Methodology & Trust Center

How we make projections: and how we grade them in public.

Nebula Insights publishes an independent model number beside the market's, logs it before the event, and grades it after against the same games the market priced. This page explains exactly how that works and shows the honest current numbers, including where we trail. Information, not picks.

Methodology updated: July 2026

How the projections are made

Our models learn from years of historical games. They use only team, player, park and schedule data team and pitcher form, park factors, rest and travel, and similar public inputs. Crucially, they never see a betting line: the projection is a genuinely independent estimate, not a repackaging of the market.

1 · Features
Public data only

Team & pitcher form, park factors, rest and schedule, built as of game time so nothing from the future leaks in.

2 · Model
Trained & validated

Walk-forward across seasons (2021 to today), so every test game is one the model had not seen.

3 · Calibration
Honest probabilities

Outputs are calibrated so a stated 55% actually wins ~55% of the time, no false confidence. Win probabilities are isotonic-calibrated and clipped to a 2%-98% range; we never publish a game as a certainty.

4 · Publish
Beside the market

We show our number next to the book's, never a lean, arrow, or bet direction.

For MLB we publish a projected total (the over/under number) and a calibrated home-win probability for the moneyline. For the run line and for NFL and UFC, we publish the market's own numbers and factual context only, because our testing found no reliable signal to add there, and we would rather show nothing than dress up noise.

The honest numbers

Measured out-of-sample from 2021 to mid-2026, on the same games that had a real opening line. We show our error next to the market's so you can judge the model yourself.

MLB totals · avg error
Our projection3.52
Market opening line3.46
Runs away from the final score, on 12,351 lined games. We trail the market by about 0.06 runs, the expected, honest result.
MLB moneyline · Brier
Our probability0.245
Coin-flip baseline0.249
Brier score, lower is better. Our calibrated win probability beats a no-skill baseline; it does not beat the market, and we do not claim it does.
MLB run line · ATS signal
Our verdictNone found
What we publishMarket line
Testing showed no reliable signal against the ±1.5 run line. So we publish the market run line only, never a cover probability or lean.

An improvement story, honestly told. Our previous totals model missed by 3.72 runs on average; the current one misses by 3.52. That is real progress against our old self, and it still trails the market. Both facts are true, and we say both.

These are the only model performance numbers we quote publicly. Every figure above is fixed to our audited July 2026 evaluation; we do not round them favorably or restate them.

How the track record grades itself

The public track record is the trust asset. It works on four rules, and nothing about it is discretionary:

For markets where we publish no projection (the run line, NFL, UFC), the track record shows the market's own benchmark instead, clearly labelled as the sportsbooks' record, not ours.

What this is: and is not

Nebula Insights is a free, independent analytics layer. We show the math and the receipts. We are never paid when you lose, and we run no sportsbook affiliate links.

x Not picks, leans, plays, or bet recommendations
x Not a sportsbook or a place to wager
x Not a promise of profit or a guaranteed outcome
x Not financial or betting advice

Frequently asked

How are the projections made?
Our models are trained on years of historical games using only team, player, park and schedule data, team and pitcher form, park factors, rest, and similar public inputs. They never see a betting line. For each upcoming game the model produces a number, a projected total for MLB over/under, or a calibrated home-win probability for the moneyline, which is calibrated so the stated probabilities match observed frequencies. We publish that number beside the market's, never a bet.
How are projections graded?
Every projection is logged and time-stamped before the event starts, then graded after the game is final against the actual result. We compare our error to the market's error on the exact same games that had a real opening line, so the comparison is apples-to-apples. Snapshots are immutable: once logged, a projection is never backfilled or edited. What we published is what you see on the track record.
Why do you show your losses?
Because an honest, checkable record is the entire point. No public model reliably out-predicts the closing betting market, and hiding the misses would make us exactly the kind of black box we exist to replace. We show every result, including the games where our projection trailed the market. The honest gap is the product.
Is your projection better than the market or Vegas?
No, and we will not pretend it is. On MLB totals our projection has missed final scores by an average of 3.52 runs since 2021; the market's opening line missed by 3.46 on the same games, so we trail by about 0.06 runs per game. We publish both numbers because that is the honest state of the art. What we offer is an independent second opinion, built without ever seeing a betting line.
Do you sell picks?
No. There are no picks, leans, plays, or bet recommendations anywhere on the site, and there never will be. We are not paid when you deposit or when you lose, and we run no sportsbook affiliate links. We publish an independent number and a public record, and let you judge.

Nebula Insights content is provided for informational and analytical purposes only. It is not betting advice, not financial advice, and not a guarantee of any outcome. Odds move; verify all prices and availability directly with sportsbooks. 18+ only. If gambling is a problem for you, call 1-800-GAMBLER.