ESPN's BPI Model currently projects the Celtics, 76ers, Bucks, & Heat all finishing the season at 51-31 in a 4-way tie for the 1st seed
Correction: The tiebreaker would first apply to the division, so 76ers vs Celtics. They are 2-2 against each other, so it would go to division record, with Celtics at 9-7 & the 76ers at 6-9. So likely the Celtics would win that division tiebreaker, dropping the 76ers to the 4th seed.
It would then be a 3-way tie between the division winners: Celtics, Heat, Bucks, with shared winning % determining the winner.
Celtics record vs. Heat/Bucks: 4-1
Heat vs. Celtics/Bucks: 2-4
Bucks vs. Heat/Celtics: 3-4
So most likely the Celtics would come out of that 3-way tie with the 1 seed, then based on current record it would be the Bucks with the 2-seed, & the Heat with the 3-seed.
Obviously all this is still up in the air, there's a 76ers-Bucks game today, Celtics-Heat tomorrow, & a Celtics-Bucks game on April 7th.
https://www.espn.com/nba/story/_/page/BPI-Playoff-Odds/espn-nba-basketball-power-index-playoff-oddsFor comparison the FiveThirtyEight model projects the 76ers at 52-30, then a 3-way tie at 51-31 for the 2nd seed among Bucks, Celtics, & Heat.
It’s a travesty that NBA secondary standings aren’t factored by points scored/against %