Hybrid Auction/Draft

Fantasy sports leagues typically use either an auction or a draft to allocate players at the start of the season. Drafts are usually faster and simpler, but especially for the NBA right now, where two players stand out far above the rest, they mean that the draft order determines who can own LeBron James or Kevin Durant. If you get a low draft pick, you may never even get a chance to own your favorite player.
Auctions, where each owner has the same salary cap, puts all owners on a level playing field, giving each a chance to buy whomever they want. I strongly prefer auctions, but they do take longer than drafts. My long-time NBA league, the Ezra Stiles Rotisserie Association, is now in its 25th year, and this time we tried an experiment of a hybrid draft/auction.
We selected 80 players to bid on before the draft, and then after those 80 players were sold we switched to a straight round draft to fill out our rosters, with owners who had more money left getting higher picks in each round. This resulted in a faster draft than doing a full auction, but a slower one than a straight draft. But since the list of 80 players included all the top names, everyone still got a fair at Durant, LeBron or whoever they might want in the draft.
The league uses a $30 salary cap to fill a starting roster of 20 players (4 each at each position), and also a 5 person bench. We score 6 categories: points, rebounds, assists, FG%, FT%, and the sum of steals and blocks.
I generated the list by first computing some projected stats for NBA players, then computing a RotoValue based on those stats. I’ll do a separate post on those projections, but an updated version of them are visible here. I tweaked the list a bit, aiming to get the top 80 players in the league, and shuffled it, but biased so that better players would tend to be near the top.
Here was the list of 80 players we sold at auction, in auction order:

Each owner had $30 to spend on a total of 25 players, with a minimum bid of $0.10 (even for the players selected in the draft phase). So owners had $27.50 to allocate to extra bids. The list was randomized, with a bias so that better players tended to appear closer to the front.
While the 8 teams purchased a total of 80 players at auction, teams did vary in the number of players they bought. Tim won both LeBron James and Kevin Durant, the two highest priced players by far, but only got 3 other players, leaving him 20 to select in the draft phase. Joe and Lorin each bought 12 players.
After the auction phase, the draft order was set in order of decreasing remaining cash, so not spending as much at the auction meant having a better pick in each draft round. If two teams tied in cash remaining, the one needing fewer players got the earlier pick, and if that was tied also, we flipped a coin.
Here’s how the draft phase went:

This is  a rather deep league, as we draft 200 NBA players, ensuring that almost every NBA starter and 6th man is owned.
Doing a hybrid sped things up quite a bit, as even the draft phase moved along rather nicely. We got the benefit of bidding for the best players, while going faster than a full auction. While I still am partial to a full auction, this was a good compromise to balance the fairness and excitement of an auction for top talent with the speed of a draft.