Who Is Signal Syndicate?

Signal Syndicate is a sports betting research platform that publishes its record before it
publishes its advice. That order matters, and it is the whole reason this page exists.

We build models that grade sports markets, we test them against results they had no part in
predicting, and we put the outcome — wins and losses alike — on a public ledger. If you want
to know whether a method works, the only honest answer is a track record you can check. So
here it is: the performance page.

What you can actually look at

• The performance ledger — every

position we have taken, graded, including the ones that lost. The losses stay up.

• The daily analysis — each morning's read of

the slate, written from that day's frozen data and published whether it looked good or not.

• The research library — how the models are

built, what failed validation, and what we retired.

• The tools — a

closing-line-value calculator, a
stake-sizing estimator, and a variance preview. Free, no account, nothing collected.

How something becomes a signal here

A model idea starts in shadow mode: it runs against live games, its results are recorded,
and it touches nothing a user sees. We require 75 out-of-sample bets before a lane is even
considered for promotion.

Most ideas do not get there, and we write those up too. One of our MLB totals lanes looked
excellent in a 2025 backtest — a +15.98% return on 186 bets in one narrow sub-lane. We
checked the same sub-lane in 2024 and got −1.76%. It had three out-of-sample bets in the
current forward test, seventy-two short of the gate. The verdict was *monitor, do not
promote*, and that is what happened to it. You can
read the whole decision — the
numbers, the reasoning, and the vote.

That is the pattern, and it is not a flattering one to publish. A good-looking number from a
small sample is the most common way betting products sell you something. We would rather show
you the moment we decided not to.

What we are not

We are not a tipster service. There is no paid group, no membership that unlocks "the plays,"
no channel where we tell you what to bet. We publish models, evidence and process, and the
decision about your money stays entirely with you.

We also do not claim certainty about single games. Nobody has it. What we do have is a method
for checking whether a model beats the number it was offered against, over hundreds of bets,
which is a question with a real answer.

Where the numbers come from

Everything is graded against the closing line — the final price in the market before an
event starts. It is the standard we use because it is the hardest number to argue with, and
you can check ours against it: what closing line value actually means.

Historical tests use data frozen as of each game date, so a model cannot be credited with
information it would not have had. Why that distinction matters more than it sounds is
explained in the small-sample traps
and why win rate misleads.

If you want to start somewhere

Read how betting lines move and what it does not mean,
then look at the ledger, then try the CLV calculator
on a bet you placed recently. If the answer it gives you is uncomfortable, that is more useful
than a compliment.

---

*Research and education only. Estimates where noted. Not betting advice — we do not tell you
how to bet your money, and we publish the same standard we hold ourselves to.*