What a player is "worth" on a fantasy draft board is not how many points they score. It is how many points they score above the alternative you could still start. That alternative is called replacement level. The gap is value over replacement (VORP) — the same idea as VBD (value-based drafting), applied to your roster slots, scoring, and free-agent pool.
Generic ranks skip that step. They sort raw projection (or consensus vibes) and hope your league looks like the sample. LeagueFrenzy does not. This post is the engine's receipts: how replacement and VORP work, why IDP labels and thin pools make the math hard, and why a coherent board is useful once the bar is honest.
How it works — replacement and surplus
Replacement level at a position is the projected points of the Nth-best starter-quality body there. N is roughly:
teams × starting slots at that position (+ a share of flex / superflex / IDP flex)
If you start one quarterback in a 12-team league, N is about 12: the 12th-best QB is still quite good, so replacement sits high and only the true elites clear a big surplus. If you start three RBs plus flex that soaks up more backs, N might land in the mid-30s to mid-40s: replacement is a deep committee body, so the top backs open a huge gap.
That is pure scarcity math. Projected points do not change when you add a superflex slot; replacement falls, so every QB's surplus rises in lockstep. We measured that on a fixed research pool in The Superflex QB Premium, Quantified — mean +257.5 VORP with zero change in projected points. The premium is replacement and roster construction, not magic yardage.
A deep IDP dynasty shape (offense first)
Imagine a deep IDP dynasty with a shallow QB start and heavy skill + flex:
| Position shape | Why N is high or low | What replacement "feels like" |
|---|---|---|
| 1 QB / team | Shallow N (~teams) | 12th–14th QB still elite; surplus is thin above the bar |
| Deep RB/WR + flex | N in the 35–45 range | Last startable body is a committee / WR4; stars open huge gaps |
| 1 CB vs 2 S | Slot counts set which ranked body is last-startable | Last startable corner sits higher on the projection curve than last startable safety |
The board is not "who is good." It is "who is good given how many of this position you must start."
VORP-S vs VORP-D
LeagueFrenzy separates two surplus numbers you will see in product copy and roster breakdowns:
- VORP-S (single-season): this year's projected points minus replacement. It answers "who should start now?" A veteran safety projecting just above the bar can seat over a rookie who is still ramping.
- VORP-D (dynasty): the same replacement baseline, summed over a multi-year horizon with an age-based development curve and discount. Year zero matches today's projection; later years ramp young players, peak by position, then decline. That is why a young safety like Caleb Downs can show a big dynasty number even on a deep position — multi-year surplus compounds while single-season surplus stays modest until the curve catches up.
Same player, same league, two questions. Mixing them is how you bench a productive veteran for "upside" that does not clear this year's bar — or dump a dynasty asset because this season's VORP-S looks quiet.
Slot counts matter on defense too
Offense gets the airtime, but IDP is where slot design rewrites the board. Start one CB and two safeties and the last startable corner is a higher-ranked body than the last startable safety: fewer corner seats means a higher replacement bar and less room for mid-tier corners to look "startable." Flip the slots and the curve flips with them. That is value-based drafting applied to fine defensive positions, not a vibes tier list.
Why it's hard — labels, thin pools, and guardrails
If replacement is "Nth-best at the position," you need a coherent position. That sounds trivial until you leave standard PPR offense.
Fine labels vs coarse buckets
Platforms and data feeds disagree on fine positions (CB, S, DT, DE) versus coarse ones (DB, DL). The same human can land in different buckets depending on who last wrote the roster file.
Patrick Surtain is the clean teaching case: treat him as a CB and he competes in the corner pool against other corners for CB slots. Fold him into a generic DB bucket and he shares a pool with safeties under different slot rules and a different N. Same player, different replacement bar, different surplus, different board rank. That is not a taste debate; it is a bucket error.
When fine signal is missing or overwritten, thin IDP pools get worse: everyone collapses into a coarse shelf, archetype floors invent depth that is not really there, and the Nth-best "DB" stops meaning anything a manager would start.
We treat keep-more-specific on upsert as a product rule for a reason: once you have a fine label, do not casually coarsen it away on the next sync. Teaching that story once is cheaper than rediscovering it every August.
Guardrails that can backfire (and what we fixed)
Replacement needs guardrails. A thin scored pool full of zeros can crater the bar and invent huge VORP for anyone with a non-zero projection. So floors exist — archetype-implied Nth-best under this league's scoring — so a starved CB/DT/K pool cannot mint fake surplus.
Floors help until the pool empties or the floor climbs above the real player ceiling. Then every real corner looks "below replacement," surplus goes negative league-wide, and the board stops ranking humans.
We hit that shape in a real investigation on a deep IDP dynasty board with separate CB and safety slots: CB replacement sat near ~223 projected points while the best real corner in the pool projected ~191. Replacement had climbed above the actual ceiling. That is not "corners are bad"; that is a broken bar.
The fix path (already shipped) is the teaching story, not a live bug report:
- Keep fine positions when the feed tries to coarsen them — so CB stays CB and the pool does not empty into DB soup.
- Cap the floor so it cannot exceed the covered pool — a guardrail meant to stop invented value must not invent a higher bar than any real player can clear.
Frame it that way when you read older screenshots or war stories: investigation receipts, then guardrails. Current boards are not claiming a 223 CB waterline against a 191 ceiling.
Everything interacts
Dynasty horizon, market blending, FA-aware replacement, and scarcity all move together. Raise free-agent quality and the bar lifts (a starter is only worth the alternative you can pick up). Blend market consensus so the board stays grounded, and you still need VORP to carry the league-specific half — especially on IDP, where market signal is thinner than offense. None of that is a secret sauce dump of weights; it is why "just sort by projection" fails the moment two leagues share a player pool but not a roster sheet. Deeper pipeline notes live in how the value engine works.
Why it's useful — scarcity- and roster-aware valuation
Once replacement is coherent, the board answers questions raw points cannot:
- Who is actually worth a pick given your slots and who is still free — not who would rank well in a default 1QB PPR sample.
- Startable free agents above the bar: if two startable tight ends sit unrostered, burning a rookie pick on TE is usually a mistake, and the board should say so.
- Position strength vs league median: you see whether your LB room is deep above replacement or paper-thin, which changes trade and draft urgency.
- Upgrade surfaces: the gap between your worst starter and the next free agent is the real "need," not a color on a need chart that ignores the wire.
That is the same edge we walked through in Why Generic Rankings Lose IDP Drafts: consensus boards optimize for a league you do not play. VORP against your replacement level is the antidote — including on IDP and superflex shapes where the bar moves hardest.
We are not claiming a mystical learning loop that auto-wins every draft. We are claiming something narrower and testable: scarcity-aware surplus, scored under your rules, with honest position buckets. Do the math once; stop redoing the spreadsheet every time scoring or slots change.
Methodology and links
If you want the longer hubs behind this pillar:
- Value-based drafting — why replacement baseline, not raw points, ranks the board
- How the value engine works — rates → league-scored projections → VORP → market blend (mechanisms, not secret constants)
- Platform guides: MyFantasyLeague · Sleeper
- Format pages: IDP rankings · Superflex rankings
- Sibling posts: generic rankings and IDP · superflex QB premium
Illustrative numbers in this post (the CB floor investigation, named teaching cases) are receipts from a real investigation, not a live re-pull of any private league. Re-run the engine against your league for ranks you should draft from.
LeagueFrenzy is live now. Paste your league URL and see your board free — replacement, VORP-S vs VORP-D, and a board built for the slots you actually start.