Predictions for '88-89

Here they are:
Atlantic Division Central Division
Boston 55-27Chicago/Atlanta 56-26
Philadelphia/New York 44-38Atlanta/Chicago 55-27
New York/Philadelphia37-45Detroit49-33
Washington29-53CLV/IND/MIL46-36
Charlotte/New Jersey23-59IND/MIL/CLV45-37
New Jersey/Charlotte22-60MIL/CLV/IND44-38

Midwest Division Pacific Division
Dallas57-25Portland59-23
Utah53-29L.A. Lakers/Seattle55-27
Denver/Houston47-35Seattle/L.A. Lakers50-32
Houston/Denver 45-37GS/LAC/PHOE/SAC35-47
San Antonio30-52LAC/PHOE/SAC/GS30-52
Miami15-67PHOE/SAC/GS/LAC24-58
SAC/GS/LAC/PHOE 20-62

Where two teams are listed on one line, that means that either team could place in that spot, with the more likely one listed first. For example, it appears that either Philadelphia or New York could finish second in the Atlantic Division, but Philadelphia appears to be more likely to finish there. Both Philadelphia and New York have the potential to be up around 44 wins, but also could be as low as 37 wins.

I arrived at the records and rankings by looking at several things. First, each team's floor % and Pythagorean projection gave a general idea of what the record would be. Second, I looked at the five positions on the teams and assigned values for how far above or below average each position was. That usually helped narrow down a range of wins for every team. Third, I considered the possibility of the plexiglass principle taking effect. Fourth, I looked at possible anomalies in each team's won-loss record in close games, which is heavily subject to luck and usually balances out over an extended period. Finally, I got subjective. Generally, coming within four games of a final record looks to be about as good as is possible and four games is how much I allowed myself to subjectively change a team's statistically predicted record.

This is the first time I've done this and it probably has flaws. No, it definitely has flaws; I never do anything right the very first time. There are a few teams that I feel very uncomfortable about because of the drastic personnel changes they've made (Golden State, the Clippers, Phoenix, Chicago, and Atlanta). High college draft picks always cause uncertainties (the Clippers, Indiana, New Jersey, Golden State, Phoenix), but usually don't make much of an impact on a team's record. Rumors of certain players being injured hurts my confidence with other teams (Milwaukee, the Lakers, and Boston). Washington's record was literally pulled out of a hat. If everything turns out wrong, I'll scrap the system, but that shouldn't happen. The basic methods used have done rather well in other more limited and simple situations. I expect at least two very bad errors, but any more than five means something went haywire.

The Pacific Division is especially difficult this year as the bottom four teams are highly unpredictable because of their youth and changes in personnel. Golden State may be the best of the bunch, based on the veterans it has, but the youth in the other cities could easily make one or two of them very competitive teams.

Despite the division races, a division winner may not win the Championship next year. Detroit and Atlanta look to be the best bets. Dallas looks to be the best out west. Portland looks very strong and could win it all if they win 63+ games during the regular season, but the Blazers will probably lose in the second or third round of the playoffs. The Lakers can repeat again next year, but it's going to take a certifiable miracle and some controversy.


Basketball Hoopla, © 1988, L. Dean Oliver