Signal Syndicate Daily Analysis — Sep 21, 2026

3 signals graded, 1-2-0 record, -46.00% ROI, -1.38u — computed from the append-only settled ledger.

Performance Snapshot

All models incl. shadow · 1707 graded signals · 1098-601-8 · +7.13u · Live-promoted: 356 · 216-140-0 · +28.39u · updated at build time (PG)

Cumulative units equity curve (live models)

Cumulative units by model

Run Lines
+36.09u
Hits Props OVER
-37.59u
Hits Props UNDER
+5.13u
MLB Totals UNDER
+7.91u
F5 Innings
-0.82u
NFL Spreads
+1.69u
NFL Totals
-5.27u

> Ledger summary

1-2-0 · -1.38u · -46.0% ROI — 3 LIVE signals graded on 2026-09-20. Honest ledger, nothing hidden.
Shadow research (not official portfolio): 40-14-0 · +9.02u. Total portfolio (all lanes): 41-16-0 · +7.64u.

Record Units ROI
**Live** 1-2-0 -1.38u -46.0%
**Shadow** 40-14-0 +9.02u 16.7%
**Total** 41-16-0 +7.64u 13.4%

3 signals settled from yesterday's card.

What Happened Yesterday

Sunday brought three run-line³ signals and a reminder that models misfire. Two underdogs failed to cover the spread. Washington Nationals at St. Louis Cardinals fell short at +1.5, as did Boston Red Sox at Tampa Bay Rays. Toronto Blue Jays at Texas Rangers grabbed the only win, returning 0.62 units at +1.5. Net damage: minus 1.38 units on a 1-2 card.

CLV¹ on those three plays provided context. Median closing line value sat at plus 2.05 percent, mean at plus 2.87 percent — the entry price held up through settlement. Over the trailing seven days, run lines carried a median CLV of plus 2.05 percent with a mean of plus 13.51 percent. All-time, the live book shows a median of plus 0.88 percent and a mean of plus 6.89 percent across 116 settled signals. That gap between median and mean flags outliers — a few large moves driving the average up while the typical signal hovers near one percent. The cumulative live record now stands at 117-73-0, plus 30.99 units, 16.31 percent ROI across 190 signals.

Shadow research exists to test frameworks before they touch real money. Yesterday it ran 40-14 across four lanes, returning 9.02 units. Hits Props UNDER went 23-4, producing 7.10 units with a median CLV of zero and a mean of minus 12.68 percent. F5 Innings⁵ ran a perfect 3-0, adding 2.73 units; that lane tracks line movement in runs rather than percentage, posting a median of minus 4.00 runs and a mean of minus 3.83 runs. Hits Props OVER went 7-5, dropping 2.13 units with a median CLV of minus 35.45 percent — a poor entry price on a split result. Those shadow lanes sit at 827-431-4 cumulative, minus 6.41 units, across 1,262 signals. Total portfolio activity yesterday: 41-16, plus 7.64 units across 57 signals.

The projection engine covered 270 players across 15 games. Largest category deltas sat at minus 0.36 for total bases and minus 0.22 for hits. If we project a hitter for 1.2 total bases and he finishes with 0.84, the model missed by 0.36. Those deltas feed parlay builders and DFS players hunting edge against posted lines.

How the Projection Engine Did

The research projection scoreboard is offline for this cycle — no per-category level or skill-versus-baseline data to report. The engine generated 1,350 player projections yesterday, but the internal benchmarking that measures whether it sets the right overall rate and beats each player's season average is not available.

When the scoreboard returns, it will show mean absolute error⁸ per category, percentage of projections landing within one unit of settlement, and skill margin against baseline. Until then, the engine runs and the output feeds the board, but we cannot claim precision we have not measured.

Today's Card

Monday brings a three-game slate. Run lines generated two signals; Hits Props OVER generated none. The projection engine covers 270 players across those three contests. One run-line candidate cleared the Kelly gate: Toronto Blue Jays plus 1.5 at Baltimore Orioles, best price BetRivers at plus 160. The rest — including all shadow research — lives on the Signal Board.

What This Means

A losing day does not erase the process. CLV measures whether we are getting value at entry, and the live book continues to capture plus-median closing line movement. Small samples swing. The ledger stays honest.

Dig Deeper


Every figure computed from the append-only settled ledger. Educational research only — not betting advice.
Data: Signal Platform · Edited by: Signal Desk


Definitions:

  • CLV¹ (Closing Line Value): difference between entry price and closing price, expressed as a percentage. Positive CLV means the line moved in our favor after entry. Median is the robust center; mean includes outlier sensitivity.
  • Delta²: difference between projected value and actual settled value for a player stat.
  • Run line³: spread betting in baseball, typically ±1.5 runs.
  • Hits props⁴: over/under markets on a player's total hits in a game.
  • F5⁵ (First Five Innings): markets that settle after five innings rather than a full game.
  • ROI/units⁶: return on investment and standardized bet sizing (one unit = one standard wager).
  • LIVE/SHADOW/total⁷: LIVE = official published board; SHADOW = research/validation lanes; TOTAL = combined activity across all lanes.
  • Mean absolute error⁸ (MAE): average size of projection misses, ignoring direction.

Definitions

  • ¹ CLV (Closing Line Value) — whether the price we got beat the closing price. A measure of process quality, not profit or loss.
  • ² Delta — the gap between our projection and the market's line for a player prop; the side where we'd have value.
  • ³ Run Line — baseball's spread bet, typically ±1.5 runs.
  • ⁴ Hits Props OVER/UNDER — a bet on whether a player records more or fewer hits than the posted line.
  • ⁵ F5 Innings — first-five-innings result, isolating the starting pitchers.
  • ⁶ ROI / units (u) — return per unit staked; +1.00u = one unit of profit.
  • ⁷ LIVE / SHADOW / total portfolio — LIVE is the official real-money book; SHADOW is research tracked publicly but excluded from the official record; total = live + shadow combined.
  • ⁸ Mean absolute error (MAE) — the average size of our miss, ignoring direction; lower is better.
Go deeper in Signal Syndicate

Blog posts are public education. The app has Research, signals, and Ask Signal.

Open the App Read the Methodology

All figures are estimates. Past analysis is not a guarantee of future results. Not betting advice.