Two sides, two lines, and the space between them.

Two books hang different numbers on the same spread or total, leaving a range of results where both sides of the position settle as winners. Call that range the middle zone. The zone is usually narrow and margin is paid on both legs, so the question is how wide the gap is, never whether one exists.

That is as far as this page takes the concept. Where the gaps come from, a worked totals example, and an honest account of how limited the appeal is are covered properly in the learn article rather than compressed here. What follows is about reading a candidate once you are looking at one.

The checklist is this: how wide the zone is in results, not percent; what each leg costs after margin; what the position returns when the result lands outside the zone on either side — that is the most common outcome and worth stating plainly; whether both lines are the same market; whether both books will take the stake the split calls for.

A candidate is a claim about two prices at a moment, and both halves decay. Timestamp discipline keeps a candidate from turning into a loss. A middle that looked promising thirty seconds ago may not be there anymore, or the books may have already corrected one leg. This is why speed matters, not because you are racing other bettors, but because the gap that made the position viable closes as books align. The board shows the candidate as it was captured; you need to verify it is still there before committing.

The zone is a small number of results wide, margin is paid on both legs so the position starts behind, and the books that create middles work to close them. Most candidates are therefore not worth taking, and a board that surfaced every one would be noise rather than signal. The three structural reasons are set out in full in the learn article.

What that means for this board: the useful question is never whether a middle exists, but whether this one is wide enough to be worth two stakes and two accounts. That is the judgement the numbers on screen are there to support.

Both accounts must be funded and unrestricted before a candidate appears, because moving money takes longer than the gap lasts. Books may restrict accounts for consistent paired-line activity as they do for arbitrage. If you see a middle you want to take but one bookmaker has cut limits to small stakes, the board will not surface it to you — the circuit that determines what shows up already knows that both sides have to be takeable at the split amount. That filtering keeps you from chasing phantoms.

PhotonOdds never accepts wagers, holds funds, or places a bet on your behalf. A middle is a position you construct yourself at two separate bookmakers, using two separate accounts and two separate financial connections. Everything about execution — timing, stakes, which book first, whether to hedge — is your responsibility. The board gives you the candidate and the arithmetic. What you do with it is your decision.