Best option within reason
On the Draft Report, the last column asks a fair version of "who should they have taken?" The naive answer — the best player drafted at any later slot — is unfair: it would fault a team that picked 3rd for "missing" a star who slipped to 77th, when nobody had him there. So we bound the comparison to a reasonable range, and we let the draft value curve decide what reasonable means.
- Every draft slot is given a value from the curve below, built from the career WAR of everyone drafted there (2013–2021). It's normalized so 1st overall = 1.00, and every later pick is scaled to that.
- A later pick is "within reason" while it's still worth at least 50% of your pick on the curve. Because the curve runs about 1 ÷ pick number, that works out to roughly 2.1× your slot — the multiple is read straight off the curve's slope, not set by hand. So a #1 is measured against the next pick, a #10 against the next ~10, a #30 against the next ~30. Early picks are sharply distinct in value, so the window is tight; late picks are nearly interchangeable, so it's wide.
- The column shows the best player taken later within that window. If nobody in range beat your pick's own career, you took the best player available and it says so; otherwise it names the better player you passed on.
Value relative to the 1st-overall pick (= 1.00), from career WAR by slot, 2013–2021 skaters. The 1st pick is worth roughly twice the 5th and ~9× the 31st; by the third round the curve is nearly flat.
| Pick | Value (#1 = 1.00) | Within reason through | Window span |
|---|---|---|---|
| #1 | 1.000 | #2 | 1 pick |
| #5 | 0.478 | #10 | 5 picks |
| #10 | 0.366 | #21 | 11 picks |
| #20 | 0.237 | #41 | 21 picks |
| #30 | 0.126 | #62 | 32 picks |
| #60 | 0.049 | #124 | 64 picks |
| #90 | 0.038 | #186 | 96 picks |
| #160 | 0.023 | #224 | 64 picks |
| #224 | 0.007 | #224 | 0 picks |
A description, not a grade — career WAR is heavy with luck, and whether anyone drafts better than the field at all is settled by the null on Scouting Value. The companion Pick Value page shows this curve next to the published Dom Luszczyszyn and Eric Tulsky models. The window is read off the curve: a later pick counts while it's worth ≥ 50% of yours, ≈ 2.1× the slot. Tighter than that ratio is one knob if you want it.