The 2018 draft

Fifty-six men in the file have a 2018 pick, more names than the older classes, because the source data gets deeper. The file itself runs out in 2025-26. From 2018 that is eight seasons. The menu will offer you one or five. It will not offer ten.

Eight seasons is the whole runway

A start year has to leave room for five seasons before it is listed at all, which is why 2018 is still available and 2021 is not. Room for five is not room for ten. 2018, 2019, 2020, 2021, 2022, 2023, 2024 and 2025 are what a run from here can touch. Any peak dated after 2025-26 does not exist in the game yet. Any peak inside those eight years is in play for a five-season run only if it falls in 2018-19 through 2022-23. A peak in 2023-24 or 2024-25 is real in the file and outside The Sprint.

That sounds like bookkeeping. It changes who is actually valuable. A man whose best year is 2024 is a fact about him, and a five-year run started in 2018 never plays that year.

The board, with the years attached

• Deandre Ayton, pick 1, Phoenix, centre. Rookie rating 69. Peak 71 in 2021. Both numbers are inside a five-year run. He is what the first pick looks like when the later years do not run away from the rookie year: good, and close to his ceiling in this file.
• Marvin Bagley III, pick 2, Sacramento, forward. Rookie rating 65, and 65 is the highest season the file has for him, in 2018 itself. There is no hidden peak waiting in 2022. The second name is the second name.
• Luka Dončić, pick 3, Atlanta, guard. Rookie rating 73, the best rookie number of the top of this class. Peak 89 in 2023. The file stores Atlanta as the club that drafted him. The 73 is already ELITE if it were a team rating, and it is simply his rookie season. The 89 falls in 2023-24, which a five-year run from 2018 does not reach. You get the early years, which start high. You do not get the peak unless the menu someday offers a length that includes it. It does not, today.
• Jaren Jackson Jr., pick 4, Memphis. Rookie 62, peak 77 in 2024. The peak is outside a five-year run.
• Trae Young, pick 5, Dallas, guard. Rookie 72, peak 84 in 2019. This peak is the following season. A one-year run misses it. A five-year run includes it. He is the rare man on this board whose best year in the file arrives almost immediately.
• Shai Gilgeous-Alexander, pick 11, Charlotte, guard. Rookie 60. Peak 92 in 2024, the highest peak in the class, and one of the highest in the whole file. 2024 is outside a five-season run from 2018. Inside the five years you can play, you have a 60 and whatever the seasons through 2022 actually rate, not the 92. Drafting him for the 92 is drafting a year your run is not allowed to reach.
• Jalen Brunson, pick 33, Dallas. Rookie 56, peak 78 in 2023. Same timing problem as the 92. The pick number is the bargain. The year of the peak is the catch.
• Michael Porter Jr., pick 14, Denver. No 2018 rating. First number 59 in 2019, peak 73 in 2025. He is on the 2018 board at the 59, then away until 2019. The peak is the last season in the file, well outside a five-year run.

Collin Sexton, pick 8 to Cleveland, peaks at 74 in 2023. Miles Bridges, pick 12, is stored against LAC and peaks at 72 in 2024. Both peaks sit late. The rookie ratings, 60 and 55, are what a Sprint can be sure of collecting in year one.

How to draft it anyway

Prefer the years you will actually play. Young's 84 is in range of a five-year run. Dončić's rookie 73 is in range on night one, and his 89 is not, if you only get five seasons. Gilgeous-Alexander's 92 is the best story in the class and a poor reason to spend a pick in a run that ends in 2023. Ayton's 69 is less exciting and more honest about the window. Bagley's 65 does not improve later in the file. Passing on him does not require a theory. It requires reading the column.

The other clubs still cannot see 2024. They can see a 73, a 72 and a 69. Your extra information is smaller here than in 1996, because the run is shorter than the careers. That is the late-era bargain. The classes are famous. The campaign is not long enough to redeem a pick whose payoff is the season after you stop.

Twenty seasons is on offer for every start year through 2006-07, so 2003 has it and 2018 does not. 1976 can run longer than twenty, out to the end of the file, but that is extra length, not the only place the twenty-season button exists. The same rule is on how to play: a length appears only when those seasons are still in the file.