for L.A., and 54 percent of simulations when the Heat have the home floor. (In the custom set-up, it appears the tool gives home court to the higher seeded team and goes to the team with the best record if they have the same seed.)
After running through a series of operations like this, I was able to reach some pretty solid conclusions about the simulation. Since most of these jive with my impression of the playoffs in general, here are some observations:
1. First-round upsets are much more likely in the West.
In addition to Denver, I'm fond of Memphis and Portland as first-round upstarts, but this tool liked another team even more -- New Orleans. It gave the Hornets a 33 percent chance of beating Dallas in Round 1, and several iterations placed the Hornets into the conference finals ... I even got them in the Finals once, after a riveting conference finals against Memphis that was the lowest-rated in league history. (The simulation didn't actually say that part, I just figured it was likely.)
That said, Memphis, Portland and Denver all had better odds than similarly seeded teams in the East; the likely No. 6 seeds were the only slot where the Eastern team had a better chance, and only marginally so (35 percent for New York to 33 percent for New Orleans).
2. The last two rounds of the East playoffs are basically toss-ups.
Provided Chicago takes the top seed, which it appears the Bulls will, the 1-4 and 2-3 matchups in the East become coin tosses. The biggest difference is between Chicago and Orlando, where the Bulls have a 56 percent chance of advancing; the tightest is a Miami-Chicago conference finals, which gives the Heat a 51 percent chance.
Take away home-court advantage from the Bulls, however, and this tool is much less bullish on their chances. In a conference finals, Chicago has a 53 percent chance of beating the Celtics with home-court advantage, but only 43 percent without it.
3. Lakers or Spurs? Take your pick.
This tool is not a big fan of Dallas, giving the Mavs a decent shot of going out in the first round against New Orleans (33 percent) or Portland (23 percent), and preferring the Lakers 54-46 even if the Mavericks have home-court advantage; without it the preference for L.A. goes to 62-38.
Meanwhile, the tool gives the Thunder a decent chance against San Antonio in the second round (31 percent), which is almost as good as Dallas' chances against the Lakers. As a result, San Antonio's seeding doesn't appear much more advantageous than L.A.'s.
In the conference finals, the app sees Lakers-Spurs as a 50-50 proposition -- one of only a few matchups right now that rate as true coin flips. (Among the others are a Dallas-Chicago Finals and a Denver-Memphis second-round pairing.)
4. In the Finals, go Boston or go West.
Despite all the power in the East this season, Boston was the only team that fared well in Finals matchups. The Celtics win 52 percent of the time against L.A., 50 percent against San Antonio and 57 percent against Dallas.
Their East brethren don't fare nearly as well. San Antonio beats Miami 59 percent of the time, Orlando 60 percent of the time and thumps Chicago 65 percent of the time. In addition to the slight edge against Miami noted above, the Lakers pound Orlando (65 percent) and eke past Chicago (51 percent). Even Dallas has a decent chance, rating a pick-'em against the Bulls, 53 percent against the Heat and 57 percent against Orlando.
In closing, let me sum it all up for you. I made this handy chart to show you how all the contenders fare against each other based on today's ratings. Obviously, it's going to change every day, which is part of the fun -- you can go back in every morning and re-simulate the playoffs with the latest seedings and information, and you can keep trying until you get that Hawks-Blazers Finals you've always wanted.
