A middle is not a hedge — it is an option that converts a strong position into two chances to win. When an early spread bet moves through 3 or 7, the right question is not whether to lock a guaranteed profit but whether the buyback price creates a window where both tickets cash more often than the juice costs.
Identifying the middle window
Suppose you bought the Eagles -2.5 on Sunday morning. By Saturday night the line moved to Eagles -4. You can now take Cowboys +4 as a buyback. Your window: any Eagles win by exactly 3 middles both tickets, and any Cowboys cover (win, or loss by 3 or fewer) wins the second ticket. The Eagles -2.5 ticket still wins any Eagles win by 3 or more.
| Final margin | Eagles -2.5 result | Cowboys +4 result | Net |
|---|---|---|---|
| Eagles win by 5+ | Win | Loss | +1 unit (net of juice) |
| Eagles win by 3 | Win | Win | +2 units (middle) |
| Eagles win by 1–2 | Loss | Win | Roughly flat (juice) |
| Tie / Cowboys win | Loss | Win | +1 unit (net of juice) |
The 3-point middle occurs in about 15% of NFL games. At -110 on both legs, the combined juice is 9.5 cents per unit. The middle EV: 0.15 × 2 units = 0.30 units expected, minus 0.095 juice per unit on the losing leg = positive in most scenarios. The math works when the window contains a key number and both tickets are near -110.
When to take the buyback vs hold the original
Not every line move justifies a buyback. If the line moved from -2.5 to -3 (across 3), buying back at +3 gives you no window — both tickets sit on the same number. If it moved from -2.5 to -3.5, you have a window of exactly 1 point (3 to 3.5) which is too narrow to be worth the juice. The window needs to span at least 1.5 points and contain a key number to be worth playing.
Also check whether taking the buyback converts your original edge into a juice trap. If you got Eagles -2.5 because you had a genuine edge on an underpriced spread, the buyback at +4 captures the middle but also means you now lose money on your original position whenever the margin is between 0 and 2.5. Do not buyback a position that moved in your favor just to create a middle — protect positive CLV bets. See closing-line value and bet tracking for how to log these setups correctly.
Middle math: the price that makes it work
The key formula: (key-number frequency) × 2 > (combined juice cost per unit). At -110/-110, the combined cost is about 9.5 cents. For a 3-point middle (15% frequency): 0.15 × 2 = 0.30, which exceeds 0.095 — profitable. For a 7-point middle (9% frequency): 0.09 × 2 = 0.18, still above the juice bar. For a non-key middle centered on 6 (5% frequency): 0.05 × 2 = 0.10, barely above break-even — not worth the effort.
Juice degrades the math quickly. At -120/-120 combined cost is 18.2 cents. A 3-point middle at -120/-120 barely breaks even (0.30 vs 0.18 per unit at risk). Never pay more than -120 on either leg of a middle, and avoid any window that does not contain 3 or 7.
NFL ATS cover-margin distribution
Bars count completed NFL schedule rows by closing-spread cover margin using the repository canonical home-margin grading convention.