Contenders vs. each other in Accuscore Simulator (home-court advantage team listed first)
EAST SEMIS
Chicago 56% Orlando 44%
Boston 54% Miami 46%
Boston 52% Orlando 48%
Chicago 49% Miami 51%
WEST SEMIS
San Antonio 69% Oklahoma City 31%
Los Angeles 62% Dallas 38%
Dallas 46% Los Angeles 54%
EAST FINALS
Chicago 53% Boston 47%
Boston 57% Chicago 43%
Chicago 49% Miami 51%
Boston 52% Orlando 48%
Miami 56% Orlando 44%
Chicago 56% Orlando 44%
WEST FINALS
San Antonio 50% Los Angeles 50%
San Antonio 54% Dallas 46%
NBA FINALS
San Antonio 54% Boston 46%
Boston 57% Dallas 43%
Boston 52% Los Angeles 48%
San Antonio 65% Chicago 35%
Chicago 50% Dallas 50%
Chicago 49% Los Angeles 51%
San Antonio 60% Orlando 40%
Dallas 57% Orlando 43%
Los Angeles 65% Orlando 35%
San Antonio 59% Miami 41%
Dallas 53% Miami 47%
Los Angeles 62% Miami 38%
Miami 46% Los Angeles 54%
http://insider.espn.go.com/nba/insider/columns/story?columnist=hollinger_john&page=PERDiem-110315
This neat little tool allows you to play out the postseason as many times as you like and simulate what might happen if different teams met ... or what the odds are of different outcomes in later rounds ... or what might happen if the bracket changed ... or countless other possibilities.
First, the nitty-gritty stuff. There was some confusion about this a year ago, so just to be clear: This is Accuscore's baby, not mine. My Power Rankings have nothing to do with it, although you can fill in the bracket using the current Playoff Odds.
Instead, their simulation tool looks at every matchup and plays out the series, showing you the percentage odds of each side winning and then advancing the winner into the next round.
Just looking at the early rounds, it immediately becomes clear that for anyone in the East, avoiding New York is the key. Boston, for instance, has an 88 percent chance of beating projected No. 7 Philadelphia and a 90 percent chance of beating projected No. 8 Indiana ... but shows up only as a 70-30 proposition against the Knicks. Similar odds face the Bulls and Heat in their prospective matchups.
Meanwhile, it appears the 4-5 series will be a lot more competitive in the West than the East. Orlando has an 88 percent chance of vanquishing No. 5 Atlanta in the East, belying the closeness of their seedings, but the Thunder have only a 65 percent chance of advancing against Denver in the West. (This assumes home-court advantage for the Thunder, which we might not want to assume so cavalierly: Denver is three games behind and still plays Oklahoma City twice. If the Nuggets win both, they're a game back and own the tiebreak.)
We can also set up our own brackets using the "Custom Rankings" feature at the bottom right, which is useful for two reasons. First, we can see what happens if home-court advantage flips in a series; it appears it changes the odds by about 8 percent to 12 percent, depending on the series. For the Denver-Oklahoma City case, for example, the Thunder's odds of advancing go from 65 percent with home-court advantage to 53 percent without it.
These results imply that home-court advantage in a series increases the odds of winning by about 5 percent compared to a neutral-site seventh game, which sounds about right; after all, the series first has to be close enough to warrant a seventh game before the advantage becomes a factor.
But I digress. The real fun of the "custom bracket" tool is that it allows you to rig the bracket to see what would happen if two teams met in the Finals, rather than replaying the matchups a hundred times before you get what you want. If, for instance, you hanker to see what would happen if Miami played the Lakers, no problem: Just insert a series of Clevelands and Sacramentos into the bracket with them, and voila ... after sweating out Detroit, Toronto and Milwaukee in the Eastern playoffs, the Heat make the Finals and face a Lakers team that outlasted the Clippers, Phoenix and Utah.
When they meet in the Finals, the Lakers beat Miami in 62 percent of simulations, assuming home-court advantage for L.A., and 54 percent of simulations when the Heat have the home floor. (In the custom set-up, it appears the tool gives home court to the higher seeded team and goes to the team with the best record if they have the same seed.)
After running through a series of operations like this, I was able to reach some pretty solid conclusions about the simulation. Since most of these jive with my impression of the playoffs in general, here are some observations:
1. First-round upsets are much more likely in the West.
In addition to Denver, I'm fond of Memphis and Portland as first-round upstarts, but this tool liked another team even more -- New Orleans. It gave the Hornets a 33 percent chance of beating Dallas in Round 1, and several iterations placed the Hornets into the conference finals ... I even got them in the Finals once, after a riveting conference finals against Memphis that was the lowest-rated in league history. (The simulation didn't actually say that part, I just figured it was likely.)
That said, Memphis, Portland and Denver all had better odds than similarly seeded teams in the East; the likely No. 6 seeds were the only slot where the Eastern team had a better chance, and only marginally so (35 percent for New York to 33 percent for New Orleans).
2. The last two rounds of the East playoffs are basically toss-ups.
Provided Chicago takes the top seed, which it appears the Bulls will, the 1-4 and 2-3 matchups in the East become coin tosses. The biggest difference is between Chicago and Orlando, where the Bulls have a 56 percent chance of advancing; the tightest is a Miami-Chicago conference finals, which gives the Heat a 51 percent chance.
Take away home-court advantage from the Bulls, however, and this tool is much less bullish on their chances. In a conference finals, Chicago has a 53 percent chance of beating the Celtics with home-court advantage, but only 43 percent without it.
3. Lakers or Spurs? Take your pick.
This tool is not a big fan of Dallas, giving the Mavs a decent shot of going out in the first round against New Orleans (33 percent) or Portland (23 percent), and preferring the Lakers 54-46 even if the Mavericks have home-court advantage; without it the preference for L.A. goes to 62-38.
Meanwhile, the tool gives the Thunder a decent chance against San Antonio in the second round (31 percent), which is almost as good as Dallas' chances against the Lakers. As a result, San Antonio's seeding doesn't appear much more advantageous than L.A.'s.
In the conference finals, the app sees Lakers-Spurs as a 50-50 proposition -- one of only a few matchups right now that rate as true coin flips. (Among the others are a Dallas-Chicago Finals and a Denver-Memphis second-round pairing.)
4. In the Finals, go Boston or go West.
Despite all the power in the East this season, Boston was the only team that fared well in Finals matchups. The Celtics win 52 percent of the time against L.A., 50 percent against San Antonio and 57 percent against Dallas.
Their East brethren don't fare nearly as well. San Antonio beats Miami 59 percent of the time, Orlando 60 percent of the time and thumps Chicago 65 percent of the time. In addition to the slight edge against Miami noted above, the Lakers pound Orlando (65 percent) and eke past Chicago (51 percent). Even Dallas has a decent chance, rating a pick-'em against the Bulls, 53 percent against the Heat and 57 percent against Orlando.
In closing, let me sum it all up for you. I made this handy chart to show you how all the contenders fare against each other based on today's ratings. Obviously, it's going to change every day, which is part of the fun -- you can go back in every morning and re-simulate the playoffs with the latest seedings and information, and you can keep trying until you get that Hawks-Blazers Finals you've always wanted.
Contenders vs. each other in Accuscore Simulator (home-court advantage team listed first)
EAST SEMIS
Chicago 56% Orlando 44%
Boston 54% Miami 46%
Boston 52% Orlando 48%
Chicago 49% Miami 51%
WEST SEMIS
San Antonio 69% Oklahoma City 31%
Los Angeles 62% Dallas 38%
Dallas 46% Los Angeles 54%
EAST FINALS
Chicago 53% Boston 47%
Boston 57% Chicago 43%
Chicago 49% Miami 51%
Boston 52% Orlando 48%
Miami 56% Orlando 44%
Chicago 56% Orlando 44%
WEST FINALS
San Antonio 50% Los Angeles 50%
San Antonio 54% Dallas 46%
NBA FINALS
San Antonio 54% Boston 46%
Boston 57% Dallas 43%
Boston 52% Los Angeles 48%
San Antonio 65% Chicago 35%
Chicago 50% Dallas 50%
Chicago 49% Los Angeles 51%
San Antonio 60% Orlando 40%
Dallas 57% Orlando 43%
Los Angeles 65% Orlando 35%
San Antonio 59% Miami 41%
Dallas 53% Miami 47%
Los Angeles 62% Miami 38%
Miami 46% Los Angeles 54%